AI{CIENT MII{ING
IN
RHODESIA
and adjacent areas
by
ROGER SUMMERS
Senior Keeper of Antiquities,
National Museums of Rhodesia
MUSEUM MEMOIR No 3
THE TRUSTEES OF THE NATIONAL MUSEUMS OF RHODESIA
SALISBURY
1969
CHAPTER 16
SYNTHESIS AND SUMMARY
Commencement o{ Iron Age
The earliest dated iron slag in Rhodesia belongs to the second century A.D. (Robinson,
l96le),1 but iron working may be older than this and in round figures one can date the start
of the Rhodesian Iron Age as about 2,000 years ago.
Iron was so often smelted in small-scale operations-a type of village industry-that one
hardly regards it as a subject for trade although it would have been necessary to supply it from
the interior to the Mozambique sea-plain where deposits of iron-ore are scarce. From time to
time it may have been exported but, for reasons given, it does not seem to have been an im-
portant item in the export trade (p.212).
Iron deposits are often small but they are widespread and of frequent occurrence hence lt
a
-J
very many tiny iron-ore quarries are bound to be overlooked and we can now know only the I
f,
s
largest and most recent. It is this difficulty that has lead to the exclusion of iron-ore mines st
from this study (pp. 2ll-212).
Beginnings of copper mining and trading
Smelted iron, manufactured to form tools and weapons, does not last long in the acid
Rhodesian soil and consequently iron slag is the only indication we have for the manufacture
of that metal. Fortunately, however, copper is not so easily affected by acid soils and many
Iron Age contexts contain only copper or its alloy bronze. This is no reason for postulating
the separate existence of a Bronze or Copper Age in south-central Africa for indeed the
earliest dated copper appears with the earliest dated iron in about e.n. 200 (Robinson, l96le).
Copper-ore is not so widely distributed as iron, hence one can visualise the possibliity of
a trade in this useful and ornamental metal and evidence has been adduced to suggest a local
trans-Zambezi trade in copper in the seventh century A.D. (p. 212). Since, however, we find
copper in deposits nearly 500 years earlier, we may assume that Ingombe Ilede was not the
earliest copper trading post and it seems reasonable to suggest that copper mining started on
a small scale at least as early as l.t.
200 and internal copper trading somewhat later.
Copper-ore seems to have been first extracted by means of large open stopes (p. 3l) and,
from its proximity to Ingombe llede, it has been suggested that copper mines in the Piriwiri
area (or Urungwe Mineral Belt) were some of the earliest in the region.
I The C-14 date for this site-Mabveni-is (SR-43) ,ar.s. 180*120. Another date frorn same site is (SR-79) 5701 ll0. The earlier
d:te is preferred but
the mean of the trvo is 380:: 120.
2ts
ANCIENT WORKINGS
The beginnings of the gold tade: Collecting
It has been suggested that the collection of gold from eluvial deposits on the edges of reefs
and its subsequent trading started about e.n. 600 (p. 119), but hotv and w/ry it started are
less simple of determination.
Perhaps the r'fty ? is easier to answer. As long ago as the first century A.D., the government
of the Roman Empire (theoretically the Emperor himself) became concerned at the drain of
bullion from the Roman Empire to India: Nero (e.n. 54-68) promulgated an edict for the
purpose ofarresting the flow ofgold to the "sink" ofthe Indies, but this, like other exchange
control legislation, was not too successful and gold continued to travel eastward and to remain
there (Wheeler, 1955).
Indians have traditionally kept their accumulated wealth in gold and their own country D
could not maintain the supply for ever. It has been pointed out that Southern Indian mining
may have provided the inspiration for Rhodesian ancient workings (p. I 19 ff.) and that the In
intense period of Indian mining lasted for sorne 500 years up to the third century A.D., there- rh
after falling off until it ceased altogether during the troubled period of the Islamic invasions Icr
(p. 119). Undoubtedly, therefore, Indian prospectors were looking far and wide for new sources th
of gold from the third century onwards and ultimately they came to hear of the occurrence ot
of gold in the interior of south-east Africa. The story of the chances and mischances of the Pr
Indians' hearing of "Rhodesian" gold must be a fascinating one but, as there are no clues to
the details, we must avoid embarking on a fictitious account here. Suffice to say that somehow- i:\
-r
possibly through the stories of Arab sailors-the story of gold in Africa penetrated to India
and that by some means this African gold started to flow eastwards about,c.o. 600.
It is out of place here to point out that the Piriwiri copper deposits, which-for quite oP
different reasons-we think to have been among the earliest "Rhodesian" deposits to have s!l
been exploited, contained an appreciable quantity ofgold which had been carried into the pre- st(
Cambrian Piriwiri rocks with the fluxes of copper sulphides (Chalcolite and Chalcopyrite) SU
which themselves, when oxidised, gave rise to easily smeltable ores of copper. Indian metal- na
lurgists were acquainted with gold occurrences in similar conditions, for the geology of gold sII
in Mysore was identical with that in Rhodesia (p. lt6) but were obviously happier when sp
exploiting quartz reef. en
Gold prospecting and mining \,
.1.
rL
A hypothetical scheme for prospecting for gold has been given (p. 115.). It cannot be
suggested that there was anything planned about this scheme, which must have been purely Er
empirical and which must have relied on that "eye for country" and for the fine details of the m:
landscape referred to at the end of Chapter 2 (p. l7). We do not even know if such a programme ui
is chronologically sound: it fits the dates we possess at present but may have to be amended
when more dates come along. Unfortunately, we can never get the really vital dates of com- tLr
mencement of work, as so much direct evidence has been destroyed, not only by modern alt
mining but by the "ancients" themselves who swept away their earlier work in their own efforts le<
to exploit the mines. at'
Fre
Usually our dates will have to depend on correlations with archaeological industries and
the best degree of dating we can hope to get for the beginning of mining is what Dr. Kenneth
of
Oakley has called ,A3 dating, although we can reach the highest degree (Al dating) from
ut
charcoal and skeletons which marked the cessation of work in some particular mine.
tr€
Under present conditions the following programme is suggested (pp. 158-60): thi
Umniati Basin and Sinoia area. Date of commencement probably seventh century A.D. but it
was still producing (probably only one mine-now Eldorado) in 1882. Portuguese were trading thr
2t6
SYN-fHESlS AND SUMMARY
tl) had a post near Gatoonra (Note
that trre portuguese
il!ffi'.f,:f ,Tf:#tl#ii:t$?:3.1jso
Mazoe Basin' Exploitation ton],r,.n..d'about eighth century or earlier
teenth century but fell_offduring lasr and continued to eig'-
'n t
rrvo or
utntali' Much abo,t the same ai Mazoe-but
thle frunJ..iviu^
fell off.o,rria.iuu"iy after the fifteenth
of sqbi-Lundi Basin' Exploitation co-n.,e*ed certury.
auout tne-ia; ffi. as Mazoe Basin,
larger and richer in ririnerals .ontin,,.Jlo but bei'g
flourish ,.".rri.enth.or
: Around Bulava't'o' Probabll' the last ur.u
to commence"iuiigold production.eighteenth century.
prospecting does
rln seem to rrave bee' thoror"r-srr. probabre
dates eleventn_.ignl5ntr, centurv. 'ot
Developrnent q/ gold ntining
ao
Altho'gh it is'ery probable that the ori-sinal sri'rulus
he Indian prospectors from Mysor" canre fro'r I'dia and that i'rmigrant
the erpioit"iio,r^oi .,nhodesia',.
the development seems to have been"o-t".n..d-
in the hands of Atiican groups. This is sho*.n -eold-fields,
t\
local pottery whic.h comes from ancient
workings and especialil.frorn sites clustereo
bit'e purelv
as
the tops of the ntines' The various *u..r,
known as Zi*a and Leoparo's Kopje are
rou,ri
le other Iron Age wares some of which are a1 aki' to
unconnected *itJr nrining and others arL.
IC predate the earliest mines by many kno$n ro
centuries.
o It is possible' and indied most likely, that the original prospectors
specialists but it is certain that by the seventeenth. and were
2 cent'ry 'riners
too b..on.,. a no'-speciarised.
or at best semi-speciarised, occupation, u'dertaken
This brings out one-extremily interesting poi't.
foia'ining
pri.t?firr. year only (p. 112).
e open stopes of the classic Indian paltetn
rirJ-.lrI.r, mrnes seerl to have bee^
and sonie of thern see,n io nuu" contained
e systems (p. 166), also.apparently Indian (p. u.nrituting
steeply-dipping reefs' There are viry
il9), were admirably suited to
"ff "i",iri.fr^
occasional reference, in o.i"riptions of dr.i'r,s which were
, supported by timbering' In Rhodeiia,
however, a different rnetl-rod was followed
namelv the digging of large numbers'of for flat reefs,
i small area at the bottom of the shaft,
indiviiuat rilar, ;;;'t'lie- exptoitation oia
the shafts u,-,a uotiour'"rrn,nu.., were
rJuiiu.ry
I later refilred u,it'
:#i!::'#['l,1tli,;'Hi'"t;J:i#t[the n.ed ror tinJe'.i,if.iu,g. u,'c..g,.o,na-,iopl, out
This rnethod \\'as apparently not used in
lndia but uas a standard in shallorv
Neolithic flint rnines in *lestern ir."p;-;i
about 3500 ,... 'rethod
the desired nriner.al la' irr
almost horjzontal strata: it t o, tup.ir.al ",r*r.
Europe Bronze Age copper ntines ifter
ur the tirrrbered
u'ralrg.ornd stopes in u,estern
about 2000 s.c. lt uu.,n6 t1e simplest
man faced with the problem o_f mining response of
a rejatir
"' el, llat reef t,u, ;r rr"l rtor tlteresponse of
uith a long rnining tradition behind [in,.- a nran
So both zrrchaeology and rninin-e technolog-r poi'r
the discovery' of the mines being inirr. to trre deielop'ent. as opposed to
ttunar oi..r*iir.r1,'unriuln.o people. Ne'ertheless.
although they were half-spe-cialisJa-l.i"g
Iedge of the complexities of
p.i'aril' agricultural people-and had no knoq.
ljnine, ti;;-";r.
able to get down io such .ontiJ.rufii.-ffi,;;
nor *ithout skill and courage, for
they u,ere
that usuaul. tii.r' 111;", stop untir
and that proved arr absol'te barrier they reached
to nri before the invention
ffiliill'rvater-level ^ln"r,
Despite these drawbacks of lack
of training. showrr most clearly i' the very few occasions
when thev overcanre probrenrs
i""i;;;';"a::pi".i-,r"g ;;;;.ft', rnaintained trreir mining
"f
ffi','jiffi "spite
changes in
-eovern'r.,rtinJ a.Jpit" ,,ii!,"ni, t;;;" portuguese to take
over
Skeletal remai's shoiv that the nriners. ofte' *,o'en.
the present Bantu inhabilanrs of ttl. r..,0io,.,. were ol the sarrre racial gro'p as
2t7
ANCIENT WORKINGS
In all some 4,000 mines are believed to have been opened (p. 107) and it has been calculated
that over a period of about 1,200 years they produced between 20,000,000 and 25,000,000 oz.
with a normalproduction during their heyday ofabout 20,000 oz. ayear, a very good produc- Firs
tion for miners working without any mechanical aids or explosives (p. 188).
End
Dur
Control of the gold trade Abc
During its first few centuries, external trade with "Rhodesia" was in the hands of a very Fro:
few people, possibly indians and almost certainly Asians. ?it
Later when gold collecting and gold mining came to pass, the group controlling the trade (l-
became rather larger but still it cannot have been very big and it has been suggested that it
took steps to maintain its monopoly by subterfuge (p. 256 ff.). Internally, it was probably
helped by the apparently small social units that existed: for it is highly probable that there Du:
were no "tribes" as understood today and possibly a single village or a small group of villages
constituted the political unit. If such a multiplicity of units took a long time to deal with, they
were nevertheless easy to cajole into selling gold for a low price.
During the twelfth or thirteenth century, however, there seems to have arisen an aristo- Dur
cratic group which before long took charge of the political organisation of the area, enhancing
its prestige by building stone "palaces". This group, or a successor of somewhat similar type 8th-
and aspirations, not unnaturally took control of the region's natural resources and erected Dur
stone buildings to watch over the gold mines (pp. 137-a0). Du:
.rn
The Portuguese testify to restrictions placed on gold mining (p. A7) while the increasing
number and variety of imports found in the archaeological record after the fourteenth century
(p. 125) confirms that the price of gold was rising just at a time when the quantity was falling I -1rh
(p. les). I lth
We may conclude, therefore, that the kings known as Mwenemutapa were in full control I 3th
of the economy of their realm and were successful in enforcing both a restriction in the output ?D
and a price control which not only enabled them to have sumptuous burials (pp. 192-4) but I 50:
also beggared the Portuguese colonists and ruined the Arab traders on the east coast (pp. 196- I 57(
7). B:'1
Dur
End of "ancient" gold mining
Abc
It would seem that by the seventeenth century many of the mines were becoming worked I
-i:-
out, at any rate so far as "ancient'. payability was concerned: that is to say that tenor was
falling below about 16 to 20 dwt. per ton. Very little seems to have been exported after 1700
and it may be that many mines were actually closed down before being worked out: this is one l Et:
explanation for the lack of exploitation around Lonely Mine. Abr
However, the real end of gold mining came with the overthrow of the successors to the D::
Mwenemutapas-the Rosvi Mambos of the Changamire dynasty-in about 1825-35 by a
savage Nguni group under Chief Zwangendaba which destroyed political power and social
cohesion. A following Nguni group, the Ndebele under Chief Mzilikazi Kumalo, maintained Co,:.
the existing state of uncertainty and in fact depopulated the richest gold mining area of the
region. Te:::
A few mines in Shona territory may have continued production but only one is known, the noti
Eldorado near Sinoia which was being worked by Chief Maghonda in 1882 but which ceased the I
production shortly after when the chief was killed by orders of the Ndebele King Lobengula.
After an interval of a very few years the mines began working again under British super- colle
vision in 1894 but, crude as were the methods of the Pioneers, we can no longer refer to these emp
mines as "ancient workings". Afri,
218
SYNTHESIS AND SUMMARY
Test,n I3.-CHRONOLOGICAL SUMMARY
First century A'D.: Beginning of Iron
{ge on zambezi. spreading throughout Rho-
End.2nd century: 3:Ti"?#,1.fl3#fi:T..","1"
f.urine-{tf century: coid
_. Beginning
productio"n in India iuifing of.
About 600, possibly earlier: of gord collecting rn-ft.rroieriu.
prospecting f;r gold by l"jiuor. -^---"
T'loT 7th century:
? 7th century
(between 650 and 950): Ingombe Ilede founded but does not survive for
more than about
150 years. Trans-Zambezi ftade (? to Kilwa) fro_ U_niutV
During 8th centurv: H,t{i!:1il'Xl;" in production (incrudes Darwin, Makaha
and Salisbury).
Sofala founded (not on present site).
Manica gold mines producing.
During 8th-9th centuries : Marumbene a market and port on Sabi River (? becomes
un_
usable by l4th century).
8th-9th centuries: Prospecting up Sabi and Lundi rivers.
During 9th century: Hartley gold mines started, Belingwe and Filabusi mines
During I lth century: opened.
Mapungubwe (K2) founded.
? During l2th century: sofala founded on present site, Kilwa takes over from Mogadishu
as principal of Sofala (this may have occurred earlier).
l3th century: Disaster at Aboyne Mine (Fort Rixon).
llth to 16th century: Zimbabwe flourishing as local capital.
l3th to 17th century: Mapungubwe flourishing as important regional centre.
? During l5th century: Mazoe/Ruenya production begins to decli-ne.
1505 Portuguese capture Sofala from Arabs.
1570-72: Portuguese expeditions up Zambezi successful.
By 1600: Portuguese markets and settlements throughout northern
MaShonaland.
During 17th century: Some Manica and Mazoe mines worked out.
About 1690: Portuguese driven from interior, except for Tete and
lTth-l8th centuries: Sena.
Some north Matabeleland mines closed although
not worked out,
Belingwe and Midlands mines srill workins.
Khami flourishing.
,:
1825-35: Ngorri_invasion (Zwangeadaba) puts an end
About 1840:
to most mining.
I
Zirnbabwe sacked by Ndebele.
During 1880's: Last.Mashona mine (Eldorado, Sinoia) still operated by
I Chief
Maghondi but ceases production when he rvas kiued uv
I Conclusion
ua.Gi..
The story revealed by this study contains no great
names: the eueen of Sheba, Solomon,s
Temple and Hiram, King, of Tyre,-hnd no place-f,ere
_and
nothing less than an analysis or irre truttr which underlies
ophirlrlr.u., mentioned. yet it is
the Sailor.
irr. rorourrrl stories of Sindbad
The Persian and Arab storytellers, who have passed
on stories which can still enthral us,
collected their matter
9n -the quayside in Basra-and_other ports in the persian curr. Trr"y
emphasised-not only the immense profits but also the tremendous risks
taken in the East
African trade' Shipwrecks make good stories for those wtto tnow nothing
or ttrem aJ nair-
2t9
ANCIENT WORKINGS
an
breadth escapes are themselves an escape from humdrum lives, yet Sindbad's stories have
ui, oi-u"tuuliity which is missing from Sir John Mandeville's and Baron Munchhausen's;
Sioabaa's are 6asically true while the others are for the greater part utterly false'
The storytellers' audiences identified themselves with sindbad and so were not
interested
in the people"from whose efforts he made his profits but we can perhaps appreciate their work
almost universal application in undeveloped areas: the introduc-
-oi" Jutiiy. The tale is one oftechnology, the ruihless exploitation of natural resources for an
tion of a relatively advanced
ouitia.r't profit, ihe deterioration in t""l*iqu.t and the unsophisticated solution of complex
i""it"i.uf problems, the intervention of a rivai group of explorers, the bidding for local favours'
All this has been repeated often enough but here the story changes: the appearance of local
;;;6; wittr a "upu"ity for and organisation and an ability to raise the price of the desire{ com-
il;Jity by hard bargaining restrictive practices. These are quite new variations on the too
tu-ifiit iheme and-for sJveral centuries they persist successfully but alas they do not last for
ever and we are back on the old story: invaiions by neighbouring tribes,
barbarous atrocities
and a complete collapse of both technology and economics'
gold
In the 75 years or so since modern production started (1894) about 38,000,000 oz. of
trave been won, as against the 22,000,00b oz. (more.or-less) during
the 1,200 years of 'ancient"
production: this suggesis that modern mining techniques-are nearly 30 times as efficient
as
.,ancient" ones but this improvement has come aboui only in the mechanics of mining and
ttre-ctremistry of .."ou"ty, we still know ofvery few more mines
than the "ancients" did hence
o.r. .*plotuiion methodi are little, if any, better than theirs'
It is clear that the 'iancients" made very-combined full use of their opportunities but that this was
not particularly profitable to them until they it with extremely sophisticatedmethods
of economic iontrot. It is indeed remarkable how much the Monomotapas and Mambos
l" been wasting asset and a declining economy'
*"""g.J get out of what appears to.us,to have a
to extract- sufficient
il;i;r; "eituries after production had passed its power u"9 managedpeak, they.
;r"frt};;- the gold trade to maintain themselves in .$ the country with the help
li u gradid "civil service". Such advantages are quoted with envy by recent_wrjters on
"i..f' history of pre-Colonial Africa, yeithe sud{en and utter
the econom'ic
collapse of the Mambo
,-pii.-i" if,. iu". 6t aggression by the unorganised Nguni hordes shows that satisfactory
economics alone were not enough.
220
