Dos Santos further writes, "When the Portuguese found themselves in the land of gold [Manica], they thought they would immediately be able to fill sacks with it and carry off as much as they chose; but when they had spent a few days near the mines [soil-washing places] and saw the difficulty and labour [**] they found their hopes frustrated" (VII, 218). The records further show that the Portuguese were fully aware that the gold bartered by the natives was obtained from rivers; for instance, Sousa states, "the rivers of the country [of the Monomotapa[18]] have golden sands." (I, 15).

Dr Theal states, "The natives neither knew how to dig, nor had the necessary tools. Only by washing river sand in pools after heavy rains, these barbarians obtained all the gold that was purchased at Sofala" (VIII, 364). The localities of operations of washing surface earth the Portuguese in their usual grandiloquent style misnamed "mines." Dr. Theal further states (VIII, 478), " In none of the records still preserved is there any trace of ancient underground workings having been discovered by the Portuguese."

Manica and Mazoe districts comprise many areas of square miles in extent where the surface soil has been trenched over for gold. All mining engineers and surveyors working on the present mines on the rock in these districts have always asserted that the soil trenching operations are of mediaeval and post-mediaeval times, probably the work of old MaKaranga and BaTonga people, that the ancient mines sunk to depth in the rock of these districts are undoubtedly ancient in the fullest sense, these having been naturally silted in up to 150 ft. (45 m) in the course of centuries. This is obvious to anyone inspecting both the surface workings and the mines on the rock.

Reef-mining, if any, by MaKaranga and Ba-Tonga during the Portuguese period was on outcrops only ; that is by extracting the "vein stones" from outcrops of reefs, in the same way as can be seen in the Wedza district, where MaKaranga within the last few hundred years, and even lately, have followed along the line of outcrops for miles, taking only the "vein stones" of copper and iron ores without any sinking, and from positions where alone they could use fire and water to split the rock, a process impossible of adoption in the deep mines.[19] This nibbling of outcrops for iron and copper, but not for gold, was a common native practice not so very long ago.

But assuming for the purposes of Professor Randall-MacIver’s argument that a vast amount of gold had been obtained by the natives from washing surface soil and sand in riverbeds, and the records show most conclusively that this was not the case, such an amount, whatever it might have been, does not account for a single pennyweight of the more than £75,000,000 which has been extracted from the rock of the ancient mines scattered thickly all over the country, an area of 700 by 600 miles (1,126 x 965 km)

Bocarro (II, 399) describes sacrifices by the natives that the spirits of the dead might point out in dreams where the "mines" were ; that is, where the soil contained gold dust. This practice is very common to-day in the case of lost cattle or stolen property, but divining for gold is not exactly mining prospecting; yet this is but another of many evidences that the MaKaranga and Ba-Tonga of those days, and the "mining" operations of both these people is described in the same identical terms, were absolutely ignorant of rock-mining and also of assaying reef. Divination cannot possibly explain the extraordinary skill of the ancients in estimating the value of rock, the bulk of which, not showing a speck of visible gold, could not have afforded without assay and determination any possible clue as to its value. [**]

                                                         Hall: The Conical Tower at Great Zimbabwe

                          Prior to excavation                                                  After excavation 1902 - 3

Gold of No Value to Natives, 1505-1760

The ancient mines, sunk to depth in the rock, which cover an area of 700 by 600 miles (1,126 x 965 km) testify that the prehistoric miners valued gold. The Portuguese records demonstrate that the natives of the country of the Monomotapa’s cared nothing for it except as an article to barter for loin-cloths, blankets, glass beads, and brass wire; that, as to-day, their cattle and not gold formed their internal currency[20] and that they only washed surface soil for gold, and then only between harvest and sowing only bartering during one short period of the year, just as their trading to-day is confined to certain months of the year only.

Barbosa, writing of the trade as the Portuguese found it on their arrival, says, "The natives of Benemotapa exchange gold for cloth, without weighing the gold, in such a quantity they [the Moors] commonly gain a hundred for one" (I, 94, 96). Barretto (1667) states that the native chiefs do not wish gold to be dug for in their lands, because the Portuguese might buy the land from the King, and they would be despoiled of their lands (III, 483), or only such a small amount of gold as was sufficient to buy necessaries (III, 485). Even at Masapa, "where there were rich mines," cows were worth more than gold (II, 120). We have already seen that the "mines" of Manica were "not much valued by their owners" (I, 15).

Scarcity of Gold Ornaments of Natives, 1505-1760

Soares mentions that gold beads and trinkets were brought by the natives to Sofala, these only weighing 10 to 12 meticals (a metical, according to Ferao, = ¾ of an ounce) "making out that he is sending to him [the factor] the greatest thing in the world," i.e. that the natives not having many gold ornaments prize these small quantities (I, 82) The gold ornaments given to the Governor of Mozambique by the Monomotapa "did not weigh 10 meticals, the honours and value not being equal" (Father Monclaros, III, 248) Dr Theal states in his Abstracts of the Records that these small presents of gold beads were ridiculed by the Portuguese, and he considers that such ornaments were exceedingly rare among the natives.

The Arab writers of the twelfth century state that the natives did not wear ornaments of gold notwithstanding there was gold in Sofala, "nevertheless the inhabitants prefer copper and make ornaments of it," and also that they adorned themselves with brass and not gold ornaments. The records (1 505-1 760) contain only three references to gold ornaments being worn, and in two of these instances the gold in quantity ranks after iron and copper.

Ferao, "captain of Sena," states the natives "do not know how to work this metal [gold] and never apply it to any purpose" (VII, 379). The Portuguese records show that iron-smelting furnaces were not numerous, and they mention them as rarities. Authorities on the Bantu are unanimous in stating that no tribe of Bantu people have ever been known to weld metal. Lacerda only saw one native in Zambesia wearing gold "spangles" (split pellets, just as the natives of the present-day wear copper and tin spangles in their hair) and remarks, "but no one prized them" (p. 23), the chief of the Maravi informing him that "his people did not extract gold, because they knew not what it was" (p. 80), further, they did not do so "because their ancestors had not done so." Lacerda said that the natives did not know how to mine for gold, nor how to convert it into ornaments, and also that the natives did not value gold. All the writers very frequently refer to the brass, iron, and copper ornaments worn by the natives. For instance, "The negroes make of it [copper] their necklaces, bracelets, and anklets (yergas, wires [imported]) like carpet-rods, twisted round the legs," as also worn to-day.

But the evident scarcity of gold ornaments among these people to which the records testify appears to be borne out by the "finds." We are distinctly informed in the records that the MaKaranga did not make or wear large gold bangles and necklaces (I, 32), and yet out of the £4,000 worth of ornate gold ornaments found on the lowest and original floors in the older structures at Zimbabwe and out of the profusion of gold ornaments found in other ruins of the Zimbabwe type and age, the greater portion consisted of these large bangles and necklaces.[21] This points to an occupation of Zimbabwe and certain other ruins at some period very long prior to 1500 AD and also to the very general practice of wearing such articles. Moreover, gold bangles are stated to have been worn only by the Monomotapa himself, "an honour he grants to none and reserves for himself alone" (III, 248) This confirms the statements of W.G. Neal and myself in The Ancient Ruins of Rhodesia, written before the Portuguese records were discovered and based on our finds in two distinctly different types of ruins, that at any rate since 1500 AD, if not from some very much earlier period, the MaKaranga did not make or wear gold ornaments. In the graves of the natives of (about) that period, and of subsequent times, hardly any gold beads, if any, are discovered, nor in debris of the huts of such natives; nor are they to be found on the clay (dhaka) floors which yield the class of MaKaranga and Ba-Rosie[22] [Rozwi] pottery of a late period and of poor make with which Professor Randall-MacIver so freely illustrates his work; nor have they been, or ever will be, found in certain ruins of a late date and of poor construction said by Professor Randall-MacIver, but without the slightest warrant, to be the "prototypes" of the Temple at Zimbabwe. Every future explorer must meet with the same experience concerning " finds" of gold ornaments as were described seven years ago in The Ancient Ruins of Rhodesia of the very oldest type, in which the bodies lay in a horizontal position, also on the lowest granite cement floors of ruins of the Zimbabwe order, and then they were always discovered in rich profusion, "as plentifully as nails on the floor of a carpenter's shop." Gold will not, unless a stray bead or two, be found on the upper clay {dhaka) floors, or in connection with daga structures.

This evident scarcity of gold ornaments during this period provides a striking contrast with the circumstances of some much earlier period, and points to a decided divergence in culture. From 957 A.D., when AY Wardy wrote that the natives of the Sofala country did not wear ornaments of gold, we have a long succession of emphatic statements by Arab, Persian, and Portuguese writers to the effect that no gold ornaments were worn. Was the gold ornament wearing period at some time prior to 957 AD? All the brass wire bangles so far discovered contain a "core of vegetable fibre" (M.R. 76), but in fully a hundred instances of the discovery of gold bangles the "core of vegetable fibre" had completely disappeared. This alone points to "periods," and also to a wide divergence in culture.

Gold beads are also to be found in positions where they have been undisturbed for many centuries. One instance of several may be stated. In 1906 Mr. Garthwaite, consulting mining engineer for the BSACo , found gold beads 12 ft. (3*65 M.) deep in solid gravel which had never before been artificially disturbed, in a district where there were no ruins or mines, and nowhere near any river.

Unsettled State of the Country, 1505-1760

While the ancient rock mines in Rhodesia are shown by mining experts to evidence very long periods, covering centuries, of peaceful, unmolested, and well-organised mining operations, the period 1505-1760 is shown by historic records to have been one of continuous wars and of completest unsettlement and disturbance, precluding any suggestion whatever of natives raising within that period even the veriest appreciable portion of the vast amount of gold estimated to have been extracted from the rock of the country.

On the arrival of the Portuguese in 1487 they found the Arabs and Persians of Mozambique and Sofala to be engaged in a chronic state of warfare with each other, and with various tribes on the coast. "This constant strife," says Dr. Theal, "was the key to the easy conquest of the coast regions and islands by the Portuguese." The disruption of the "empire" of the Monomotapa’s, which had taken place long before the arrival of the Portuguese, still resulted in permanent feuds and constant wars in the interior.

The small amount of gold arriving at Sofala is attributed to native wars inland (I, 66) De Brito, factor of Sofala, in 1 519 reports, "This country is ruined," and that a chief of Quiteve had "reduced all the territories round this fortress" (I, 105). In 1569 Tete trading station was temporarily abandoned owing to wars (Monclaros, III, 202). In 1570 the Portuguese were fighting with the Quiteve, the king of the hinterland of Sofala (De Conto, VI, 388). Then followed the twenty years' war (1570- 1590) of the Muzimbas with other tribes, also with the Portuguese, which stopped all trade at Mozambique, Zambesia, and Sofala, the ill-effects continuing till the end of the seventeenth century (VIII, and generally). In 1592 there was a slight revival in trade, but later "matters along the great river (Zambesi) were in a worse condition than ever before" (III, 403). In 1602 "the Cabires, a warlike tribe, were in possession of the mines of Chicova, and the principal mines of the kingdom of Monomotapa" (King of Portugal to the Viceroy of India, IV, 50). Dos Santos states there was war "nearly every year" (VII, 273), and further states, "the natives went about in bands at variance with each other" (VII, 363). De Barros reports, "No gold has been extracted from the mines for years because of the wars" (VI, 268). In the Government reports (1584-1668) it is evident that the Kings of Portugal were more concerned for the safety of Sofala and Mozambique as naval depots on the route to India, which were threatened by the Dutch and later by the English than for their value as gold trading stations. In 1607 the crews of Dutch ships robbed and sacked the town and port of Mozambique (V, 285). In 1609 the chief trading station of the Portuguese at Masapa in Mocaranga was abandoned, "all the country was in arms against the Portuguese" (Bocarro, III, 383), and the Portuguese are ordered to withdraw from the Zambesi to Mozambique "to defend it from the Dutch" (III, 384). In 1615 the Chicova fort was abandoned by the Portuguese (IV, 158). In 1628 the King of Portugal writes, "The trade of the rivers of Cuama ['five mouths, 'Zambesi] is in a miserable state, "which he attributes to wars (IV, 213). In 1634 Dutch pirates rob the Portuguese trading ships off Sofala. In 1635 the Portuguese are at war with the kingdom of Manica (IV, 278). In 1651 English ships threaten Sofala coast ports. In 1667 "the settlement [Mozambique] is almost deserted " (Barretto, III, 480). In 1687 English ships trade in South-east Africa and seized the commerce of the Portuguese (V, 296). In 1719 "disorders [in Mocaranga] are frequent" (V, 66); "the vast empire [of Mocaranga] is in such a state of decadence that no one has dominion over it" (V, 72). In 1748 French ships visit Mozambique and usurp the Portuguese trade with the neighbouring islands and "the commerce of Querimba [an important trading station between the Zambesi and Mozambique] is entirely in the hands of the French " (V, 194)

 Hall: Monomotapa (or Mocaranga) in the sixteenth and seventeenth Century according to R.N. Hall

But, in passing, be it remembered that the records refer to that period, the stormy and unsettled times just described, which Professor Randall-MacIver claims as the very period during which the bulk of the Rhodesian ruins were erected, and as part of that period the greater, or at any rate a large, portion of the more than £75,000,000 of gold was extracted from the rock at depth. But the records make the acceptance of such conclusions absolutely impossible, though there is no doubt, as was contended by me almost ten years ago, that some of the small and poorer buildings, mere parodies of the earlier structures, were erected in subsequent times. Later we shall draw attention to the strange ignorance of the Moors and Portuguese that such colossal building operations and such skilled mining of rock (not soil or sand in rivers) on an astonishingly extensive scale were going on (so Professor Randall-MacIver invites us to believe), not only over an area of 700 by 600 miles (1,126 x 965 km) of country, but actually in those very districts where were the Portuguese trading stations. But not a rumour of such colossal undertakings, which Professor Randall-MacIver states must then have been in progress, ever reached the ears of the Portuguese! But to return to the records.

Portuguese Trade at Sofala* 1505-1760

On account of the poorness of the gold trade of Sofala, which place Duarte Barbosa describes as "a village" (1, 93), Alcáçova applies for a better position in India (I, 67). De Lemos, factor of Sofala, reports in 1508, the gold of Sofala "is plentiful in the country, but the natives barter very little" (I, 72). Soares, factor of Sofala in 1513, complains of the small quantity of gold bartered for. He says, "I see so few natives and traders from the interior that from then to the present time I have not bartered 500 meticals, and this though the whole country is at peace," and "the captains by presents had laboured to secure a commerce." "Although there is gold scattered over the whole country, no one has such a quantity that it is worth his while to come so far in order to barter it"; also he says, "There is not so much gold in this country as has been reported" (I, 80, 81). Soares also reports, “A great expense for so little revenue and profit, which is scarcely sufficient to cover the charges of the said establishment [Sofala]." He suggests the abolition of certain offices, also the reduction of salaries of officials and he states, "A factor and two clerks" were "sufficient to carry on the trade [at Sofala]"; and further, "There is nothing to prevent the traders coming here, if there are any" (I, 82, 83). De Brito, factor of Sofala, reported in 1519 that in eleven months he had only received 552 meticals of gold, which, he says, was insufficient to pay the official salaries of the factory, and "I am quite ruined [by farming the trade], and I wish I had not come here at any price;" trade at Sofala, he says, "is so dull that men have no heart " (I, 105, 106). Sofala as a trading station is declared to be useless {Ibid.) In 1552 official reports show a further serious decline in the gold trade of Sofala (III, 148, 149). In 1560 Father Monclaros states, "The favourable reports of the abundant riches of Monomotapa are not borne out by facts" (III, 302). In 1580 it was officially reported that "this fortress [Sofala] yields no revenue to our lord the King, except a small trade in ivory" (IV, 1). De Barretto writes, "Not a grain of gold is to be found in [the kingdom of] Quiteve " (III, 489), Quiteve being the immediate hinterland of Sofala, extending inland for 150 miles (241 K.). In 1585 Sofala was yielding nothing except the profit on a small quantity of ivory (VIII, 406). In 1634 Rezende, "the most competent writer of his day," reports that Sofala had no garrison, only three Portuguese residing there. He states, "The only commerce carried on was in ivory" also, "the only merchandise being ivory" (II, 405). In 1635 it is reported, "The fortress at Sofala is in a ruinous state, with no men" (IV, 255). In 1667 Barretto states that the principal trade of Sofala, "which is almost deserted," is " ivory" (III, 480) ; and later states his reasons to account for the small quantity of gold produced from the whole of the country. Ferao attributes the failure of the Sofala trade in gold to civil wars in progress in Quiteve (VII, 378, 380, 381).

Much has been made by the supporters of Professor Randall-MacIver’s conclusions of the reference to the "two ships of the Moors who had laden gold [1500] from that mine [Sofala] and were going to Melinde." (I, 48). The records show that owing to sand-banks ships could not enter the port of Sofala, and that all the transport along the entire coast of Mozambique was carried on in zambucos. On P 91, I, it is stated that "they [the Moors] came [to Sofala] in little vessels which they call zambucos from the kingdoms of Kilwa, Mombasa, and Melinde." The two "ships" were "zambucos" (V, 443)/ and Father Fernandes, 1560, says (II, 83) zambucos were small open boats "where there is no room for a man to stand, sit or lie down." De Goes states (III, 77), "The ships, or zambucos, in which these Moors sailed had no decks and were not nailed together but were fastened with wooden pins and cord made of palm fibre; the sails were made of the same palm tree closely woven together like mats." These boats, which only had one mast and one sail, were always drawn up above highwater mark when in harbour. In calms and contrary winds they were propelled with oars (VI, 171). Such, then, are the reports on the state of the gold trade at Sofala. But if such was the sorry state of the gold trade at Sofala from 1505 to 1760, in what position was the gold trade of Mozambique and Zambesia during the same period?

Portuguese Trade of Mozambique, Zambesia, and Coast

But Mozambique was the Portuguese centre both of the administrative authority and of the commerce for the whole of South-East Africa, including not only Zambesia and the country of the Monomotapa, and of Manica, Quiteve, and Sofala, but of Melinde, Mombasa, Kilwa, Magadoxo, Inhambane, Lourenco Marques, and of all the islands lying along the coast from Cape Delgado to Cape Correntes. All trading or barter goods for all these places were only obtainable, except by smuggling, from Mozambique, and to the central depot at Mozambique was sent all the gold, ivory, and other articles bartered from the natives at the sub-trading stations. The records explicitly state, as shown later, that no gold bars or ingot gold were ever sent to Mozambique from any one of the sub-trading stations of Mocaranga. All gold arrived in the form of gold dust, which on arrival was converted by Government assayers into bar gold. Mozambique was the only place where the official assayers were stationed. From Mozambique the gold was exported to India, only a small quantity being sent to Portugal. Consequently the state of the gold trade at Mozambique provides a fair index as to the state of the gold trade of the Portuguese throughout the entire South-East African coast.

The records show that at Mozambique the gold trade never flourished, but on the other hand was ever in a chronic state of depression, at times of stagnation and complete ruin. The vicissitudes of the island were at times exciting, varying from the chronic unrest of its Moorish population to sieges by the Dutch, the ruination of its trade by the French, constant fears of an English occupation, and a hostile native population of Maravi, who were cannibals, on the mainland, and who for long periods together not only refused a foothold to the Portuguese in their territory but denied to them the supply of provisions.

In 1503 the kings of Melinde and Mombasa were at war (III, 103), while "the Christians and Moors of Kilwa were always at war" (III, 79) In 1572 Monclaros, a local writer, states, "The people [on the Mozambique coast] are generally poor and wretched in nearly all these parts, and the Portuguese are becoming so already through the loss of the commerce and navigation taken from them by their enemies" (III, 216). In 1584 an official report states, "This fortress [Mozambique] produces no revenue for our lord the king" (IV, 2). In 1634 Rezende writes, "The Captain of Mozambique holds a monopoly of the commerce, he carries it on alone, with only one small vessel" (II, 405). The failure of the trade reduced the Governor of Mozambique "to beggary" (IV, 279).

The records show there was never at any time any trade in gold from the mainland of the northern Mozambique coast. "The Portuguese resorted to these rivers [on the Mozambique coast north of the Zambesi] to trade for ivory, provisions, and ambergris," there being not a single reference to gold (III> 223). Moreover, the Portuguese of those days, as Dr Livingstone and Sir Richard Burton have pointed out, never went inland on the mainland from Mozambique, while the records expressly state that the Portuguese possessed no knowledge of these regions for the simple reason that the natives, the Maravi, were a most dangerous and warlike people always at strife among themselves, and on every opportunity raiding the Portuguese settlements even on the islands along the coast.

                                                           The Great Enclosure

             Hall: ‘the small cone’                                                                Conical tower and Great Enclosure after excavation

From the northern banks of the Zambesi the Portuguese traded with the Maravi, but for ivory only, and after 1645 for slaves and ivory. From the country of the Maravi, which was bounded on the south by the Zambesi, a territory extending for at least 200 by 200 miles (321 km) the records frequently mention that no gold was ever obtained, for instance, "In Maravi not a grain of gold is to be found" (III, 489), while Lacerda states that the Maravi had not the remotest idea of what gold was.

We find that the trade between Mozambique and Lourenco Marques, Cape Correntes and Inhambane was comparatively insignificant. One pangayo, a small boat propelled either by a sail or oars, was sent every year from Mozambique to Cape Correntes and Inhambane with trading goods for both those places, to bring back "ivory, slaves, honey, butter," there being no reference to gold (VII, 331). In 1630 it was said, "Inhambane, where there is a trade in ivory," gold not being mentioned (I, 22). One pangayo sufficed for the year's trade between Mozambique and Lourenco Marques, but it only brought back ivory, there being no mention of gold, and it is said the pangayo returned  “half laden." (VII, 366). One pangayo every six months was sufficient for the trade between Mozambique and Angoche, whence was obtained "ivory, tusks of the hippopotamus, some ambergris, a number of slaves and very fine straw mats," there being no mention of gold (VII, 332). One pangayo every year suffices for the trade between Mozambique and Querimba, "millet and rice, cows, goats" being mentioned, but not gold. One also suffices for a year's trade between Mozambique and Madagascar, the commerce consisting of "cows, goats, ambergris, and slaves," there being no references to gold (VII, 332).

Owing to the chronic state of depression of the gold trade at Mozambique itself, Father M. Barretto wrote (1667) that "there was a proverbial saying in the mouths of the inhabitants and persons acquainted with these parts that all which is outside Mozambique is better than Mozambique." (III, 502) The inhabitants of Mozambique were noted as having a boast which was to the effect that there were more Portuguese buried there than anywhere outside Portugal. "It is the sepulchre of thousands of Portuguese" (III, 464). The records further show that owing to the poorness of trade the officials at Mozambique frequently became ruined and were always petitioning the Home Government for their removal elsewhere.

But the official documents and revenue reports given in the records disclose a most hopeless state of affairs at the headquarters of the Portuguese trade in South-east Africa. In 1590 Friar da Zevedo writes to the King of Portugal, "The kingdom of Monomotapa and the rivers of Cuama [Zambesi] at present profit you nothing" (IV, 35). In 1593 the King of Portugal writes, "My treasury in that state [Mozambique and Sofala] not only does not receive any profit from the trade of these fortresses but has also been obliged to bear the expenses thereof" (IV, 39). In 1593 Kilwa was destroyed by the Muzimba, who ruined the trade at Kilwa and Mombasa for many years (VII, 302). In 1623 the stations at the mouths of the Zambesi were "in a most abandoned state, and in want of everything, especially of men." (IV, 230). In 1628 the King of Portugal writes, "The trade of the rivers of Cuama [Zambesi] is in a miserable state" (IV, 213). In 1635 Quilimane station "is in a ruinous state with no men" (IV, 255). In this year, the Maravi tribe besieged Quilimane and stopped all trade (IV, 278). In 1684 the King of Portugal writes, the revenues of Mozambique "would not suffice to cover the necessary expenses for the defence of Mozambique and the rivers [Zambesi]" (IV, 423). In 1688, for three years, says the King of Portugal, there has been nothing for the Royal Treasury " because the decadence of the rivers increased every day" (IV, 449). In 1711 local tribes again besiege Quilimane and stop all trading (V, 33). In 1720 it is stated that "no profit whatever resulted from the trading, debts upon debts accumulating yearly" (V, 91). In 1720 the King of Portugal writes, "The debts [of Mozambique and the rivers to the Royal Treasury] have increased annually, the northern stations [Kilwa, Mombasa, Melinde declining more and more" (V. 74) and "that vast empire [of the Monomotapa] is in a state of decadence." In 1734 the King of Portugal writes, "I have seen your letter relating to the ruinous state to which the General Council of Mozambique and the rivers is reduced, being more than 200,000 cruzados in debt" (V, 175)

Portuguese Trade of Sabia and Limpopo Districts, 1505-1760

But the most astonishing feature in the records is the fact that they are absolutely silent as to any gold being obtained in the most important districts of Sabia and Limpopo, the evidences both positive and negative being that no gold was traded during this period from these territories. On this point the records are emphatic, and, further, they clearly prove that the Portuguese never visited them, and that what trade came from them consisted only of "ivory, ambergris, and iron, sesame and other vegetables" (VII, 186), and also copper, brought by the natives to Sofala, Mambone at the mouth of the Sabi, and Inhambane, and this only very rarely and in small quantities.

Yet the immense kingdom of Sabia — the records define its boundaries — which was traversed by the Sabi river, occupied by MaKaranga, and included the Great Zimbabwe (which the Portuguese never saw, and which is not the "Zimboache"[23] of Barbosa which he stated was in the kingdom of the "Benemotapa" which "Zimboache" was the chief zimbaoe or "residence" of the Monomotapa near Masapa in the Mazoe district), yields in its mines by far the best and most substantial evidences of ancient activities to be found anywhere in the whole of South-east Africa! To attempt to deal with the problem of the ancient rock mines between the Zambesi and the Limpopo without any reference whatever to these most important territories would be tantamount to discussing the play of Hamlet with the part of the Prince of Denmark omitted.

The Sabi provided in ancient times, and has always to this day provided, the only natural and possible trade route through the mountainous barriers of the escarpment of the central South African plateau from Sofala to the Zimbabwe country, while the Limpopo provided and still provides the natural and only possible approach from the coast to the ancient goldfields of Tati, Tuli, Gwanda, Belingwe and in fact, to the whole of the ancient rock mines area lying between those districts and the Murchison Range in the Transvaal; an immense area of ancient mining activities of the existence of which the Portuguese were absolutely ignorant. Evidently the scores of millions sterling of gold did not come from these two areas between 1505 and 1760, nor had the Moors any traditions concerning the great wealth which was once obtained from these sources.

Inland Trading Stations

None of the inland markets of the Portuguese, excepting Sena, were established until sometime after De Barretto's expedition in 1569, while the greater number were founded considerably later, some not being opened until over one hundred years after the Portuguese had arrived at Sofala in 1505, while there is no reference to others until 1749, about twenty years before the Portuguese power in inland South-east Africa was broken. The records further show that these markets have very precarious existence, also that some were only used for a very few years. For instance, Masapa was first abandoned in 1616 (I, 39), and there are references to later abandonments. The Portuguese were driven out of Bocuto in 1609 (III, 379). Chicova was established in 1614, and finally abandoned in 1616 (I, 41, 42). In 1572 Monclaros mentions Tete as "where the Portuguese formerly traded" (III, 226) and says, "Tete was deserted by our people" (III, 239), and he describes Sena as "a small village with straw huts" (III, 223). In 1616 and also in 1628 both Sena and Tete were raided by the Monomotapa (I, 43 ; II, 429). The factory of Chipiriviri was very short-lived, and being near Chicova, probably shared the same fate. In 1635 there were only "six men" in the kingdom of Manica (IV, 7), while the records show that the stations in Manica were abandoned at times for years together, some finally abandoned, and that in 1720 there was an attempt "to re-establish the fair of Manica" (V, 95). The only mention of Zumbo is in 1749, and only as a mission (V, 215), and of Massi-Kessi somewhat later still. Bandire was abandoned soon after its establishment (VII, 381).

The general unsettlement of the country from 1505 to 1760, as described earlier, accounts very largely for the short lives of these markets, but the introduction of the slave trade in 1645, as dealt with later, finally put an end to the Portuguese trade in gold. The Jesuit letters of 1740 (Wilmot's Monomotapa, P179) state that "all live in continual wars." But the policy of the Portuguese was equally as unaccountable for the general unsettlement as was the chronic state of native warfare or the Portuguese traffic in slaves. The Portuguese were always sending out punitive expeditions and to such an extent that in 1634 De Rezende wrote, "Those [natives] in the interior are in revolt against us and as we have often chastised them in war, they cherish a hatred of us" (II, 411) and he further explained the "petty commerce" by stating, "The negroes of these parts resent the punishments we have inflicted upon them." (416)

Moreover, these markets were only "annual fairs," that is, they were only open at one period of the year. The natives, the records show, did not wash in the rivers for gold except between harvest and sowing, also their trading was confined, as it is to-day, to one period of the year only. The Portuguese went annually to Masapa, Luanze, Bocuto, and Bandire, "where the natives from the interior go to wait for them at certain times."[24] (VI, 368 ; VII, 381)

                                                     The Great Enclosure

         Summit of the East Wall                                                    Chevron pattern on the East Wall

Portuguese Farm the Gold Trade

The various fiscal policies adopted by the Portuguese at different periods between 1505 and 1760, as detailed in the records, afford further evidence of the slackness of the gold trade during the whole of this period. At one time customs are farmed, at another the whole gold trade of Mozambique is farmed, at other times the Home Government works the trade on its own account, while intervening periods of absolute free trade alternate with those of rankest protection, monopolies, and prohibitive tariffs. The fiscal policy was always in a chronic state of variation to either or any extreme, no system proving satisfactory. In 1598 "the mines of Sofala" were leased for £4,050 per annum (IV, 46). It does not speak much of the value the Portuguese placed upon their gold trade when we consider the terms of one of these farming contracts. In 1614 the whole of the Portuguese trade in South-East Africa in gold and ivory was farmed to the Captain of Mozambique, who is debited with £7,500 annually, and against this amount he is credited with expenses of the trading stations, forts, officials, and troops, but not with the tribute payable to kings and chiefs, and the balance to be sent to India (VIII, 460). Little wonder is it that the records show that in many instances the contractors became ruined, and that the Government was always being defrauded by its officials.

There are, however, two rare instances cited where those who had "farmed" the gold output had been successful. But these two fortunate individuals enjoyed the monopoly of all trading goods brought into the country, and though only a small trade in gold was transacted, yet the margin of profit between the price of such goods when landed in the country and their actual selling price to the natives was so enormous — amounting in some instances to a thousand percent — that with a small turn-over a most successful trading could be carried on. Thus we read, "A good deal [of profit on ivory] is gained by it in consequence of the trifling value here [Sofala] of the merchandise with which it is purchased" (I, 85) Correa mentions that "for a piece of cloth worth 150 reis was paid gold worth 750 reis" (II, 28) Monclaros, writing of the great profit to be made in trade, says, "One hundred cruzados may well be made to yield three thousand crusades" (III, 234). Cloth valued at 66 reis when landed at Sofala was worth for sale 2½ meticals of gold dust; that is, 66 reis bought gold dust worth 1,167 reis (I, 104) Loincloths sold for one metical of gold dust "apiece," that is, 467 reis per loincloth! (Ill, 234, and also, I, 85). But these "farmers" are shown to have made more money out of their extortions and also by licences and permits to trade sublet by them and by the issue of slave-trading licences, for which there was a brisk and constant demand, than they made by their own trading.

  Still the majority of "farmers" were ruined by their contracts. Most wanted to leave the country and go to India or Portugal and boldly asked the king for their removal from East Africa. Promotion "elsewhere" depended on two conditions precedent, bribes  (one "farmer" modestly describing “bribes"[25] as recompense for obtaining office (II, 414)) and glowing reports. The constant aim of "farmers" and officials is shown to have been promotion "elsewhere," nothing in South-East Africa

being considered good enough. "I pray your Highness graciously to send me away from these parts [Sofala] " (Covesma, I, 56). De Brito writes, "I am quite ruined [by farming the trade] and I wish I had not come here [Sofala] at any price" (I, 105, 106). One applicant for promotion "elsewhere" pathetically commences, "For God's sake!" (I, 26) and no wonder when the Governor of Mozambique himself was "reduced to beggary" because, as is stated, he had paid too big a bribe for his position (IV, 279). The poet Camoens, induced to take office at Sofala, met with rather unpoetic experiences in money matters at that port (I, 20). Alcáçova, who wrote that he was "ruined" and therefore unable to buy promotion " elsewhere," sends in a most glowing report based on "the information derived at second hand from an untrustworthy source, viz. the reports of the Arab intermediaries who traded to Sofala." (Medieval Rhodesia, p. 60), and it is his glowing report which exactly five hundred years later led Professor Randall-MacIver so seriously astray.

But the factors' reports allude to the smuggling of gold by the Moors. Surely Professor Randall-MacIver would not suggest that the £75,000,000 of gold was smuggled out of the country between 1505 and 1760? There is no doubt the Moors did engage in illicit trade. Dr Theal very fairly and impartially sums up the references in the records to such a traffic when in his History of South Africa (p. 285) he writes, "That a considerable trade was carried on by the Mohammedans with the Bantu in defiance of the Portuguese is highly probable, but that it amounted to a very large sum is not at all likely." The enforcing of the customs against the Moors was so rigid, that "according to Durate Barbosa they were reduced to such straits (' abject misery ') that they began to cultivate cotton and manufacture loincloths themselves, but this, if correct at all, can only have been on a very limited scale." But Alcáçova states that the gold from the country of the Monomotapa "does not go out through any other part [that is, not from Zambesi ports] except through Sofala, and something [smuggled gold] through Angoche, but not much." (I, 66).

           Randall-MacIver: Iron implements and weapons, Great Zimbabwe

Portuguese did not Mine for Gold, 1505-1760

The records are very emphatic in showing that the Portuguese did not mine in these territories for gold. Though excellent miners in Brazil, they had no mining engineers or assayers in South-East Africa. In 1619, after being in the country for over one hundred years, they had to admit to the Home Government that they had no mining experts and asked for some to be sent out (IV, 162), and the first assayers were sent in 1649, but to Mozambique only (IV, 307), and also that they had no miners (IV, 158, 161, 162). In 1634 pieces of conjectured silver ore had to be taken to Lisbon for assay (I, 42 ; II, 41 1 ; III, 235, etc). Two hundred years after their arrival they confess they had not found the gold mines! Probably not, for the ancient mines at depth on the rock can be shown to have been naturally silted in and almost buried long centuries before the Portuguese arrived in this country.

In 1629 the Monomotapa donated to the Portuguese certain mines, but, says the Governor-General, "up to the present time there is only the word of the said natives in proof of this [their existence]" (IV, 159). So incompetent were the Portuguese that in 1635 they most seriously declared, "The metal [gold] is formed by the sun on the surface of the earth" (IV, 286). They describe "golden quarries " with lumps of gold as large as a man's head [the Chartered Company would be exceedingly glad to locate and "reserve" these "golden quarries"] also that gold grew in trunks of trees, and that it "grows " and "sprouts' presumably after the fashion of cabbages (I, 22; III, 355, etc.) "like a large yam" (VI, 367). "Mountains of silver" were reported as having been seen (IV, 159)

Dos Santos, a local writer, describes how the deceitful natives had buried two pieces of silver ore in the soil for the Portuguese to discover and that when they were dug up with a "plantation mattock" " there was rejoicing and delight, and the trumpets and drums of the camp assisted in celebrating the discovery." (VII, 283) they also discovered a "red earth that has not yet been converted into gold, but which shows by its colour that it will so become." (281).

But the Portuguese, the records tell us, did actually attempt at a very late period to mine in the northern portion of Mazoe, and there only, and they scattered Nankin china, this "very valuable dating material round about their workings by the hundredweight.” Mr. Telford Edwards has shown this was the case. Pieces of their timbering for the roofs in their extensions of the ancient adits were taken to London for examination and about three hundred years was allowed by experts as their utmost age. The Portuguese cleared out some few small ancient rock mines[26] of the silted soil with which they had become filled up, but, as the records show, they soon abandoned what was only an attempt, "for commerce native truck trading] is more profitable," "and they left them [the mines]" (III, 233), "the trade in cloth being more profitable, especially in loincloths" (III, 253). Further, Monclaros, a local writer, states, "The Monomotapa gave some gold mines [probably alluvial areas] to several Portuguese, but because the expense of extracting the gold was so great, and so little was taken out every day, they would not have them." (III, 233). " Gold mines" to the Portuguese were evidently not worth acceptance, even as gifts.

In a letter written by the King of Portugal (1640) we find, "The steps taken up to the present have not shown that there would be any advantage in working the said mines, on account of the small profit obtained from them." (IV, 287). In 1645 we find the Portuguese trade in slaves to Brazil and India from Mozambique paid far better, on their own admission, than either gold-trading or goldmining, while the records show (the references cited before) that the ivory trade at Sofala was more profitable than the trade in gold ! We read, too, that the Portuguese slave trade destroyed the confidence of the natives in the Portuguese and ruined the trade in gold. Little wonder the King of Portugal's "Ultramarine Council" at Lisbon became so exceedingly downhearted, as the official correspondence demonstrates. One thing is evident, the "Ultramarine Council" did not receive any of "the missing £75,000,000 worth of gold " which had been extracted from the rock at depth in those very territories of Southern Rhodesia containing the immense area of deep rock mines, of which the records do not make a single mention , and from which , it is explicitly stated, the natives traded no gold, and in the recent exploration of which rock mines no thirteenth- or fourteenth-century article has ever been found.

The records definitely state that the Moors of 1505 - 1760 did not mine for gold, the reason assigned being that they found trading "easier." "They [the Moors] have no other occupation than trading" (I, 103). De Barros, speaking of the gold of the Moors, says, "The gold which the Moors obtain from the negroes" (VI, 169). The records contain innumerable references to the gold obtained by the Moors having been obtained by barter only.

Export of Gold Dust and Bar Gold, 1505-1760

From 1505 to 1593 the whole of the gold from Sofala, Zambesi, and Mozambique was exported in the form of gold dust only. On this point the records are very explicit. From 1593 to 1649 gold was exported from Mozambique to India and Portugal in the form of dust and bars. In 1593 there is an official order that "bars" of gold (which appear from the records to have been in the form of bar adopted by the Portuguese in Europe) were to be stamped on their extremities and in the middle with the royal arms (IV, 40). In 1649 the King of Portugal, being advised of "frauds discovered at Mozambique with respect to the gold dust, which is brought from those parts,  dispatched assayers to the fortress of Mozambique and the rivers of Cuama [Zambesi] to remedy the frauds." (IV, 307). The first suggestions as to smelting gold dust and making bars appear in 1593. In 1635 are further orders that these bars were to be stamped with the arms of the crown, "in the same manner as in the Spanish Indies," in order to prevent any misappropriation (IV, 260).

The whole of the gold sent from the subsidiary trading stations of Sofala, Manica, Zambesi and from the country of the Monomotapa’s, to Mozambique, where alone the assayers were stationed, was sent, during the whole of the period from 1505 to 1760 in the form of gold dust only (IV, 360) moreover, "the currency is gold dust." (Monclaros, III, 208), therefore gold ingot moulds were not used in the interior by the Portuguese, nor by the natives, as these, it is stated, only traded the gold in dust form. [**]

                              Randall-MacIver Great Zimbabwe excavated artefacts

             Faience (Nankin) china imports                                                             Gold jewellery

Slave-trading Runs the Trade in Gold

In 1645 the Portuguese, being disappointed with the small amount of trade in gold, commenced to trade in slaves (IV, 302). The slave trade was at times a Government monopoly, and at other times it was "farmed." Licences to trade in slaves were sold at a high figure, and there was a brisk demand for them, which returned a revenue, as the records state, exceeding that derived from the gold trade, beside which the records show that the slave trade itself was far more remunerative than the slow business of bartering loincloths and glass beads for gold dust. But gold-trading could not be carried on simultaneously with the slave trade. With the introduction of the slave trade in 1645 commerce in gold ceased. Raids for slaves were, for at least one hundred years, carried on most extensively in the very territories which had formerly yielded the gold supply.

The Portuguese complained with most astonishing naivete at the disappearance of the gold trade, small as it had been, and whined because "the natives fled to other lands" Their fleeing is hardly a cause for wonder. Whole populations were decimated, and entire territories devastated. Dr Livingstone, in 1865, stated that the effect of this slave trade could be observed in Zambesia even in his day, while the traditions of slave-trading by white men, not Arabs, still exist among the MaKaranga of Mashonaland, who still possess many traditions as to the Portuguese occupation. (see Chapter V), Clearly there could have been no mining on the rock in such troubled times. We find from the records that this cruel and inhuman traffic completely destroyed the confidence of the natives in the Portuguese. The slaves were extensively exported to Brazil and other Portuguese colonies and were also supplied to the French colonies. Should anyone be interested to read of slave raids, the prices of slaves, and the profits of the trade, he should peruse The Records of South-Eastern Africa.

The Gold Trade - an Admitted Failure

The records contain, as can be seen, numberless official statements expressing the keen disappointment of the Portuguese at the non-success of their trade in gold and they admit they had nothing whatever to show in results to compensate them for the vast expenditure in troops and administration and for the appalling loss of life, "except a few dilapidated forts." Dr. Theal in his Abstract of the Records (Records, Vol. VII) points out that the official documents of the Portuguese Governments testify to the utter failure of their trade in gold. This failure was piteously but quite frankly admitted by all those most directly concerned in the "Conquest of the Mines." This "conquest" the records show was never effected, for the very simple reason that the Portuguese never penetrated as far as the ancient mines' area, concerning which they never obtained any information, apparently being altogether unaware of its existence. No rumour of large mines on the rock ever reached their ears, not even in the form of Arab tradition, or native legend. The records are absolutely silent as to the existence of any such mines.

Beyond the immediate vicinity of the Zambesi [river] they never penetrated. The entire period of their occupation was but brief. They only succeeded in obtaining a temporary and precarious hold on Manica and the northern portion of North Mazoe District, and even from these places they were repeatedly driven out. Yet the ancient mines' area extended beyond those districts for almost 500 miles (804 km) to the south and several hundred miles to the south-east and south-west, an immense country covered thickly with prehistoric mines sunk deep in the rock and completely filled in and buried in the course of centuries by a process of natural siltation.

De Lima, in his Possessdes Portuguezas (1859) in summing up the results of the "conquest" observes, "O illusoro Potosi de Chicova[27] ou 0 fabuloso El Dovardo de Quiteve”[28] but this applies only to that small portion of the country penetrated by the Portuguese. All modern Portuguese historians, without exception, confirm the story of the utter failure of the gold-trading ventures of the early pioneers of their nation in South-East Africa, and as did De Lima, describe the "conquest" as "illusory" and "fabulous."

In 1570 Father Monclaros, a local writer, reported, "Of the mines and abundance of gold and silver [of the country of the Monomutapa] many have written at great length, but the sum of what is known is much less than the reports which are current in Portugal" (III, 253). Again, "He [the king] had more favourable reports of the abundant riches of the realms of Monomotapa than were borne out by facts or came within our experience." (III, 202) Soares writes, "There is not so much gold in this country as has been reported." (I, 80-81), etc.

In 1619, over one hundred years after their arrival, the Portuguese were still engaged arranging for "the discovery and conquest of the mines [of the Monomotapa]" (IV, 161). In the same year it was stated, "there is no certainty of their existence" (IV, 160), and "experienced officials should first certify their existence." In 1622, presents are ordered to be given to the Monomotapa to secure his assistance in the facilitation of "the discovery of the mines." (IV, 184). In 1626 the Captain of Sofala is instructed "to search for the mines of Monomotapa" (IV, 194). In 1628 are fresh orders from the king for "search to be made for the mines of gold and silver" (IV, 218). In 1634 De Rezende writes, "Up to the present no gold or silver mines have been found." (II, 411). In 1667 Father de Barretto, a local writer, ridicules the existence of the gold mines and calls them "pretended mines" (III, 480). In 1697 the King of Portugal grants to Dominicans tithes of a reported mine at Sena, but taught by sad experience of misfortune, added, "if it is discovered." (IV, 496).

Thus, for one hundred and ninety years after the arrival of the Portuguese in the country they had never discovered the gold mines on the rock. As the Portuguese power inland became broken about 1760, there was only sixty years left in which to make the discovery, and this was not made even within that time. As late as 1719 the King of Portugal was doubtful about the existence of gold mines in the country, and he ordered inquiries to be instituted as to whether the Monomotapa (in 1607) did actually donate any gold mines to the Portuguese, "so that in the future, when there is a better opportunity, we may avail ourselves of his donation." (V, 72).

Summing up the history of the Portuguese possessions in East Africa during the seventeenth century, Andrade Corvo says, "We dragged out a sad existence, without progressing in colonisation, without developing commerce or industries, and without the famous gold and silver mines giving the marvellous results which were expected from them."

Professor Randall-MacIver’s attempt to twist the Portuguese records to fit in with his impossible theory is humorous indeed, but his assumption of a knowledge superior to that possessed by De Lima, Andrade Corvo, and the modern Portuguese historians is simply amazing. "O illusoro Potosi de Chicova! O fabuloso El Dovardo de Quiteve!"

The Prehistoric Gold Mines of Rhodesia – When was the gold extracted from the rock? Not between 900 and 1760 AD

PART II

Arab and Persian Period, 900-1505 AD

"The dates of the Arab and Persian settlements, which the Portuguese of the sixteenth century found facing the coast from Cape Delgado to Cape Guardafui, are known from the chronicle of Kilwa. The most ancient is Magadoxo, which was founded not earlier than the tenth century. Sofala itself as the chronicle states, was first colonised from Magadoxo, and there is, therefore, no justification for ascribing to it an earlier date than the eleventh century AD" — Professor Randall-MacIver, R.G S Journal, April 1905, P335.

"Zimbabwe, being the great distributing centre, must have owed its very existence to that trade with the coast first opened up by the Arabs of Magadoxo, the earliest possible date for any settlement there is the eleventh century." — Mediaeval Rhodesia, P86.

The historic records of the Portuguese, and the evidence presented by the mines sunk to depth in rock in Rhodesia, as we have seen in Chapter II, most conclusively prove that the overwhelming bulk, if not all, of the more than £75,000,000 worth of gold obtained from the mines on the reefs was most certainly not extracted between 1505 and 1760 AD. The evidences of the earlier historic period, 900 to 1505 AD, will now be examined in order to ascertain whether such an approximate amount was extracted from the rock between some indefinite time in the eleventh century and 1505, when the Portuguese became masters of the Mozambique and Sofala coasts and islands, or even since 915 AD when al-Masudi visited the coast.

East and South-East Africa before 900 AD

Since the Christian era, if not earlier, Zanzibar or its vicinity had been the place of resort for the Arab and Persian traders and the slave dealers of the Red Sea, the Persian Gulf, and the coasts of Scinde [Sindh] and Gazerat.[29] "The Periplus" [80 B.C.], says Mr Bent,[30] "mentions that the Arab settlement of Rhapta [believed to be on the Rufiji river delta] was subject to the sovereign of Maphartes, a dependency of Sabia or Yemen." M. Grandidier, in his monumental work,[31] quotes documents to show that the Comoros Islands, stepping stones between Madagascar and Rhodesia, were peopled in the reign of Solomon "by Arabs, or rather Idumean Jews from the Red Sea." Further, he says, "There is nothing surprising in the presence of an Idumean colony in Madagascar, for we know that from the very earliest times the Arabs of Yemen had frequented the East African seaboard at least as far as Sofala." Also, "the Jews and Arabian Semites were not the only peoples who had formerly commercial relations with the inhabitants of the African seaboard. From time immemorial these southern seas were navigated by the fleets of the Egyptians, probably even of the Chaldeans, Babylonians, Assyrians, Phoenicians, Tynans." And again, "From the earliest times the Indian Ocean was traversed by Chaldean, Egyptian, Jewish, Arab, Persian, Indian and other vessels."[32]

Much later, about 739 A.D., the Zaide Arabs from Arabia settled on the East African coast and extended considerably to the southwards, as far as, if not south of, the Zambesi, and also into the interior, eventually becoming by mixture of blood incorporated with the natives, till they became

hardly distinguishable from the natives.[33] From Sir Richard Burton's work[34] it would appear that Persian intercourse with South-east Africa had preceded by some considerable time the arrival of traders and settlers of Arab race, an hypothesis to which the author of The Bantu of the Tenth Century states colour is lent by other considerations of a philological nature.

Professor Dr Keane, in his reply to Professor Randall-MacIver’s assertion that the ancients had no knowledge of the southern regions, said,[35] "Mr. MacIver makes the hazardous statement that the ancients, even so late as the Roman Empire, had no knowledge of those parts, and he refers in proof to the Periplus of the Red Sea, which in those days meant the Indian Ocean. On the Periplus, fresh light has recently been thrown by Edouard Glaser, who has discovered that it was composed by a certain Basil of Alexandria, who was evidently much interested in the trade of the Far East. This Basil describes the navigation of the east coast of Africa, down not merely to Cape Rhapta, but all the way to the extremity of the continent. He mentions the island of Menuthias, which must be absolutely identical with Madagascar and with no other island on the east coast. He says, further, that the Indian Ocean bended round westwards, and eventually mingled with the waters on the other side of Africa — that is, the Atlantic Ocean. Now, Basil lived in the time of Nero; consequently, it was then known that Africa was an island. But it was also known to be an island long before that period. Herodotus tells us plainly (IV, 42) that King Nechos, about 610 BC, equipped an expedition conducted by Phoenicians, which sailed round Africa in three years and came back by the Pillars of Hercules to Egypt, having started from the head of the Red Sea. He says, moreover, that Africa was thus proved to be circumnavigable, all except the isthmus of Suez. Mention is made by the author of the Periplus of the 'Fire men/ a most appropriate epithet for the inhabitants of these Comoro Islands, where active volcanoes still exist almost in sight of Rhodesia. [!] Therefore, there was no conceivable trouble for the Arabians and the Persians, who had fleets in the Indian Ocean, to reach these regions in quest for gold."

Hall: The East coast of Africa from Mogadishu to Cape Correntes traded by the Arabs prior to European arrival

                                                                 Hall: The Great Enclosure

             Looking in during excavations                   Unusual perspective from the Conical Tower

Historic South-east Africa

About 930 AD Arabs from the Persian Gulf founded Magadoxo and Brava on the east coast, and during the eleventh century spread to Sofala. After the settlement of Magadoxo, and about the end of the eleventh century, Persians settled at Kilwa, and according to the Chronicles of Kilwa in the course of time, about 1314, they usurped the Arab sovereignty and commerce along the whole Mozambique coast from Zanzibar, at least as far as and including Sofala, if not to Cape Correntes. On the arrival of the Portuguese at Sofala (1505) the Persians had been in possession of the coast for about one hundred and ninety years, De Goes stating that the King of Kilwa's dominions "extended from Cape Correntes [south] almost to the town of Mombasa [north], a distance of 400 leagues of coast, besides a number of islands which lie along it"[36] and the records also state that the Persians or "Moors" of the Sofala district had, sometime before the advent of the Portuguese, rebelled against the King of Kilwa and become independent (I, 16.)

But, though Professor Randall-MacIver claims that "there is no justification for ascribing to it [Sofala, town and port] an earlier date than the eleventh century AD," the date of the first historic mention of the political foundation of Sofala does not and cannot by any means set a bar or limit to inquiries as to the earliest gold and ivory exporting activities from the coast or land of the country known centuries earlier as Sofala [Arabic = "low land" or "low coast"] for Sofala was the name of the entire coast and also of a portion of the hinterland, and much later the name was bestowed upon the port in the same way as the names "River of Sofala" and "Mines of Sofala" were at a still later date respectively employed for the Buzi river and the mines of Mocaranga [Rhodesia].

If, therefore, Professor Randall-MacIver’s claim as to the date of the founding of the settlement at Sofala were intended to set such a bar to inquiries as to more ancient activities on the Sofala coast, then such a suggestion can only provoke a smile; for, as will be seen later, there are ample historic evidences that for at least two centuries before the first historic mention of the political settlement of Sofala by the Arabs of Magadoxo, and how much earlier no one can say, gold, ivory, and amber were being regularly exported to Arabia and India from the coast of Sofala, and also forwarded to China.

Al-Masudi, "the Arabian Herodotus" visited East Africa some time before 915 AD,[37] later journeying to Malacca and the shores of China, and the Red and Caspian Seas, publishing his Golden Meadows in 943. In Mr. W. Hammond Tooke's valuable treatise on The Bantu in the Tenth Century which gives the translation of al-Masudi description of these regions and of its inhabitants and export trade, we find al-Masudi mentions Sofala. "They [the Zeng[38]] established themselves in this country [east and south-east coast] and extended as far as Sofala." Also, "Just as the Sea of China extends to the country of Sila [Japan], so the limits of the Sea of Zeng[39] are the countries of Sofala and of the Waq-Waq"[40] and "navigators advance over the Sea of Zeng as far as the island of Kanbalu and Sofala. The merchants of Syraf are in the habit of navigating this sea."

Al-Masudi also states that the best amber was found along these coasts, and that it was exported to Iraq and Persia. In the Portuguese period amber was one of the principal articles of export from the Sofala coast. But al-Masudi also describes an immense trade, evidently established long before his time, in ivory, which was sent from these coasts to Oman and "exported thence to China and India," and gives the value of the tusks which were sent to Bagdad. He speaks of the country of Sofala as where the Arabs of his time went habitually to obtain gold from the natives.[41]

Thus within a few years of 900 AD we have most definite historic evidence that almost, if not quite, two hundred years before the first historic mention of the political settlement of the Magadoxo Arabs at Sofala there was an extensive and long-established export trade in gold, ivory, and amber being carried on between the country of Sofala and Arabia and Persia. How long this trade had been in existence before al-Masudi’s time no one can say, for ancient trades with remote parts of the world, such as he mentions, were not wont to be of sudden growth and development. At any rate, long before al-Masudi wrote, far-distant China was receiving ivory from the Sofala coast. Most probably, as many scientists affirm, and this recently, and in the face of Professor Randall-MacIver’s assertions, the trade in al-Masudi’s time was but the survival of a trade in gold and ivory which might very well have existed here even before the commencement of the Christian era.[42]

Sir Harry Johnstone, one of the greatest living authorities on African peoples, especially on the Bantu, in replying to Professor Randall-MacIver’s conclusions, stated,[43] "I should say my own convictions remain relatively unshaken that there was at a period at least as early as the birth of Christ — I believe earlier — an incursion into this country [Rhodesia] of a Semitic race of teachers. I cannot otherwise explain the goldmining, the soapstone birds, the phalli, and the several other features in 'those remains ' which are so utterly unlike anything that has ever been made by any race of Bantu negroes. Neither is there any evidence to show they could have been made by Hottentots. My own belief is that the presence in Africa, south of the Zambesi, of Bantu negroes is a relatively modern phase. The first violent eruption of the Zulus may have driven away the pre-Islamic Arabs, and yet not have completely caused the goldmining to cease. I do not think Mr. MacIver having found Nankin pottery in the foundations[44] of one or more of these structures should induce us to give up too readily the belief that at some period of possibly more than 2,000 years ago, Arabians, or people from the direction of Arabia, did make their way down the coast of Africa in search for gold."

Professor Randall-MacIver attempts to ignore al-Masudi’s record and to limit and underrate in such a peremptory manner, and without justification, the discoveries and traffics of the ancients, and the influences which, from surviving local conditions, they evidently appear to have exerted by contact with the barbarous inhabitants of South-east Africa, is somewhat astonishing. Especially is this the case seeing that South-east Africa was far nearer to Arabia than China and Japan, and other remote places so frequently mentioned by ancient geographers, and that the monsoon both to and from the Sofala Sea would, as navigators of to-day assert, very greatly facilitate intercourse between Arabia and the Mozambique coast.[45]

Before "Sofala' Sofala was

But long before the first historic mention of the political establishment of the Magadoxo Arabs at Sofala at some time in the eleventh century, Arabian and Persian writers had frequently alluded to the ivory and gold trade of Sofala. Even before Magadoxo itself had been established (about 930 AD) al-Masudi, as we have seen, had described the country of Sofala, its people, and its gold and ivory export trade. Ibn Ay Wardy[46] in 957 had also described Sofala 't il Dhab ["the low land of gold"] stating that "the most remarkable produce of this country is its quantity of native gold."

Further, at the time which Professor McIver claims as the first settlement at Sofala, a Persian geographer, Albyrouny, had already described the Sofala coast as far as Cape Correntes and mentioned the standing export trade in gold and ivory which existed between Sofala, the Red Sea, India, and China. Now follows in order of time the first mention of the political settlement of the Magadoxo Arabs at Sofala at some time in the eleventh century there being, Professor Maclver claims, "no justification for ascribing to it an earlier date." Then follows in the twelfth century Idrisi’s mention of the established trade of "Sofala " with India, and of "the abundance of gold in the mountains behind Sofala."

But before "Sofala," Sofala was. History explicitly shows that the Magadoxo Arabs did not originate the gold export trade of Sofala, for, as we have seen, Arab and Persian records prove that it had been in existence at least two centuries earlier and then was a continuance of a still earlier trade. The merchants of Oman and Syria, in the course of their trade with the country of Sofala, must have used some ports on the Sofala coast, and the consideration of the location of the actual sites of such ports or trading stations is altogether immaterial to the argument as to when the first contact of the natives of Sofala and of its auriferous hinterland with Arab, Indians, or other foreign peoples took place. Most probably the sites were at the mouths of the Zambesi, or at the mouths of the great waterways into the interior of the Sofala country provided by the Pungwe, Buzi (near Sofala), Sabi, and Inhambane rivers, or on Sofala Bay, if not at the very site of Sofala itself. All that we need concern ourselves with at present is the historic fact that the intrusion of foreign influence on the natives of ancient Rhodesia, brought about by trading, was in full operation before 900 AD, and that even then it had already been exerted for some indefinite period earlier.

Unsettled State of Sofala, 900-1505 AD

The history of the Sofala country and its hinterland from 900 to 1505 AD is one of unsettlement and wars. Abou-Zeid Hassan, who wrote in the tenth century, mentions wars as going on in the interior. Al -Masudi records[47] that just before his visit to the Mozambique (915) a great wave of the barbarous Zeng, known to be the Bantu, had extended from the north and settled on the coast and hinterland of Sofala, whose advent Mr Bent, and also the mining experts who have examined the ancient rock

mines in Rhodesia, as well as authorities on the Bantu race, believe caused the final abandonment of the rock-mining operations in what is now known as Rhodesia.[48] Of one thing we may be certain, whether this conjecture as to the abandonment of the rock mines be correctly founded or not, such a descent or extension of the Bantu as is mentioned by al-Masudi must have dislocated for a very considerable period the mining and the washing for "native gold" and the export trade from these coasts. Whether the appearances and evidences noticed by the subsequent Arab writers warrant any inference that the natives of 900 to 1505 AD were skilled in rock-mining will be considered later.

But the period from the eleventh century to 1505 comprises two different periods of occupation, or, more correctly, of connection by trading — that of the Arabs of Magadoxo and that of the Persians of Kilwa, which latter during this period usurped the sovereignty and commerce of the Arabs. The Magadoxo Arabs are shown to have won their possessions by the sword, and their supplanters the Persians, as shown in the chronicles of Kilwa, were long engaged in wars with their rivals before they were able to establish a footing on the adjoining islands along the coast. Both rivals needed to "fortify themselves against the CarTres," and both were continuously at war with coast tribes, while both kingdoms were subject to internal rebellions and civil wars. How trading in gold fared during these periods may well be imagined.

Further, this period was also marked towards its close by the disruption and dismemberment of the most powerful and most extensive Bantu kingdom yet known to research. The large kingdoms of Sabia and of the Quiteve successfully rebelled against the Monomotapa, the King of Mocaranga, which country had yielded the gold supply. The civil wars which preluded this disruption are referred to by tradition in the records, as also the subsequent state of chronic warfare between the Monomotapa’s and their erstwhile rebellious kingdoms, all of which are matters of Portuguese history. On the arrival of the Portuguese in 1505 they found the MaKaranga to be engaged in "continual wars inland," and that the former gold export trade had reached vanishing point (see ante, Chapter II, Unsettled State of the Country, 1505-1760). The much-vaunted gold-trading port of Sofala too had become but "a village" (Barbosa, I, 105), and its inhabitants but "low-class Moors " who were "more native than Moors."

Natives and Gold, 900-1505 AD

The export trade in gold and ivory from Sofala could only have originated in the first instance by a foreign demand, for al-Masudi (915), speaking of the natives of the land behind Sofala, stated they never used for their own purpose the gold and ivory with which the country abounded. "They employ iron in their ornaments instead of gold or silver." AP Wardy (957) wrote, "The most remarkable produce of this country [Sofala] is its quantity of native gold…in spite of which the natives adorn their persons with ornaments of brass." Idrisi, in the twelfth century, wrote, "In all the country of Sofala they find gold in abundance. Nevertheless the inhabitants prefer brass, making their ornaments of it."[49]

But al-Masudi described the natives of his time (915 AD) exactly as we find them to-day, so truthfully that from the mass of evidence he and other Arab and Persian writers provide "we may," as Mr. Bent says, "affirm that for a thousand years at least there has been no change in the condition of this country and in its inhabitants." This opinion but correctly reflects the unanimously expressed belief of every single student of the Bantu people.

The natives, "black and naked," "barbarians," and "cannibals," of 915-1505 were not of any higher intelligence than were the natives of 1505-1760 as described in the Portuguese records. Could the natives of 915-1505 have been the "skilled miners and assayers" who worked on the rock at depth in Rhodesia ? There can only be one answer to such a question. The Portuguese (1505-1760) confessed that they themselves did not possess the mining knowledge to enable them to work the mines on the rock in Mocaranga. That being so, one cannot suppose the natives of 1505-1760 to have been more skilful and intelligent than the Portuguese; neither can one imagine, with historic record testifying to the contrary, the barbarians of 915-1505 to have been more intelligent than the natives of 1505-1760.

From the Arab writers, and from the evidences obtained by the Portuguese on their arrival on the Sofala coast as to the low condition of the degenerated half-breed Moors, and the vanishing point which the gold export trade had then undoubtedly reached (see authorities cited, Chapter II), and from the indisputable evidences presented by the actual rock mines, as will be seen later, it is impossible to conceive that the mines at depth on the rock in Rhodesia had been sunk between 900 and 1505.

The early Arab writers show that the gold was brought to the coast by the natives themselves, just as Barbosa (1514) in describing the trade of the Moors, states that "the natives bring to Sofala the gold which they sell to the Moors," and al-Masudi (915 A.D.) had previously stated the Arabs of his time "obtained gold from the natives." We learn that the Moors were restricted to the coast.[50] There were no Moorish settlements and markets in the interior, though at a very late date, and only just before 1500 (VII, 478), the Moors had a settlement near the present Sena on the Zambesi within easy reach of the coast, this being their only station inland. The natives of those times consequently had to rely upon their own initiative. They had no foreign directors of mining. The Arab writers, as well as the Chronicles of Kilwa, are absolutely silent as to any rock-mining on the part of the natives, and nothing they contain in the slightest suggests, much less warrants, such an inference; while the records of the Portuguese unmistakably declare that no rock-mining operations were being carried on between 1505 and 1760, nor were there any traditions of such operations (see Chapter II).

The gold trade of Sofala from al-Masudi’s time down to 1760 was, as described by the old writers, exactly similar in every respect to the gold trade carried on at the west coast of Africa. In describing the washing soil and sand in rivers for gold by the natives on the east coast nothing different has been written to what has been written to describe the gold-winning operations on the west coast. A description of one is an accurate description of the other. Both used quills, as both use them to-day, but quills are not very suggestive of rock-mining operations. Both were of Bantu stock, but why the east country natives should, as Professor Randall-MacIver’s argument would suggest, be so far superior in intelligence and mechanical skill to the natives on the west coast is not explained. The alluvial areas of Rhodesia have been most extensively worked in diggings in the soil by natives at some much later time than that during which the mines on the rock were sunk.

                   Randall-McIver: ‘Philips’ ruins in the Valley, Great Zimbabwe                                                   

The Mines "Ancient"

But the mines were declared by early Portuguese writers to have been "ancient," and of a profound antiquity. The dozen references in the records to certain of the ruins being "ancient" and "very ancient" give a clue to what these writers meant when they stated the mines were "ancient," i e. their frequent references to their having been built in Solomonic times. In three instances regarding mines they credit them to the time of Solomon, while for the others they employ the description "ancient " and "most ancient." De Conto, in referring to the mines of Masapa, states they were so old that they were the mines "whence the Queen of Sheba obtained the greater part of the gold she offered for the temple of Solomon" (VI, 367). De Barros states, "The mines…are very old, and no gold has been extracted from them for years," and contends the region is the Agysymba of Ptolemy (VI, 268). He further states that the mines of Toroa[51] were "the most ancient known in the country" (VI, 267). Dos Santos uses the expression, "In the midst of the ancient mines," and he relates the Arab tradition that the gold of the Queen of Sheba came from these mines, the antiquity of which he himself also ascribed to Solomonic times and connects their age with that of the Fura Ruins,[52] which he describes as "some fragments of old walls and ancient ruins of stones" (VII, 275). Father Dos Santos visited the interior, and the district Manzovo (Mazoe), where these ruins and mines are situated. He is considered to be the most accurate of all the Portuguese writers. Dr Theal states that he was by far the best educated man who wrote upon this country, and also the most reliable. Livio Sanuto, in his Geographia dell Africa, declared the mines of Manica were "ancient." There are several other references in the records to the mines being "ancient."

Prehistoric Rock-mining

The evidences from the rock mines in Rhodesia prove that the prehistoric miners possessed only the crudest mining appliances, and yet they must have mined from depth far more than 100,000,000 tons of reef, much of which was rock of a refractory nature, this estimate being made by admittedly competent mining experts before half the area of the ancient mines had been discovered.[53] Nowhere else in the world are such extensive prehistoric gold mines sunk to depth in rock to be found as in Rhodesia.

In many instances the mines were sunk to water-level, beyond which the old miners could not sink. Many average from 70 ft. (21 m) to 150 ft. (45 m) in depth, while the reefs have been hewn out at these depths for lengths of 300 ft. (91 m) and 500 ft. (152 m) to 1,750 ft. (533 m) the Mystery Reef showing a worked out length of 4,000 ft. (1,219 m) at depth, while the Urungwe Reef shows distinct lines of deep workings extending to 7,500 ft. (2,148 m) But an infinitude of other instances could be cited from reports of mining surveyors of the Rhodesian companies, which are accessible to all, and from which, almost eight years ago, I extracted and published descriptions of over one hundred ancient rock mines found in widely different parts of Southern Rhodesia.[54]

But to reach these lower measures, and to work them, the ancients removed the rock from long distances on either side of the reefs at these depths and along their lengths, in one instance to a width of 150 ft. (45 m), thus making huge open workings. To do this they must have taken away rock, other than the gold-bearing reef, to an amount four or five times greater than the actual reef which they removed. This is a feature seen over an area of 700 by 600 miles (1,126 x 965 km) Many mines, however, were worked with shafts, vertical and incline, and adits, the pillars of reef left to support the roof or "hanging wall" at depth still remaining.

Every sub-district on the gold areas of Rhodesia yields evidences of millions of tons of gold-bearing rock having been mined in some prehistoric time. From the M'topata district and within a small area "no less than one and a half million tons of reef" are reported as having been taken; in the Umsweswe district "millions of tons of reef" have been extracted. On one property, located on the site of an ancient mine, and now crushing with declared results of over 10 dwts. (131 gm) to the ton (1.01 milliers) over the plates, and the tailings only estimated, "are several ancient workings on the same line of reef, and these ancient workings are in size not above the average of those generally found. One chamber of one of these numerous workings on the property gives an astonishing result. Taking the length, breadth, and depth of the reef extracted by the ancients from this chamber with the fire assay values of the reef untouched, disregarding altogether the possibility of the ancients having also worked rich 'leaders' (and these are numerous on the property), and making liberal allowance for loss caused by crude methods of recovering gold, we find that the gold extracted from this one chamber out of many on the property was, in present-day value, not less than £32,000 sterling."

Mr. J. Hays Hammond, a gold-mining expert of worldwide reputation, in the early days of 1894, before half of the ancient mines area had been discovered, and after an inspection, as he states, of only a few of the then known ancient districts, reports : "That an enormous amount of gold has been obtained from these workings in the past is, however, unquestionable. Millions of pounds sterling worth of gold has undoubtedly been derived from these sources." [55]

But what heightens the mystery is that while the ancients worked both high-grade and low-grade rock with equal persistency, and most laboriously worked millions of tons of exceedingly low-grade ore, "they picked out rich shoots, patches, and pockets with marvellous cleverness, proving themselves most skilful miners and metallurgists."

As evidence that the skill of the ancient prospectors is fully recognised by highly trained and experienced mining surveyors of today, it may be stated that quite two-thirds of all the registered gold claims in Rhodesia have been pegged on the lines of ancient workings ; "the experience in this country now amounts to this, that, given a regular and extensive run of old workings on a block of claims, it is almost a certainty that a payable mine will be found on development of the ground" {Rhodesia Chamber of Mines' Report 1899) Mr Walter Currie, consulting engineer, also states that his "experience has invariably proved that where old workings exist, they indicate more or less accurately the length of the pay shoot below." That these old miners were skilled is beyond doubt. Their skill was not "gradually evolved" in this country, for it is displayed in its most perfect excellence, chiefly in those rock mines which are admittedly the most ancient rock mines in the country. It was introduced here already perfected, and by people who were acquainted with mining in Arabia or India, or both.

These ancients knew exactly the value of the reef they were working upon, though the rock, not showing a speck of gold, and presenting no outward evidence of its value, could only have been estimated by assay. But they were also able to remove reef, which is of such a refractory nature, that today mining companies working on identically the same reef, and mining the same class of rock, can only deal with it by the liberal employment of dynamite, the cost of which is always a heavy item in the "working expenses" of almost every mine in Rhodesia. When one considers the enormous amount of rock removed by the ancients, and the difficulty experienced in modern operations on the same sites as worked by the ancients, one must be fairly staggered with surprise, not only by their skill, but by their industry.

No chief, at any rate since al-Masudi time (915 A.D.), however powerful he may have been, could have called his people to work such mines on the rock on this vast area. Chiefs cannot create "culture." Such an academic suggestion, recently most seriously made, has considerably amused mining authorities both in Rhodesia and on the Rand. On the alluvial or shed-gold areas local Bantu chiefs probably did direct the soil-washing operations of their people, though the records state the natives paid no tribute in labour except to "weed, dig, sow, and gather" the chief's crops (VII, 222), the alluvial workings on the surface being obviously of a much later date, the evidences, as shown in Chapter II, pointing to their having been worked by very old MaKaranga people. But any gold so obtained, whatever its amount may have been, does not account for a single pennyweight of the more than £75,000,000 obtained from the rock of the country.

Present Condition of Rock Mines

If any further evidences were needed in order to establish the prehistoric character of the mines in Rhodesia, it is to be found in the fact that all these gigantic workings in rock are now buried in naturally silted soil, there having been no artificial filling-in of these old workings. There is not a clean ancient stope at depth to be found, for so completely buried out of sight are the great majority of the large mines, that even mining engineers and qualified prospectors have passed and repassed close to and across their sites and yet have never been able to notice anything suspicious in the nature of the ground to suggest their proximity to ancient mines. Yet many of these mines are in positions where filling in by natural siltation would be an exceedingly slow process, requiring, one might very reasonably imagine, centuries of time.

So entirely buried and obliterated are some of these ancient mines that even now, with the skilful methods of modern experts, it is almost impossible to locate their sites, while the presence of others is only evidenced by slight unevenness in the surface of the ground. Yet these mines are silted up naturally from a depth, in some instances, of from 150 ft. (45m) to the levels of their mouths and the huge spaces once occupied by deep and yawning concavities in the rock are now level plains covered with forests of full-grown trees, including the non-indigenous mahobohobo, which is only to be found on the ancient mines' area, while fresh ancient mines are still being discovered every month.

Many of the modern shafts are sunk straight down for from 50 ft. (15 m) to 100 ft. (30 m) through this silted matter, which in the course of long centuries has set so hard that it will allow of such shafts being sunk through it. Descending in a cage some 50 ft (15 m) to 70 ft. (21 m) or more through this natural siltation one can examine its composition and where it ends on the ancient stope one can notice just where the ancients left off work, below which point the shaft passes deeper through the virgin rock.

But the condition of these rock mines also shows that the mining operations covered many centuries of time. The whole face of the country for 700 by 600 miles (1,126 x 965 km) testifies to ancient activities of an extent almost beyond human comprehension. Mr J. Hays Hammond remarks, "It is not improbable, in fact the consensus of opinion is that these reefs have in most cases been worked at different periods." All the areas were not worked at the same time. The work was progressive, and as one area was worked out and exhausted another area was prospected, worked out and exhausted and area after area was mined, till eventually territories several hundreds of miles in extent became honeycombed to water-level with deep ancient workings, all testifying to most tediously slow and unmolested operations carried on stage by stage. The expert mining opinion is "that sections of the country were successively brought under the influence of ancient Asiatics, and their operations extended over many centuries."

Further, these vast operations testify to continuous and unbroken centuries of time devoted to rock-mining. There are no evidences of rude dislocations in labour caused by wars or dynastic changes, or by the caprice of local potentates in various parts of the huge area. No single Bantu chief has been known to rule over such an extent of country and yet identical methods of working were adopted in every single mine. The venture, for it appears to have been a single venture, must have been a sovereign one, slave-labour being strongly presupposed and bartering unnecessary. A visit to these numerous workings makes "loincloths" "glass beads" and "coconuts" for which the later natives, from al-Masudi time (915 A.D.) until 1760, traded the gold, and also the "quills" for holding gold dust, somewhat trivial and ridiculous. Most probably there was no trading required for the gold so extracted. At any rate, there is not any suggestion that there was mining by individual natives, or even by individual tribes of Bantu, in the different territories on the area covered.

Hand-crushing the Rock

But what possible evidence could be produced that over 100,000,000 tons of rock were ever crushed and ground by hand by any Bantu people "not undifferentiated," as Professor Randall-MacIver asserts, "from other South African people," since al-Masudi’s time? We know exactly what a ton of average quartz ought to yield under modern stamps and over the plates. But the ancients had no batteries, no roller-mills, no dry-sorters, no cyanide reduction plant, and none of the modern appliances for reducing the ore. With their crude methods and laborious processes of hand labour, they must have failed to obtain anything approaching the full value of gold out of every ton of rock they crushed. Therefore we can safely increase the estimate of the tonnage of rock they mined and crushed even without allowing for crushing "blank," of which, of course, they must have had more than the usual quantity.

Anyone watching natives, on their knees, grinding corn on a flat stone, a two-handed process, knows well how exceedingly laborious grinding corn actually is. But to contemplate the crushing by hand of far more than 100,000,000 tons of rock,[56] much of which was refractory, and of exceedingly low grade, by Bantu people as we know them to-day, or as al-Masudi and the Arab writers and the Portuguese records describe them, is simply impossible. There is no reference since 915 AD to any such a process; there is not a particle of local evidence forthcoming to show that such crushing and reducing operations ever took place within historic times. Historic references, expert mining opinion, and authorities on the Bantu unanimously and emphatically repudiate any such suggestion.

                                                           Hall: The Great Enclosure

    Parallel Passage within the Great Enclosure         Exterior wall of the Great Enclosure

Present Bantu not the original Rock-miners

Whether the rock mines at depth were the work of any Bantu people as labourers is problematical. Many authorities consider this impossible. But, at any rate, they were not the work of any present known Bantu people, who, as Professor Randall-MacIver claims, "were not undifferentiated from other South African people." Yet where to-day are such other Bantu who have ever been known to work many scores of millions of tons of rock which, without assay, they could not have known contained a particle of gold? Such an off-hand assertion by one totally unacquainted with Bantu peoples, that Bantu of later times than the tenth century sank the rock mines in Rhodesia, and this in the face of a great volume of expert mining evidence, the expressed convictions of students of the Bantu, whose lifelong experiences of Bantu people qualify them to speak with far more than mere academic authority, and the evidences to the contrary contained in the Portuguese records, unmistakably places the onus probandi on Professor Randall-MacIver.

At present those who, with most excellent reason, hold the contrary view can very well afford to wait with amused interest until Professor Randall-MacIver’s first attempts to make out his case. Had these rock-mining operations been carried on by any present Bantu within the comparatively recent period as he suggests, the impression on the natives caused by such operations over so vast an area, and covering centuries of time, must have been such as to have been obvious on such Bantu of to-day. But where are such impressions?

Even historic records of five hundred years ago prove conclusively that in those remote times the Bantu had received no such impression. Quite the contrary. Some writers say the royal insignia of the Monomotapa was an agricultural hoe, "as a sign that he is a cultivator of the land." The records show that the "mining" in earth was with plantation mattocks! The only tribute in labour paid to their kings by the natives was "to weed, dig, sow, and gather the crops reserved in their village for the king. This is the only tribute they pay to the king, and nothing further." (Dos Santos, VII, 222) [**]

Dos Santos, who spent eleven years in the country of the "mines" which he describes from personal visits as alluvial earth-washing areas, in giving a list of native articles does not include any mining tools (VII, 209), and among the implements of the MaKaranga of to-day there is not a single object which by any possible straining can be claimed as a survival of any rock-mining tool (see Chapter II, Natives and "Mines" for definite historic statements that the MaKaranga of 1505 to 1760 not only did not mine the rock, but had no implements for such a purpose.

But the MaKaranga can be shown to have been far less migratory than any other Bantu people south of the Equator. Today they occupy in exactly the same territory as they were occupying over five hundred years ago. In 1505 their kingdom had already been disrupted, but how long before that time their kingdom had existed no one knows, still less when the MaKaranga founded it, for powerful Bantu kingdoms are not usually born readymade in a night, or still less as to the time they first arrived south of the Zambesi. We can only judge they were occupying the same territory some centuries before the Portuguese arrived by three evidences —

1. That according to Bantu authorities, and evidences presented, the MaKaranga had arrived south of the Zambesi at so early and indefinite a time before certain other later intrusions of Bantu from the north, with whom they had in some former period, and in the north, been in intimate contact, that their vocabularies had become entirely distinct, and their racial characteristics surprisingly different.[57]

(2) That in 1560 it is recorded that generations of Monomotapa’s had been buried in the Beza Ruins (after the abandonment of the buildings by their original occupiers), which "serve them for a cemetery" (III, 356)

(3) That of some hundred placenames mentioned in the earliest records the bulk are of purest Chikaranga, and very many of these remain identically the same names for rivers, mountains, districts (some of the latter still surviving in the same localities today) as are shown on modern maps, and are in use by the present MaKaranga (see p. 412). Such topographical nomenclature, covering a vast area of country and enduring for at least five hundred years, bespeaks a very long occupation of MaKaranga, extending back to some indefinite time long prior to 1505.

If, therefore, the MaKaranga, being such a pre-eminently conservative people, and having occupied the same territory for the best part of a thousand years, were the miners and hand-crushers of over 100,000,000 tons of rock from depth, in the comparatively modern time claimed by Professor Randall-MacIver, and which the records emphatically contradict (see Chapter II), then we may be safely assured that the MaKaranga ought to present obvious and indelible impressions derived from such alleged operations, and what applies to the MaKaranga would apply, so Bantu authorities contend, with still greater force to any ordinary Bantu people. But these impressions, most patently, do not exist. Yet according to Professor Randall-MacIver’s dates, were he able to substantiate them, the MaKaranga, and no other people, must have been responsible for the rock-mining operations.

               Randall-MacIver: Great Zimbabwe Hill Complex

Rock Mines and Ruins

But it may be asked: If these rock mines are of so ancient a period, are not the ruins consequently of similar age? This by no means follows. The foreign demand for gold from prehistoric Rhodesia came from the east coast. The vast amount of gold extracted was never employed for local purposes. The country was, as mining experts have shown, mined stage by stage, area by area, until the operations extended inland 600 miles (965 km) or 700 miles (1,126 km) from the east coast. It is most improbable that Great Zimbabwe at 250 miles (402 km) from the coast, and [Mount] Fura at 300 miles (482 km) and Beza at over 400 miles (643 km) from the coast were erected until very long after the rock-mining operations had been in full swing in these inland territories.

The originating cause of the rock-mining activities was undoubtedly the demand for gold, and the influence from outside, whether direct or indirect, which finally resulted in the adoption of stone building, was in all probability not felt in the interior until long after the mining and trading in gold with the coast had been firmly established as a permanently payable industry. The foreign influence was first manifested in mining. Gold was the primary consideration, stone building being merely secondary and consequential, and therefore of later date.

Personally, I have always agreed with those authorities who, having long been acquainted with both ruins and mines, have always claimed that the age of the Zimbabwe Temple, old as it undoubtedly is, does not determine the date of the commencement of these ancient gold-mining activities, that many of the rock mines are far more ancient than Zimbabwe, but that the Zimbabwe Temple represents a resultant phase of intrusions of influence, not necessarily of dominance, into these territories of both Arabs of South Arabia and of people from Persia and Western India in ancient times.

On the one hand I cannot, after working at Zimbabwe for three years, accept the age assigned to the Elliptical Temple by Mr Theodore Bent, viz. 1100 BC and, on the other, I most firmly believe, as shown later, that the evidences, as far as they have been secured, are overwhelmingly opposed to the acceptance of the comparative modernity assigned to it by Professor Randall-MacIver, though his estimate undoubtedly applies, as has long been contended, to certain of the poorer buildings of obviously late squatters at Zimbabwe.

    Randall-MacIver: part of the Hill Complex at Great Zimbabwe

Further, it must be borne in mind that not a single one of the hundreds of ruins in our country has been examined (see Ruins Unexplored, Chapter XIII), that not nearly a tenth part of Zimbabwe itself has yet been explored, while there are several ruins larger, but not so intact as Zimbabwe, some believed to be even older than Zimbabwe. If, therefore, a very small portion of Zimbabwe only has yielded relics— stone birds, phalli, "cup and ring" linga, carved stone beams, immense soapstone bowls, and a wealth of most chaste gold ornaments, as to the origin of which antiquarians are by no means agreed, what may be anticipated when more of the ruins are examined? To dogmatise on our present partial evidences is imprudent, and suggests a lack of scientific balance, and a mere game of "archaeological Bridge."

Moreover, no final solution of either mines or ruins problem can be accepted until explanation is forthcoming as to (i) the Semitic features, superior intelligence, larger brain, and their small hands and feet, small pelvis bones and narrow hips, most polished speech, peculiar construction of language, and great difference in vocabulary of the MaKaranga, and their more than two-score customs of distinctly Semitic origin, all of which differentiate them from the ordinary Bantu, and are strongly indicative of a foreign influence, their customs being suggestive of a decidedly pre-Koranic origin, and (2) the form of ceremonial once practised at Zimbabwe, which all Bantu students affirm is unknown among any past or present Bantu people.

Professor Randall-MacIver’s Impossible Theory

The Arab, Persian, and Portuguese records, as also the local evidences, clearly show that the Magadoxo Arabs and Persians of 1000-1505 AD had no settlements inland, it being most definitely stated they were restricted to the coast, some of the expressions used being, "the gold which the natives bring to Sofala," also "the gold which the Moors obtain from the natives" and several other expressions all to the same effect. Therefore it cannot be conceived that Arabs and Persians of 1000-1505 originated or suggested to, or still less taught the inland natives any rock-mining operations. But we are told by Professor Randall-MacIver "there is no justification" for supposing any intrusion of influence brought about by trading, earlier than that of the Magadoxo Arabs in the eleventh century. Yet we are seriously invited to believe that as late as the eleventh century a branch of the great Bantu people — a race most notorious as possessing extraordinary conservative characteristics, and who had never been known to mine rock or dress granite for building in the territories from which they had arrived — had suddenly conceived at the back of its Bantu mind and ultimately developed a high degree of skill both in rock-mining and building construction, and carried on operations on so extensive a scale, had within that period, 400 years, as suddenly become so ignorant that they possessed no trace or impression caused by such operations, and had no shred of tradition of ever having carried them on, they placing everything to the credit of the "Devil" for in comparison with their power and knowledge it did not seem possible to them that they should be the work of many. Nor had the natives of some indefinite time long prior to 1560 any tradition concerning the people who had been occupying the Temple for centuries.

Thus, according to Professor Randall-MacIver, within four hundred years are crowded, first the conception in the Bantu mind, and then the development to a high state of perfection, its display in operations on a scale so gigantic as to be almost inconceivable, its gradual deterioration, and finally its utter oblivion. Such an hypothesis amounts to a reductio ad absurdum.

Mr W. Hammond Tooke, in his work, The Bantu of the Tenth Century, states, "Nor can we accept it [Professor MacIver’s theory] more readily when we remember the Chronicle of Kilwa, with its record extending over the period we are considering, written up probably by Mahomedan scribes residing at Kilwa, not many days' sail from Sofala, with which we may assume there was a frequent intercourse, is absolutely silent as to rock-mining or stone building in Rhodesia."

"It is scarcely credible," he continues, "that, within a period equivalent to that between the building of the Town House, Capetown, and our own day, these remarkable structures [rock mines at depth] should have been built [sunk], inhabited [mined], utilised [exhausted of gold reef], deserted [completely buried in natural siltation], left to ruin [no surface traces existing], and all knowledge of them so completely forgotten as to be replaced by superstitious legends."

This, as is shown, is an impossible theory, but it serves to remind us that so intricate a question as this which Rhodesia presents cannot be solved by a hasty and superficial examination -  even though conducted on the most scientific principles.

But Professor Randall-MacIver frankly admits that during his brief visit to the country he was unable to examine any of the rock mines and that his sole purpose was to approach the Rhodesian problem from the standpoint of an archaeologist. The problem of an ancient intrusion into this country will never be satisfactorily settled on the lines of archaeology alone — and serious exception is taken on archaeological grounds to the great majority of his conclusions— and will not be until the ethnological, philological, anthropological, and archaeological researches have been completed and even then, not until the happy medium of all these studies can be struck. Rock-mining and stone building must be considered together. Unfortunately, he has departed from his archaeological standpoint by emphatically limiting the gold export trade of this country to between the eleventh and fifteenth centuries and has repudiated all possibilities of there ever having been an ancient intrusion of foreign influence into South-East Africa.

But in the opinion of the most able experts of England, France, and Germany, as recently expressed in various terms," Professor Randall-MacIver has by no means finally closed the door against the discussion of any ancient intrusion into Rhodesia." This door is likely to remain open for many a long year to come.

                         Randall-McIver: Iron hoes and implements found at Great Zimbabwe

The Zimbabwe Pattern Ingot Mould not used, 915-1760 AD [**]

Expert Reports on Ancient Rock Mines.

In addition to the reports given as to the ancient character of the rock mines by Mr John Hays Hammond and Mr E. H. Garthwaite, the consulting mining engineers of the British South Africa Company, the following consulting mining engineers, mining engineers, and managers of mines, have also from time to time reported on the ancient character of the oldest type of rock mines:

Messrs. J. Anderson, T. Bayne, E. Bennington, B. H. Blaine, G. H. Borrow, Wallace Broad, A. Brown, J. A. Chalmers, A. L. Chambers, W. Crossley, Walter Currie, Telford Edwards, R. G. Elves, J. R. Farrell, Gordon S. D. Forbes, George Grey, Colonel R. Grey, Messrs. H. D. Griffiths, J. N. Griffiths, T. R. Harvey, Major M. Heaney, Colonel Heyman, Messrs. E. E. Homan, W. Howard, D. B. Huntley, A. E. Ingram, J. S. Jenkin, Frank Johnson, H. G. Jones, A. Little, H. MacAndrew, A. Mackinnon, J. G. McDonald, J. S. Park, C. E. Parson, Geo. Pauling, H. Pearson, H. A. Piper, H. Power, H. A. Pringle, F. A. Purdon, A. Reid, C. T. Roberts, C. J. Robinson, Ross, Sheffield & Co., J. W. Salthouse, Dr. Hans Sauer, Messrs. W. Teague, L. E. Tylor, E. Clement Wallace, W. T. E. Wallace, H. H. Webb, Lewis Webb, Franklin White, R. Williams, J. A. Woodburn, P. B. S. Wrey, and very many others.

The opinions of mining experts, and their references to the ancient rock mines, are to be found —

(1) In the reports of mining engineers filed with the prospectuses and memoranda of association on the registration of the various mining companies of Rhodesia, several of which also give plans and sections of ancient workings, and also a quantity of interesting information regarding them.[58]

(2) In the official reports issued annually to the shareholders of the various companies. The earliest reports give the best information. The mines having been located on ancient workings, and the ancient stopes having been sunk through, the later reports consequently are not much concerned with the ancient workings on the higher levels.

(3) In the particulars and plans filed at the Mines Department on registration of claims, and in reports and diagrams filed on entering footage for Inspection purposes.

(4) In three volumes of Mining in Rhodesia for 1899, 1900, and 1902, issued by the British South Africa Company.

(5) In the files of Rhodesian papers of 1896- 1902, which contain several series of reports furnished by mining engineers as to the character of the ancient workings on their respective properties.

Rock Mines Conclusions

It must not be forgotten that the mining engineers who have reported on these rock mines have long resided in the country and being highly qualified experts are also perfectly conversant with the old native methods of nibbling at outcrops of iron and copper ores. Yet they draw a very wide distinction between the methods of the ancients and those of old native people. Their unanimity in these conclusions is not at all surprising, for no one could possibly mistake the character of the two classes of mining or fail to notice the patently skilful work of the old gold-seekers.

From the earliest pioneer days of Rhodesia down to the present time there has always been a consensus of opinion on the part of experienced mining authorities that the oldest of the rock mines showed the highest form of skill in mining ; that the culture in rock-mining was directly introduced by Asiatics into the country in its already perfected form ; that after some centuries of display over a vast area, and to an extent that is almost beyond human conception, the culture degraded very markedly until it was represented in the tenth to the seventeenth centuries only in the river sand washing operations as described in the early Portuguese records ; 'that there was no" natural evolution of mining" on the part of the unaided Bantu and that the oldest rock mines were not the work of any present known Bantu people.

The only conclusion possible is that the oldest rock mines date from some prehistoric times, that is, from before 915 A.D., at which time the main rock-mining operations appear to have ceased. This conclusion is warranted mainly by the condition of the rock mines themselves, also by the silence of the Chronicles of Kilwa, and of the Arab and Persian writers of the tenth, eleventh, and twelfth centuries as to mining in rock, by the descriptions of the Bantu given by those writers, by the positive statements in the Portuguese records of the sixteenth and seventeenth centuries to the effect that the gold was obtained by washing river sand and not by working deep rock mines, and finally by the conclusions of the recognised ethnological authorities on the Bantu that no known Bantu were responsible for the older type of rock mines.

Note. — Since this chapter was written, Mr. H. A. Piper, the Consulting Engineer of the Goldfields of South Africa Consolidated, Ltd., has reported that an examination of the prehistoric workings on the Globe and Phoenix property proves that gold to the value of fully one million pounds sterling has been extracted by the prehistoric miners.

References

J.T. Bent. The Ruined Cities of Mashonaland: Being a Record of Excavation and Exploration in 1891. Longmans, Green and Co, London 1892

F.R. Burnham. Scouting on Two Continents. Doubleday, Page & Co, 1926

R.N. Hall and W.G. Neal. The Ancient Ruins of Rhodesia (Monomotapae imperium) Methuen and Co, London, 1904. Internet-Archive https://archive.org/details/ancientruinsrho01nealgoog

R.N. Hall. Great Zimbabwe, Mashonaland, Rhodesia: An Account of Two Years' Examination Work in 1902-4 on Behalf of the Government of Rhodesia

R.N. Hall. Prehistoric Rhodesia; Prehistoric Rhodesia: an examination of the historical, ethnological and archaeological evidences as to the origin and age of the rock mines and stone buildings, with a gazetteer of

mediaeval south-east Africa, 915 A.D. to 1760 A.D., and the countries of the Monomotapa, Manica, Sabia, Quiteve, Sofala and Mozambique. Fisher Unwin, London 1909. https://archive.org/details/prehistoricrhode00hall/page/486/mode/2up?view=theater

D. Randall-MacIver. Mediaeval Rhodesia. Macmillan and Co, London 1906

R. Summers. Zimbabwe, A Rhodesian Mystery. Thomas Nelson and Sons (Africa) (Pty) Ltd, 1965

F. White. Notes on the Great Zimbabwe Elliptical Ruin. The Journal of the Anthropological Institute of Great Britain and Ireland. Vol 35 (Jan-June 1905) P39-47

Biography of Richard Nicklin Hall (1853 – 1914)[59]

Hall was a solicitor and amateur archaeologist who moved to Rhodesia and led a very active life in Bulawayo. In 1896 he published a novel, Life among the Matabele Kopjes. From 1897 he was secretary of the Rhodesian Landowners and Farmers Association and also the Bulawayo Chamber of Commerce. He became editor of the Matabele Times and was correspondent for several London newspapers. In 1899 he served as Commissioner for Rhodesia at the Greater Britain Exhibition in London and in 1901 at the Glasgow Exhibition. Fellow of the Royal Geographical Society

Hall used the records of the Rhodesia Ancient Ruins Ltd provided by W.G. Neal and they co-wrote:

Architecture and Construction of Ancient Ruins in Rhodesia, a paper presented to the Rhodesia Scientific Association in 1901 and published in the Association’s Proceedings (Vol 2, P5-28) and then The Ancient Ruins of Rhodesia (Monomotapae imperium) published in 1902.

In May 1902 he was appointed curator of Great Zimbabwe site by the British South Africa Company. The S2A3 Biographical Database of South African Science states, “though his instructions were to preserve, rather than to investigate the site, he conducted an intensive study of the ruins. During his excavations he removed extensive stratified archaeological deposits which, representing occupation by Africans, he considered of little importance. His destructive digging was probably partly motivated by the unfounded belief that the buildings were erected by foreigners in pre-Bantu times but also reflected his poor understanding of the purpose of archaeology and how to interpret its evidence. Though his work was acclaimed locally at the time, his treatment of the site led to the ending of his contract in May 1904.”

In 1910 he was appointed editor of the Rhodesia Journal

His various other publications included:

A paper on "Great Zimbabwe" (Report, 1903, pp. 504-515) to the South African Association for the Advancement of Science

Great Zimbabwe, Mashonaland, Rhodesia: An account of two years' examination work in 1902-4 on behalf of the government of Rhodesia (London, 1905)

Visitors guide to the Great Zimbabwe ruins, Mashonaland, Rhodesia, South Africa (1907)

Prehistoric Gold Mines of Rhodesia in African Monthly (1907)

Prehistoric Rhodesia; an examination of the historical, ethnological and archaeological evidences as to the origin and age of the rock mines and stone buildings, with a gazetteer of mediaeval south -east Africa, 915 A.D. to 1760 A.D., and the countries of the Monomotapa, Manica, Sabia, Quiteve, Sofala, and Mozambique (1909)

A paper on "The present position of the discussion as to the origin of the Zimbabwe culture" (Report, 1911, pp. 325-337) to the South African Association for the Advancement of Science

"Antiquity of the Bushman occupations of Rhodesia" (1912) in the Proceedings of the Rhodesian Scientific Association.