26

Innocent Pikirayi on the Zimbabwe Plateau and adjacent regions

constructed in the same architectural tradition as

Great Zimbabwe in Historical Great Zimbabwe (Beach 1980; Mudenge 1988).

Referred to as madzimbabwe or Zimbabwe , liter-
Archaeology: Reconceptualizing ally meaning houses of stone, they were capitals
Decline, Abandonment, and of local and regional political dynasties that dom-
inated the history of the Zimbabwe Plateau until
Reoccupation of an Ancient Polity, the end of the 19th century (Figure 1). Although
A.D. 1450-1900 16th-century Portuguese references describe Great
Zimbabwe as a "fortress" under the control of a
ABSTRACT prince, the Portuguese do not seem to have made
a direct expedition to it (Pikirayi 2006). The first
knownand
A recent anthropological critique of the archaeology European to have visited Great Zimbabwe
cultural heritage management of Great Zimbabwewas Karl
refers to Mauch, who in 1871 publicized the
"the silence of unheard voices and untold stories"stone
"the un-
structures to the Western world (Burke
represented pasts of local communities," and "the silence of
1969), resulting two decades later in antiquarian
anger - the alienation - and desecration of Great Zimbabwe"
investigations
(Fontein 2006). Fontein sees a lack of representation of local (Bent 1893; Hall and Neal 1902;
Hall
histories, not only in the literature, but also in museum 1905).
displays
The origins of the culture that
and in the archaeological narratives (Pikirayi 2001), including dominated Great
heritage management reports (Ndoro 2005). Admittedly,Zimbabwe
this is probably lie some 300 km to the
one of the reasons why Great Zimbabwe is a contested site
south, in the middle Limpopo Valley, following
and cultural landscape. In this paper it is argued that Great
Zimbabwe's contribution to the understanding of the the demise of the state based at Mapungubwe
origins
(1220-1280).
of later Karanga and other regional histories is poorly under- Mapungubwe, whose wealth was
enhanced
stood. Archaeology, in collaboration with other disciplines, can by trade in gold, ivory, animal skins,
play a useful role in writing the story of Great cloth,
Zimbabweand glass beads with the Swahili on
and - in keeping with the plenary session themes - relating
the Indian Ocean coast, declined following the
it to other transformative global developments of the early
modern era, when the site was clearly experiencing abandonment
decline and of the region either due to climate
change
eventual abandonment. Detailed local histories, though or transformations associated with the gold
useful
trade (Pikirayi
in understanding sociopolitical dynamics on the Zimbabwe 2001; Huffman 2007). Iron Age
Plateau, may account for the invisibility of Great Zimbabwe
farmersthen developed chiefdom-level societies
since A.D. 1550 until its "discovery" by Europeans during
at Chivowa and Gumanye hills in south-central
the late 19th century. Underlying these processes is the fail-
Zimbabwe
ure by archaeologists to understand decline or collapse of a (Sinclair 1987). They transformed from
sociopolitical system once based at Great Zimbabwe,simple
and kin-warranted
its domestic corporations, rely-
global implications. ing mainly on land and cattle, to long-distance
traders. With this newly acquired wealth, they
Introduction financed the building of stone structures. By
about 1270, a powerful elite emerged at Great
Great Zimbabwe is a complex of stone-walled Zimbabwe, laying the foundations of an elaborate
urban complex and the center of a state. Great
structures located in south-central Zimbabwe (20°
16' 23" S, 30° 56' 04" E). Dating from the early
Zimbabwe reached its peak during the 14th and
second millennium A.D., these structures cover15th centuries when the erection of elaborate
an area of more than 700 ha. The structures
stone structures - evidently symbolizing prestige
comprise massively built stone monumentaland
archi-
status - was extended towards outlying areas.
tecture, and the site has been interpreted as the
During its fluorescence, displaying elite residences,
center of an ancient kingdom or state (Garlake ritual centers, public forums, markets, houses of
commoners and artisans, and with a population
1973; Pikirayi 2001). This interpretation emanates
from the fact that there are more than 300 sites estimated at 18,000 inhabitants, it became the

Historical Archaeology, 2013, 47(l):26-37.

Permission to reprint required.
INNOCENT PIKIRAYI - Great Zimbabwe in Historical Archaeology 27

FIGURE 1. Map of the Zimbabwe Plateau showing the location of Great Zimbabwe in
(Map courtesy of Shadreck Chirikure, 2012.)

largest metropolis in southern societies,

Africa (Figure
either 2).
through contact and inte
From about 1300, stone buildings of a scale
and/or through and textual sources, both
magnitude unparalleled on the written.
entire Zimbabwe
The texts are mainly external in
Plateau were constructed. Presiding
andover a polity
usually indirect in regard to the soci
which they
located mostly in south-central regions of the are referring. Great Zimba
Zim-
babwe Plateau, Great Zimbabwebeen
became the most in such a way in Mu
documented
Portuguese
dominant political authority south sources, concentrating mor
of the Zambezi
for up to 250 years. Solomonic legend and the location of th
Dating from 1270 to 1550, Great
OphirZimbabwe
(Carroll 1988). The silence of the
has, in conventional archaeological terminology,
ments a
on the developments at Great Z
prehistory, a protohistory, as well as an historical
in the 15th century presents difficulties i
period. These terms are not without their
standing own
its decline and subsequent aband
problems (Schmidt and Mrozowski [2013]). In
although it is apparent that this is con
by economic
this paper, I use the term protohistory and political changes tha
to demar-
the character
cate a period between the conventional of Indian Ocean commerce in the
boundaries
of prehistory and history, during
early modernaworld.
which culture Archaeology must therefore
such as Great Zimbabwe is knownreconstruct
through the associated
other events and tie them with
28 HISTORICAL ARCHAEOLOGY 47(1 )

FIGURE 2. A view of the Great Enclosure and som

permission from National Museums and Monume

revolving around
the meager information the gold trade. In
available order to
from the
sources. However, even when
understand these both
processes, oral and
Great Zimbabwe
ten sources become available after A.D. 1500, must be situated within the framework of aban-
Great Zimbabwe remains peripheral to main- donment studies in global archaeology. This is
stream developments on the Zimbabwe Plateau the subject of the next section.
and the western Indian Ocean zone. Clearly it
must have been abandoned then, but not for- Abandonment in Archaeological Contexts
gotten, given its size as a former capital and
its regional political and economic influence. Studies of abandonment remain extremely
From a scholarly perspective, Great Zimbabwe's popular in archaeology and the social sciences,
"silence" (Fontein 2006) comes, firstly, from the the objective being to understand why certain
interpretation of the site's chronology, domi- complex societies eventually succumbed to
nated by radiocarbon dating, which presents it failure and came to their fateful end (Tainter
essentially as a product of prehistory; secondly, 1988; Diamond 2005). However, most of these
the failure to understand post- 1500 Portuguese studies situate the fates of these societies largely
accounts, despite some of these accounts making in environmental terms and negate the global
references to the site; and thirdly, the meaning dynamics of which they had become an inte-
of the spread of the Zimbabwe tradition else- gral part. The way archaeologists use the term
where on the Zimbabwe Plateau, which, to a "abandonment" is problematic as it obliterates
large extent, was largely by global developments a range of human behaviors, often reducing
INNOCENT PIKIRAYI - Great Zimbabwe in Historical Archaeology 29

Great Zimbabwe.
complex processes to simple, localized The evidence for this comes
events
(Cameron 1993:4). From a legal perspective,
from a careful reading of archival literature on
abandonment implies that people give
Great up their
Zimbabwe, including antiquarian literature
claims to and interest in a place
(Bentwhen
1893). they
move away (Colwell-Chanthaphonh and of
In terms Fergu-
cultural formation processes, aban-
son 2006:37). While communities did abandon
donment affects the quality of cultural remains
sites on the Zimbabwe Plateauentering
and adjacent
the archaeological record. In this context,
it is precolonial
regions in southern Africa during considered as a stage and key process in
times, human/land relationshipstheimplied
formationthat
of an archaeological site (Schiffer
they did not give up their claims
1995). to places to Cameron (1993): "[Aban-
According
they had settled originally. Thisdonment
contrasts, for
processes condition the entry of cultural
example, with the U.S. Southwest, where
material a lot
into the archaeological record." When
of abandonment studies have been conducted, think of the abandonment of
we archaeologists
Great "abandon-
and where archaeologists use the term Zimbabwe, do we imagine catastrophe,
mass depopulation,
ment" to imply that the sites concerned were no or a regional exodus of
longer occupied. Cameron (1993)people to elsewhere as implied in oral tradi-
demonstrated
that abandonment processes aretionsmuch more1962)? What was the cause for
(Abraham
complex and should not be treated as single
the abandonment? Is there any material culture
events. According to Colwell-Chanthaphonh and abandonment? How do we
patterning to suggest
Ferguson (2006:57): "[Cļontinued read
employment
"abandonment" ofin the stratigraphy? Did we
a single term to define a state of cultural
interpret theseand
remains with the assistance of
social practices is unconstructive because
relevant casual
historical ethnographies on abandonment
readers are unlikely to understandand
'abandonment'
human/land relationships?
Abandonment
beyond its colloquial meanings." Some takes place or occurs at two
sites may
have been abandoned but in fact continue to be levels, the local scale and the regional scale.
used in the present - in a spiritual sense, such Settlements
as can be part of a regional system.
the case of Great Zimbabwe (Ndoro 2005; Fon- Abandonment of sites also occurs in a regional
system as Binford's (1978a, 1978b) study of
tein 2006). One of the reasons scholars see Great
Zimbabwe this way is conditioned by the reading hunter gatherers indicates. This is often a gradual
process but can be rapid as well as catastrophic.
of behavioral archaeology (Schiffer 1995), which,
according to Colwell-Chanthaphonh and Ferguson Ritual may condition abandonment behavior,
(2006:38), structures elements as either a systemic
resulting in the deposition of unusual quantities
context or an archaeological context. This of
is acertain kinds of refuse. According to Nelson
dichotomy that leads researchers to see objects(2000:52-54),
or who also researched in the U.S.
artifacts as refuse and sites as abandoned. Southwest, archaeologists conceptualize aban-
Ordinarily, it is difficult to perceive a site donment
or differently when looking at foraging
a region as abandoned when the people once and state-level societies. Foraging societies are
living there show continuous connections with often referred to as employing a mobility strat-
such places. The problem is those archaeologists egy, while for state-level societies, this is often
who place too much focus on the "mystery referred of to as a collapse or decline. However,
leaving" and not enough attention on many ways Nelson argues that state societies also employed
that connections to homelands were maintained mobility strategies. Archaeologists use the term
(Colwell-Chanthaphonh and Ferguson 2006:38). abandonment when they are interested in the "the
leaving," such as the depopulation of regions
It is proposed here that the inhabitants of Great
Zimbabwe did not completely abandon the or sitediscontinued occupation of a site or struc-
during the 15th and 16th centuries as inferredture. Abandonment together with migration are
from the reading of the archaeological evidence
processes, strategies, outcomes, and causes of
but, rather, maintained connections with the site
social change. People move as part of a strategy
and the associated landscape over centuries,toasaddress social issues. Abandonment is also a
17th-century movements to the south suggest continuous process of transformation, a movement
(Beach 1980). I consider this as a reconnection
that results in the absence of active residence - it
with their ancestral homeland and, thus, with is about moving residence (Nelson 2000:55).
 30 HISTORICAL ARCHAEOLOGY 47(1 )

Using concepts centuries. However,

framed around the authors were obviously
abandonme
is possible to limited by the loss
reconsider bothof muchthe
of the overburden
archae
and historical evidence from Great Zimbabwe removed through antiquarian activities and,
and discuss the site in the context of both local
therefore, did not discuss in detail the termi-
and regional histories on the Zimbabwe Plateau. nal periods of the settlement. In this section,
It is the dynamics of these histories that tell I show, with the help of Arabic sources, that
of Great Zimbabwe's declining global role and considerable information has been recovered
its eventual demise as a center of considerable about Great Zimbabwe's gold and copper trade
political and economic power. During the 14th during the 14th and 15th centuries.
One of the reasons for the "silence" of Great
and 15th centuries, Great Zimbabwe may be situ-
ated in a time period archaeologists may refer to
Zimbabwe even before the availability, of Portu-
as protohistory, as some of the events pertain-
guese written sources is the peripheral treatment
scholars have given the site in the context of
ing to the site can be gleaned from examining
the development of early modern urbanism, spe-
external written accounts and through extracting
historical information from datable ceramics and cifically the Swahili towns, along the East Afri-
glass beads that arrived at the site. In this per-
can coast. The problem lies in the interpretation
spective, the early-20th-century work of David of Arabic sources. In this respect, the works of
Randall-Maclver (1906) is significant, especiallyArabic writers, such as ullah Muhammad ibn
his "Medieval Period" dating of Great Zimbabwe Battuta (1304-1369), must be reconsidered and
(which should be contrasted with Richard Hall's put into a broader regional context. Ibn Bat-
Prehistoric Rhodesia [1909], primarily a futile tuta was the only medieval traveler known to
rejection of Randall-Maclver 's groundbreaking have visited the lands of every Muslim ruler of
work on the site). There is also need to considerhis time, including Mogadishu, Mombasa, and
seriously post- 15th-century migrations by the Kilwa. From these writings, as well as those
Karanga and other groups, which are reported of the much earlier Arab writer Al Masudi (ca.
from oral sources as a process of expansion from 888-957), "the Herodotus of the Arabs," who
south-central Zimbabwe northwards towards the also visited eastern Africa, a perspective can be
Zambezi basin (Beach 1980). However, during gained of the 14th century and even the much-
the 17th century, the Karanga are reported in earlier gold trade coming from the hinterland,
both oral traditions (Beach 1980) and Portuguese for which the Zimbabwe Plateau played a domi-
written accounts (Pikirayi 1993) to have migratednant role. Often poorly understood is the grow-
southwards, leaving much of northern Zimba- ing intercontinental demand for gold during the
bwe under the control of the Mutapa State but 14th century, at a time when Great Zimbabwe,
increasingly under Portuguese exploitation. Whileaccording to available radiocarbon dates, saw
the value of these traditions is in clearly show-
intensification and expansion of building activity
(Huffman and Vogel 1991; Pikirayi 2001, 2006).
ing a reversal of earlier expansion or migration,
and a subsequent reconnection with the landscape There was so much wealth around Great
of Great Zimbabwe, the events also serve to Zimbabwe, as well as in all the Swahili towns
underline how Great Zimbabwe lost its global existing at the time, including Kilwa, that it
significance in a period usually associated withled to these cities and towns erecting some of
the growth of globalization. the most remarkable monumental architecture
on the western Indian Ocean coast and in the
Great Zimbabwe in Protohistory adjacent hinterland (Sutton 1990). This wealth
emanated from the gold, demand for which
So far, current descriptions of the stratigraphy increased considerably not just in eastern Africa,
of Great Zimbabwe have failed to reconstruct but also in the Persian Gulf, India, the Far East,
the historical events associated with the site and Europe. According to Sutton (1990) it was
during the 14th and 15th centuries. The work Europeofthat exerted so much pressure to exploit
Huffman and Vogel (1991) redefined theallchro- known and workable sources of gold. This
nology of the site demonstrating, using available development was significant in that "gold had
radiocarbon dates, that the main settlement at
a disproportionate impact on the directions of
Great Zimbabwe did not last more than two commerce and its intensity" (Sutton 1990:65) as
INNOCENT PIKIRAYI- Great Zimbabwe in Historical Archaeology 31

the volatile world market became characterized in regard to wares traded within the western,
by unpredictable booms and sudden slumps.
northern, and eastern Indian Ocean zones, includ-
This affected the poorly developed sites of ing
the the adjacent hinterlands, between the 9th and
gold-trading network, with destructive impacts.
15th
It centuries. This route supplied vast commer-
will be seen in the next section the impact cialthe networks linking China with Southeast Asia,
demand for gold had on the mines around Great the Near East, and eastern Africa. According
Zimbabwe, according to Portuguese accounts. to Pirazzoli-ť Serstevens (1985:284), the marine
When he visited Kilwa in 1331, ibn Battuta network through which the wares were distrib-
was told about the gold coming from Sofala, "a uted was never completely broken up, but the
fortnight's sail away," which in turn came from restrictions imposed by the Chinese government
the hinterland of "Yufi," "a month's journey" at the end of the 14th century and the arrival of
away. According to Sutton (1990), this was the Portuguese in the Indian Ocean at the end of
certainly a reference to the Zimbabwe Plateau,the 15th century affected its organization glob-
although the identification was clearly confusedally. The main challenge for researchers working
with the West African Ife. Interestingly, a copperon Great Zimbabwe and other sites on the Zim-
coin recovered at Great Zimbabwe was inscribed babwe Plateau that have yielded some celadon
"al-Hasan [ibn] Sulaiman," the sultan of Kilwa sherds and, later, vast quantities of blue-on-white
from about 1320 to 1330 (Huffman 1972a, porcelain - see Pikirayi (1993) for the site of
1972b). It has been suggested that although the Baranda, in northern Zimbabwe - is interpreting
coin was minted at Kilwa, the source of the these economic processes in terms of importation
copper was the Zimbabwe Plateau or elsewhere,policies, constraints, and opportunities. What is
possibly northern Zambia or southern Congo. apparent is a transformative shift from the world
The trade in copper connected Great Zimbabwedominated by Asian powers that, for centuries,
with societies in south-central Africa, and one controlled much of the commerce in the Indian
of the sites that featured in the copper tradeOcean, the gulf, the Indian subcontinent, and
during the 15th century, if not slightly earlier,continental Asia, to the world dominated by
was Ingombe Ilede, on the middle Zambezi. Thethe European powers that arrived in the Indian
fluorescence of Ingombe Ilede coincides with theOcean in the late 15th century.
loosening of Great Zimbabwe's hold over the
gold trade to the Swahili coast and the rise ofPortuguese Written Sources, Karanga
the Zambezi as the most important trade routeMigrations, and Great Zimbabwe
in south-central and eastern Africa. From now on
the Indian Ocean trade was channeled through Portuguese written sources further attest to the
the northern regions. However, the merchants continued decline of Great Zimbabwe, as the
could not escape the expansionary tendencies of commercial shift toward the more auriferous gold
rulers, some of them once based at Great Zim- belts of northern Zimbabwe undermined the city
babwe. During the late 15th century, the Mutapa (Pikirayi 2006). There is a tendency by scholars
dynasties in northern Zimbabwe conquered the to misread, let alone misinterpret, Portuguese
copper mines of the northwestern plateau, upset-sources making references to Great Zimbabwe.
S. T. Caroli is mistaken when he mentions that
ting Ingombe Ilede's hold on the trade. At the
that there were eight direct references by the
same time, the Portuguese were arriving on the
Mozambican coast, disrupting commerce at thePortuguese to Great Zimbabwe (Caroli 1988),
lower end of the trade routes. This impacted when in fact those sources were pointing to
Great Zimbabwe, a major player in global andthe state capitals of the Mutapa Kingdom in
regional commerce. the northern Zimbabwe Plateau (Pikirayi 1993).
A further illumination of Great Zimbabwe The closest and most detailed description of
during the 14th and 15th centuries comesGreat fromZimbabwe comes from João de Barros,
imported ceramics, glassware, and glass beads published in his book Da Asia in 1552, being
from the Persian Gulf, India, and the Faran account of the activities of the Portuguese in
East,
some of which have been recovered on the site the Orient. This particular source corroborates
(Garlake 1968). Of importance is the "ceramic the observations of some Arabic writers during
the 14th century:
route" defined by Pirazzoli-t' Serstevens (1985)
 32 HISTORICAL ARCHAEOLOGY 47(1)

There are other mines in a district called Toroa, clearly attest to their knowledge of the existence
which by another name is known as the kingdom of
of Great Zimbabwe. More importantly, João de
Butua, which is ruled by a prince called Burrom, a
vassal of Benomotapa, which land adjoins the afore-
Barros's account - collated in 1538 and possibly
said consisting of vast plains, and these mines are the making references to events three decades old -
most ancient known in the country, and they are all does not say Great Zimbabwe was completely
in the plain, in the midst of which there is a square abandoned. Some members of the royal family
fortress of masonry within and without, built of stones
were still living there despite the abandonment
of marvellous size, and there appears to be no mortar
joining them. The wall is more than twenty-five spans
of the gold mines in its vicinity due to civil war.
in width, and the height is not so great considering the The late 15th century was characterized by
width. This edifice is almost surrounded by hills, upon civil wars, which saw the emergence of the
which are others resembling it in the fashioning of the Torwa dynasty in the southwestern regions of
stone and the absence of mortar, and one of them is the Zimbabwe Plateau. Oral accounts from before
a tower more than twelve fathoms high.
The natives of the country call these edifices Sym-
the mid- 17th century point to the movement of
baoe, which according to their language signifies court, Karanga clans from the area of Great Zimbabwe
for every place where Benomotapa may be is so called; northwards. Traditions refer to a region called
and they say that being royal property all the king's "Guruhuswa," literally meaning "area with tall
other dwellings have this name. ... When and by whom
grass," which may be identified geographically
these edifices were raised, as the people of the land with south-central or southwestern Zimbabwe
are ignorant of the art of writing, there is no record,
but they say they are the work of the devil, for in (Garlake 1973), or simply the Zimbabwe high-
comparison to their power and knowledge it does not lands (Beach 1980). This Karanga movement
seem possible to them that they should be the work seems to coincide with the emergence of the
of men. ... The distance of this edifice from Sofala
Mutapa State in northern Zimbabwe. The Mutapa
in a direct line to the west is a hundred and seventy
State has been identified archaeologically with the
leagues, or thereabouts and its is between 20° and 21°
south latitude. ... In the opinion of the Moors whoexpansion of the Zimbabwe tradition northwards
(Beach
saw it, it is very ancient, and was built there to keep 1980; Pikirayi 1993). Stone- walled settle-
possessions of the mines, which are very old, and no
ments architecturally similar to Great Zimbabwe
gold has been extracted from them for years, because
have been located in the region, dating from the
of the wars. ... It is guarded by a nobleman, who has
15th century onwards. Some of these Mutapa
charge of it after the manner of a chief alcaide, and
they call this officer Symbacayo, as we shouldcapitals
say were still being constructed in stone when
theofPortuguese arrived on
keeper of the Symbaoe, and there are always some the Zimbabwe Plateau
Benomotapa's wives therein, of whom this Symbacayo
in the early 16th century.
takes care (Theal 1 898- 1902[6]:267- 268).
The traditions about Guruhuswa origins seem to
convey centuries-old historical processes of migra-
tions across the Zimbabwe Plateau landscapes,
Huffman and Vogel (1991) have used this
source in connection with available radiocarbon
with communities searching for vital resources,
dates to shorten Great Zimbabwe's chronology
such as salt, gold, ivory, game, pasture, and water
in other regions (Abraham 1962). The traditions
and argue for its abandonment mainly during
gleaned by Abraham (1962) about a movement
the 15th century. This would support available
from
archaeological evidence, irrespective of the Guruhuswa to the Zambezi lowlands in
stratigraphie integrity of the terminal phasessearch
of of salt need to be utilized with caution,
settlement at Great Zimbabwe. What this exer- as they have been interpreted to mean that Great
Zimbabwe was abandoned due to a shortage of
cise fails to understand is the process of leaving
Great Zimbabwe, and how this is reflected onthe commodity. However, it is not unusual to find
the site and elsewhere. salt being traded in the northern, northwestern,
It is evident from 16th-century Portuguese and even western Zimbabwe Plateau and areas
accounts of the Mutapa State in northern Zim- beyond, such as the Makgadikgadi Pans (Pikirayi
babwe that by that time Great Zimbabwe was 2001). Karanga expansion northwards during the
not as important a center as it once was. João 15th century must have been triggered by the dis-
de Barros's account was heavily derived from covery of more lucrative goldfields in the region
coastal (Sofala) Swahili traders; for example, that they tried to control and that may have
his references to square buildings and ancient affected Great Zimbabwe in terms of population
inscriptions seen above the door of the buildingdepletion. In other regions such as the southwest,
INNOCENT PIKIRAYI- Great Zimbabwe in Historical Archaeology 33

control of the lucrative goldfields

date.seems to
Available have evidence would suggest
historical
fallen under the Torwa dynasty, evidence
which to was
theclearly
contrary, however, demonstrating
a rival of the Mutapa State and successor to that developments elsewhere on the Zimbabwe
Great Zimbabwe. Plateau trigger the decline of Great Zimbabwe
during the 16th century. From a regional per-
Further contributing to the "silence" of Great
Zimbabwe is the rise of another polity in thespective, the rise of successor states in the north
southwestern regions of the Zimbabwe Plateau.and west played a major role in this, but, more
The rise of the Torwa and the State of Butua importantly, they succeeded in undercutting the
is another poorly understood development incommercial
the links Great Zimbabwe had previously
15th and subsequent centuries. The Torwa are enjoyed
first with the Indian Ocean.
encountered in Arabic accounts written by AhmadConventional archaeological approaches used to
ibn Madjid, dating to 1494, in connection with
definea the chronology of Great Zimbabwe (Huff-
civil war with the Mutapa State. The geographi-
man and Vogel 1979, 1991; Hall and Vogel 1980;
cal location of the State of "Butua" - the area
Chirikure and Pikirayi 2008) give the impression
that the settlement is primarily "prehistoric,"
ruled by the Torwa, - is, according to Portuguese
written accounts, in the southwestern parts of the"history" only starts with the appearance of
since
Zimbabwe Plateau (Beach 1980:197). It wasEuropean
rich texts in the early 16th century. A pre-
in gold and supported large cattle herds.historic
This dating for Great Zimbabwe would identify
coincides with the rise of the center of Khami
the authorship of the site with the "ancients," a
in the region (Robinson 1959). term coined in early antiquarian accounts (Bent
Torwa origins are obscure, but the general
1893; Hall 1905), apparently designed to create a
assumption until most recently is that they were
disconnect between present-day Karanga speakers
a direct successor to Great Zimbabwe (Beach
living in the area and the perceived builders of
1980; Pikirayi 2001). Khami inherits some of the
the town. This results in the production of histo-
architectural as well as material culture elements
ries that are part of a broader process of silencing
from Great Zimbabwe, but it appears the cultureor erasing histories, particularly later histories.
developed locally and earlier, as evidenced byThis has been further conditioned by traditional
Iron Age sites whose remains are found at Leop- labels in archaeological scholarship, where deep
ard's Kopje (Ntabazingwe) and other sites found time is regarded as "prehistory," while the more
within the general area. Khami partly explains therecent pasts, which are illuminated in one way or
"silence" of Great Zimbabwe in the 15th and later the other by the availability of written and oral
centuries, as it seems to have been the dominant texts, are categorized as "protohistory" or "histori-
force in the southwest during this time. Accord-cal" (Schmidt and Mrozowski [2013]).
ing to Huffman (2007) it controlled gold and The absence of Great Zimbabwe in post- 16th-
other resources, which affected the wealth that century historical records partly accounts for the
was once channeled through Great Zimbabwe to "silence" of the site, but, more importantly, to
the Indian Ocean coast. Khami was therefore the the poor understanding of settlement processes
capital of the Torwa State, which the Portuguese elsewhere on the Zimbabwe Plateau. It is evident
found wrestling for control of the northern trade that development in the northerly parts of the
routes through the Zambezi Valley. Its destructionZimbabwe Plateau during the 15th century with
in the 1640s by the Portuguese further confirms the rise of the Mutapa State (Pikirayi 1993) and
the dominance of the Torwa in the southwest the southwestern regions with the emergence of
(Pikirayi 2001). the Torwa represents peer-polity interaction that
eventually shifts political power and dominance
Great Zimbabwe in Post-1500 Historical from the south, where Great Zimbabwe was based.
Developments on the Zimbabwe PlateauBy the time the Portuguese arrive on the Zimba-
bwe Plateau in pursuit of gold and ivory at the
A strict interpretation of available archaeological
beginning of the 16th century (Beach 1980), these
evidence from Great Zimbabwe would suggest areas had become firmly integrated into the global
that the site was no longer an important settle- networks of trade and interaction, which now
ment after 1500, further suggesting that it involved
may mercantile powers from Europe. These
have been abandoned then or even before this developments overshadowed Great Zimbabwe
 34 HISTORICAL ARCHAEOLOGY 47(1 )

as a vibrant city and

on the Zimbabwemay have
Plateau, and how it lost its result
its demise, reducing domineering it into
role in internationalan ordinary
commerce at the
Sixteenth-century very same moment that European
Portuguese expansion was
secondhand r
of the site allude creating to this the modern development.
globalized world. There is a
What also remains need for poorly documented
a coordinated research program focus-
process of late- 16th- ing on the decline
and of Great Zimbabwe and the
17th-century
migrations or movements development of successor
from states on the
the Zimbabwe
northe
of the Zimbabwe Plateau to the south-central Plateau and surrounding regions. Much remains
regions, where Great Zimbabwe is situated.unaccomplished
What insofar as telling the story of
seems to trigger these movements was the grow- Great Zimbabwe is concerned, as fairly recent
ing, unstable, political, economic, and environ- research suggests (Pikirayi 2006, [2013]; Chirikure
mental conditions in the Mutapa State (Pikirayi and Pikirayi 2008). Scientific and historical redat-
2001, 2009), exacerbated by exposure to global ing of the site is important to refine the existing
elements of the gold and ivory trade, which relative and radiocarbon chronology. At present
triggered considerable violence. This resulted in radiocarbon dates do not relate to the
available
the repopulation of the south-central parts ofperceived
the terminal periods of the site, although
Zimbabwe Plateau, which remained shielded from some of the samples were secured from upper
the rapacious demands of external traders until stratigraphie
the levels. A reexamination of Bent
19th century, when the region became exposed (1893), Hall (1905, 1909), Hall and Neal (1902),
to developments associated with the Nguni Randall-Maclver (1906), Gertude Caton-Thompson
movements from the south and the expansion (1931),
of Robinson (1961), and Summers (1961)
Afrikaner and British settlers towards the South presents a clue to the nature of archaeological
African highveld. Karl Mauch's map of the south-deposits removed by early investigators, while
central regions (Burke 1969) suggests an increaserecent condition survey reports would be help-
ful in indicating which of these deposits remain
in population and settlement activity in the region
of Great Zimbabwe up to 1870 (Figure 3). available for study without further destroying the
While all this information is critical in building
site. An archaeological study of imported ceramic,
up a story for post- 1500 Great Zimbabwe, little glass
is beads, and other valuable items brought
to Great Zimbabwe through trade and exchange
known about the actual developments within the
needs to be undertaken.
town or city save for the late- 19th-century Euro-
pean accounts that publicized it to the Western The history of gold mining on the Zimbabwe
world. Thus, Great Zimbabwe remains effectivelyPlateau requires a reexamination (Summers 1969;
Phimister 1976) to document which kingdoms
without a history, its authors are conveniently
associated with an "ancient race" that has no and regions benefited directly from the gold trade
connections with the "decadent" tribes that were when Great Zimbabwe was at its prime. Already,
Kilwa provides useful clues, some of which have
found living within it by European explorers, and
been recovered at Great Zimbabwe in the form of
therefore remains a "mystery" in terms of origins.
minted coins. However, research needs to estab-
This historiography, which puts considerable
lish how much of this trade ended up in the pal-
emphasis on the prehistory - narrowly conceptual-
ized here as the period before written documen-aces of the Persian Gulf, India, and the Far East,
whose written accounts may help define Great
tation - has perpetuated the silence observed in
Zimbabwe's protohistorical period. Of value here
anthropological and historical research (Fontein
2006). This is posing considerable conflict is ininformation relating to the decline in the price
terms of management of the site, as communitiesof gold during the protohistorical period and
its coincidence with references from João de
living in the area around it have been ostracized
from it in one way or the other (Ndoro 2005).Barros 's account of some of the gold mines on
the Zimbabwe Plateau. The paucity of imported
Conclusion ceramics from India and the Far East during the
15th century has been interpreted to mean the
Future research on Great Zimbabwe must decline of Asian trade with Great Zimbabwe.
adopt two themes - how the settlementWhile this is technically correct, one should also
is con-
consider the royal policy that banned exports
nected to modern Karanga and other histories
INNOCENT PIKIRAYI - Great Zimbabwe in Historical Archaeology 35

w » ar ^

I
I - ^ i jw

I ar
o» »v c.
MaâandùW,
m t A 'L , < ,, ^ ^
'i'Ķ A'* ' vi- )s «■
tí i b ti
i i Uteunif»/-'
M a biH ¡^sāĻļji
H ¡¡¡faŽáí
, Ì 'K A vj
)

^Maika , ^ zJÊ1
.V. «ì ' ( /^ MajUhMBAa /l

^ ļ jļf5y ; 25.*™ py
FIGURE 3. Map by geologist Karl Mauch illustrating settlement activity around Grea
regions of the Zimbabwe Plateau in 1870. (Redrawn from Burke [1969].)
 36 HISTORICAL ARCHAEOLOGY 47(1 )

to the West. All of these factors demonstrate Carroll, Scott T.

1988 Solomonic Legend: The Muslims and the Great
that while much of Great Zimbabwe's history
Zimbabwe. International Journal of African Historical
predates the period of European expansion
Studies 21(2):233-247.
typically associated with the transformations
brought about by modern globalization, itCaton-Thompson,
was Gertrude
nonetheless associated with relevant transforma- 193 1 The Zimbabwe Culture: Ruins and Reactions. Clarendon

tive themes. Press, Oxford, UK.

Chirikure, Shadreck, and Innocent Pikirayi

Acknowledgments 2008 Inside and Outside the Dry Stone Walls: Revisiting
the Material Culture of Great Zimbabwe. Antiquity
Earlier versions of this paper were delivered 82(318):976-993.
at the conference of the Five Hundred Years
Colwell-Chanthaphonh, Chip, and Thomas J. Ferguson
Initiative in Cape Town in 2008, which sought
2006 Rethinking Abandonment in Archaeological Contexts.
to redefine trends in historical archaeology in SAA Archaeological Record 6( 1 ):37-4 1 .
southern Africa. I would like to thank confer-
Diamond, Jared M.
ence participants for their highly constructive
2005 Collapse: How Societies Choose to Fail or Succeed.
comments and feedback. This paper is part of Viking Penguin, New York, NY.
wider research on the collapse of complex soci-
Fontein, Joost
eties, funded through the South African National
2006 The Silence of Great Zimbabwe: Contested Landscapes
Foundation Blue Skies Research Programme
and the Power of Heritage. UCL Press, London, UK.
(BSRG) Grant 81694.
Garlake, Peter
References 1968 The Value of Imported Ceramics in Dating and
Interpretation of the Rhodesian Iron Age. Journal of
Abraham, Donald P. African History 9(1): 1 3-33 .
1 962 The Early Political History of the Kingdom of Mwene 1973 The Great Zimbabwe. Thames and Hudson, London,
UK.
Mutapa (850-1 589). In Historians in Tropical Africa:
Proceedings of the Leverhulme Inter-Collegiate
Hall,
Conference , pp. 6 1-92. University College of Rhodesia Martin, and John C. Vogel
1 980 Some Recent Radiocarbon Dates from Southern Africa.
and Nyasaland, Salisbury, Rhodesia [Zimbabwe].
Journal of African History 21(4):431-455.
Beach, David N.
1980 The Shona and Zimbabwe 900-1850. Mambo Press. Hall, Richard N.
Gweru, Zimbabwe. 1905 Great Zimbabwe. Methuen, London, UK.
1909 Prehistoric Rhodesia. F. T. Unwin London, UK.
Bent, Theodore J.
Hall,
1 893 The Ruined Cities ofMashonaland . Reprinted 1 969 byRichard N., and W. G. Neal
Books of Rhodesia, Bulawayo, Rhodesia [Zimbabwe]1 .902 The Ancient Ruins ofRhodesia. Methuen, London, UK.

Binford, Louis R. Huffman, Thomas. N.

1972a An Arab Coin from Zimbabwe. Arnoldia 5(32): 1-7.
1978a Dimensional Analysis of Behavior and Site Structure:
Learning from an Eskimo Hunting Stand. American 1972b The Rise and Fall of Zimbabwe. Journal of African
Antiquity 43(3):330-361. History 13(3):353-366.
2007 Handbook to the Iron Age: The Archaeology of
1978b Nunamiut Ethnoarchaeology. Academic Press, New
York, NY. Pre-Colonial Farming Societies in Southern Africa.
University of KwaZulu-Natal Press, Scottsville, South
Africa.
Burke, E. E. (editor)
1969 The Journals of Carl Mauch. National Archives of
Rhodesia, Salisbury, Rhodesia [Zimbabwe]. Huffman, Thomas N., and John C. Vogel
1979 The Controversial Lintels from Great Zimbabwe.

Cameron, Catherine M. Antiquity 53(207):55-57.

1993 Abandonment and Archaeological Interpretation. 1991 The Chronology of Great Zimbabwe. South African
In Abandonment of Settlements and Regions: Archaeological Bulletin 46:61-70.
Ethnoarchaeological and Archaeological Approaches,
C. M. Cameron and S. A. Tomka, editors, pp. 3-7. Mudenge, Stan I. G.
Cambridge University Press, Cambridge, UK. 1988 A Political History of Munhumutapa. Zimbabwe
Publishing House, Harare.
 INNOCENT PIKIRAYI- Great Zimbabwe in Historical Archaeology 37

Ndoro, Webber Robinson, Keith R.

2005 The Preservation of Great Zimbabwe : 1Your Monument,
959 Khami Ruins. Cambridge University Press, Cambridge,
Our Shrine. ICCROM Conservation StudiesUK. 4. Rome,
Italy. 1 96 1 Excavations on the Acropolis Hill. Occasional Papers
oftheNational Museums of Rhodesia 3(23 A): 1 59-1 92.
Nelson, Margaret C.
2000 Abandonment: Conceptualization, Representation Schiffer, Michael B.
and Social Change. In Social Theory in Archaeology , 1 995 Behavioral Archaeology: First Principles. University
Michael Schiffer, editor, pp. 52-62. University of Utah of Utah Press, Salt Lake City.
Press, Salt Lake City.
Schmidt, Peter, and Stephen A. Mrozowski (editors)
Phimister, Ian R. [2013] The Death of Prehistory. Oxford University Press,
1976 Pre-Colonial Gold Mining in Southern Zambezia: A Oxford, UK.
Reassessment. African Social Research 2 1 : 1-30.
Sinclair, Paul J. J.
Pikirayi, Innocent 1987 Space, Time and Social Formation: A Territorial
1993 The Archaeological Identity of the Mutapa State: Approach to the Archaeology and Anthropology of
Towards an Historical Archaeology of Northern Zimbabwe and Mozambique, c. 0-1 700 AD. Societas
Zimbabwe . Societas Archaeologica Upsaliensis, Archaeologica Upsaliensis, Uppsala, Sweden.
Studies in African Archaeology 6. Uppsala, Sweden.
200 1 The Zimbabwe Culture : Origins and Decline in Southern Summers, Roger
Zambezian States. AltaMira Press, Walnut Creek, CA. 1961 Excavations in the Great Enclosure. Occasional
2006 The Demise of Great Zimbabwe, AD 1 420-1 550: An Papers of the National Museums of Southern Rhodesia
Environmental Reappraisal. In Cities in the World, 3(23A):236-288.
1500-2000, A. Green and R. Leech, editors, pp. 3 1-47. 1969 Ancient Mining in Rhodesia. National Museums of
Society for Post-Medieval Archaeology Monograph 3. Rhodesia Museum Memoir 3. Salisbury, Rhodesia
Leeds, UK. [Zimbabwe].
2009 Palaces, Feiras and Prazos: An Historical Archaeological
Perspective of African-Portuguese Contact in Sutton, John E. G.
Northern Zimbabwe. African Archaeological Review 1990 A Thousand Years of East Africa. British Institute in
26(3): 163-1 85. Eastern Africa, Nairobi, Kenya.
[2013] Waterand Social Formation in Pre-colonial Zambezia:
Rethinking the Development and Demise of Complex Tainter, Joseph
Societies in Southern Africa. In Water and Humanity: 1 988 Collapse of Complex Societies. Cambridge University
Historical Overview , Vol. 7, V. Scarborough, editor. Press, Cambridge, UK.
UNESCO-IHP Manuscripts, Paris, France.
Theal, George M.
Pirazzoli-t' Serstevens, Michèle P. 1898-1903 Records of South-Eastern Africa , 9 vols. C.
1 985 The Ceramic Route. In The World Atlas of Archaeology, Struik, Cape Town, South Africa.
M. Beazely, editor, pp. 284-285. G. K. Hall, Boston,
MA. Innocent Pikirayi
Department of Anthropology and Archaeology
Randall-MacIver, David University of Pretoria
1906 Medieval Rhodesia. Macmillan, London, UK. Private Bag X20
Hatfield, Tshwane, 0028
South Africa