Content uploaded by Shaw Badenhorst
Author content
All content in this area was uploaded by Shaw Badenhorst on Aug 24, 2015
Content may be subject to copyright.
148 South African Archaeological Bulletin 64 (190): 148–155, 2009
Research Article
THE CENTRAL CATTLE PATTERN DURING THE IRON AGE OF
SOUTHERN AFRICA: A CRITIQUE OF ITS SPATIAL FEATURES
SHAW BADENHORST
Department of Archaeozoology, Transvaal Museum, P.O. Box 413, Pretoria, 0001, South Africa
and
Research Associate, Institute for Human Evolution, University of the Witwatersrand,
Private Bag 3, WITS, 2050, South Africa
E-mail: shaw@nfi.museum
(Received February 2009. Revised September 2009)
ABSTRACT
The Central Cattle Pattern (CCP) is the dominant model to interpret
theIronAgeofsouthern Africa. Protagonists of the CCP arguethatthe
model can be traced back to the Early Iron Age, when the first farmers
settled in the region. In this paper, I suggest that the spatial features of
the CCP are not diagnostic, and not restricted to patrilineal, Eastern
Bantu-language speakers. The spatial features addressed include hut
shape and village layout, centrally positioned livestock kraals, storage
pits and burial data. The spatial features of the CCP, when applied to
Early and Middle Iron Age sites, are inappropriate and mimicked by
other cultures.
Keywords: Central Cattle Pattern, cattle, caprines, Iron Age,
farmers.
INTRODUCTION
The Central Cattle Pattern (CCP) remains the dominant
model for interpreting the Iron Age of southern Africa (over-
view by Mitchell 2002). Although many have criticized the CCP
(Hall 1986, 1987a,b; Lane 1994/1995; Maggs 1994/1995), the
impact has been minimal. This is partly due to the (perceived)
problematic nature of other models countering the CCP, a lack
of newer alternative models, and responses to the original
criticisms by the protagonists of the CCP (e.g. Huffman 1990,
2001, 2007). It may seem as if the antiquity of the CCP in
southern Africa has been accepted (Mitchell & Whitelaw 2005).
In this paper, I provide evidence countering the CCP. I suggest
that the model is not appropriate when applied to Early and
Middle Iron Age farmers of southern Africa. I first provide a
brief overview of farming settlement in southern Africa, as well
as the CCP, and counter-arguments to the model.
THE ARCHAEOLOGY OF FARMERS IN
SOUTHERN AFRICA
BRIEF OVERVIEW
The first farmers that settled in southern Africa did so
during the first millennium AD, in the time generally referred
to as the Early Iron Age (c. AD 200–900). These farmers
cultivated plants, and kept domestic animals, although some
exceptions may have occurred. Faunal evidence indicates that,
generally, sheep was the numerically dominant domestic
animal (based on both NISP and MNI for large samples).
Human populations were low at the time. People lived in
pole-and-daga structures. They manufactured ceramics, and
usedmetaltechnology. Towardstheendofthefirstmillennium
AD (the Middle Iron Age dating to between ca. AD 900–1300),
some settlements such as Mapungubwe in the Limpopo Valley
became incorporated in the trans-Indian Ocean trade network.
Communities such as those on the Zimbabwe Plateau supplied
gold, ivory and other commodities in exchange for imported
goods such as glass beads. Great Zimbabwe succeeded
Mapungubwe in controlling these trade networks during the
Later Iron Age (c. AD 1300–1820s). Construction in stone
became widespread during the second millennium AD. Sheep
continued to dominate faunal assemblages until the early second
millennium AD, when cattle began to outnumber caprines.
During the past few centuries, travellers and missionaries
documented the role of cattle in the social realm of farmers.
From recent times, it is well known that the Eastern
Bantu-language speakers of southern Africa used cattle in
ceremonies, bride wealth payments and exchange (e.g. Kingon
1919; Schapera 1953a; Maggs 1984; Plug & Voigt 1985; Plug
1996, 2000; Mitchell 2002; Badenhorst 2008).
Ceramic data suggests a link between modern Shona-
Venda speakers and first millennium AD farmers in southern
Africa (Huffman 2007). Descendants of modern Nguni and
Sotho-Tswana speakers arrived during the early to middle
second millennium AD in southern Africa. These new inhabit-
ants brought with them new building styles, different settle-
ment locations, ceramics and other forms of material culture.
They interacted with and absorbed those farmers who had
already occupied the region (e.g. Mitchell & Whitelaw 2005).
THE CENTRAL CATTLE PATTERN
Thomas Huffman’s Central Cattle Pattern (Fig. 1), which is
derived from an ethnographic model of Nguni and Sotho-
Tswana settlement patterns (Kuper 1982; also Bruwer 1956;
Mönnig 1967), explains the internal structure and layout of
Iron Age farming villages in southern Africa (Huffman 1982,
1986, 1990, 1993, 1997, 1998, 2001, 2004, 2007). Briefly, the CCP
holds that the head man of the community resides opposite the
entrance and that his wives live in huts on either side of his hut
and that all these buildings are arranged around a central cattle
kraal (enclosure or byre). The central part of the village is a focal
point of ceremonies and iron forging. Important people are
buried in the central region, usually in or close to the cattle
kraal. Men are associated with the central area, to which there
is restricted access for women. There may be several cattle
kraals in the centre of a village, arranged in a variety of ways.
Heads of villages store grain in the central region, whereas
women, associated with huts, store theirs in granaries in the
courtyard of their homes. The CCP is associated with Eastern
Bantu-language speakers, with hereditary male leadership,
and a patrilineal form of descent in which cattle is the preferred
bride wealth. Space in a village is organized with opposing
South African Archaeological Bulletin 64 (190): 148–155, 2009 149
values of male:female, public:domestic, top:bottom and
left:right (see also summaries by Mitchell 2002: 279–288;
Whitelaw 2005: 123).
According to Huffman (2004: 67), the CCP derived from a
theory about the relationship between worldview, social
organization and settlement pattern. In addition, the features
of the CCP are not limited to any ethnic group or political
community, but can be found among patrilineal Eastern
Bantu-language speakers who use cattle for bride wealth
(Huffman 1982: 140). As a structuralist model, it is projected
back in time and applied to farming settlements by using the
direct historical approach. Using this approach, it is concluded
that, since the ethnographically-derived model can be applied
as far back as the Early Iron Age, then the worldview and
beliefsmust have been the same (Huffman 2007: 328). Evidence
for the CCP has been reported from Early Iron Age sites in
South Africa, including Ndondondwane (Greenfield et al. 2000;
Greenfield & Van Schalkwyk 2003; Greenfield & Miller 2004),
Nanda (Whitelaw 1993) and KwaGandaganda (Whitelaw
1994) in KwaZulu-Natal, as well as Broederstroom (Huffman
1990, 1993) in the North West Province.
Although Huffman (1989: 158–159) defined other spatial
aspects such as the interior layout of huts, these have little or no
archaeological visibility, and are not used to distinguish
between settlement patterns of Eastern and Western Bantu-
language speakers in the archaeological record (cf.Huffman
1997:152–153,2001).Asaresult,theyarenotconsideredfurther
here.
During the Later Iron Age, stone walls clearly demarcate
many spaces in villages, and the model is applied with success
(e.g. Pistorius 1992). However, when the CCP is applied to
settlements of the Early and Middle Iron Age, which lack stone
constructions, the following criteria are typically considered as
evidence supporting the model (cf.Huffman 1982: 140, 1997:
152–153, 2001; Whitelaw 1994: 49, 2005: 123):
•A central cattle kraal(s);
•huts arranged around the central cattle kraal(s);
•storage pits associated with kraals and huts;
•elite burials in kraal(s), usually of men; and
•iron forging in the central part.
An ethnographic study in the Eastern Cape of South Africa
has shown just how complex the relationship between social
relations and material expressions are with regard to settlement
layout (Davison 1988). In fact, even applying ethnographic
information to Late Iron Age settlements in KwaZulu-Natal
proved difficult (Hall 1984; also see Segobye 1998). I do not
review the criticism of and counter-arguments to the CCP in
great detail here. Nonetheless, it suffices to mention that signif-
icant changes have occurred over the course of the last two
thousand years among farming communities in southern
Africa (cf.Lane 1994/1995). Some of the major changes and
differences include (1) dominant matrilineal descent during
the Early and Middle Iron Ages, with dominant patrilineal
descent only associated with the Late Iron Age (cf.Oliver 1982;
Holden & Mace 2003; Badenhorst, in press); (2) dominant
caprine herding during the Early and Middle Iron Ages, with
dominant cattle herding associated with the Late Iron Age
(Plug & Voigt 1985; Hall 1986, 1987a,b; Badenhorst 2008, 2009);
(3) human teeth modification during the first but not second
millennium AD (Morris 1993); (4) greater reliance on fish
during the first but not the second millennium AD (Maggs
1994/1995); (5) settlements located on valley bottoms and next
to rivers during the first but not the second millennium
AD when sites are typically located in uplands (Maggs & Ward
1984); (6) settlements placed in savanna environments (both
grasslandsandwoodedregions)duringthefirstbutnotsecond
millennium AD when sites are located in grassland regions
(Maggs 1994/1995); (7) changes in ceramic fabric, thickness,
surface treatment, form and decoration (Maggs 1980a); (8) the
presence of pits with bottomless pots at first but not second
millennium AD sites (Maggs 1980b); and (9) closer interaction
with hunter-gatherer communities during the first than second
millennium AD (Maggs 1980b). While some of these traits, such
as teeth modification are dismissed as low level details not
included in the CCP (Huffman 2001), they suggest that great
caution is necessary when applying the model to farmers from
theEarlyandMiddle Iron Age (cf.Plug & Badenhorst, in press).
Although tooth modification is widespread today in many
parts of southern Africa (Van Reenen 1987), they are not found
at Late Iron Age sites. I concur with Maggs (1980a,b, 1984,
1994/1995; Maggs & Ward 1984) that the cultural changes listed
above indicate a significant difference between the worldview
of first and second millennium AD farmers.
A CRITIQUE OF SOME SPATIAL FEATURES OF THE CCP
None of the previous critiques of the CCP dealt thoroughly
with spatial issues, an aspect I will address in this paper, with
reference to the Early and Middle Iron Age. I do not consider
issues pertaining to metal working (but see Van Schalkwyk
1987; Miller & Whitelaw 1994; Greenfield & Miller 2004;
Whitelaw 2005). With evidence for significant socio-political
and economic changes during the last 2000 years, the question
arises as to why archaeologists find evidence supporting the
CCP from the Early Iron Age onwards? To answer this
question, two hypotheses can be formulated. First, that the
CCP can indeed be traced back to the Early Iron Age of
southern Africa. Second, that the CCP does not extend back
to the Early and Middle Iron Age (i.e. before the arrival of
ancestral Nguni- and Sotho-Tswana speakers) and that the
evidence presented in support of the model is incorrect,
inappropriate, and/or mimicked by other cultures.
In this paper, I provide support for the second hypothesis,
by showing that the criteria used for recognizing the CCP are
not diagnostic when applied to Early and Middle Iron Age
farming settlements. I next present data countering the CCP
model.
FIG. 1. Simplified layout of CCP villages (redrawn from Huffman 2007:25).
150 South African Archaeological Bulletin 64 (190): 148–155, 2009
WESTERN BANTU-LANGUAGE SPEAKER SETTLEMENT
PATTERNS
One of the strongest arguments used in support of the CCP
is that the settlement pattern of Western Bantu-language
speakers differs from those of Eastern Bantu-language speakers
(e.g. Huffman 1986, 1989, 1997, 2007). Following the CCP, the
farmers of west-central and central Africa live in rectangular
houses, often arranged in a more or less rectangular form or in
parallel opposing rows on a street (see also Herskovits 1930: 72;
Ruel 1962). In addition, the men’s court was at the end of the
street or open spaces near the headman’s house, and storage
pits were not placed in the village’s centre. V-shaped pits were
sometimes dug behind the village to process palm oil. Most
people are also buried in cemeteries outside the village in
extended positions, and graves often have shrines on them
(Huffman 1989: 158, 1997). Others have also noted that rectan-
gular and circular huts occur in West and East Africa, respec-
tively (e.g. Torday 1921: 380–381).
Rectangular huts have been present in regions such as
Gabon and Cameroon since the first millennium BC onwards.
However, this does not include regions such as Angola and the
Democratic Republic of the Congo (DRC), where no archaeo-
logical information on hut shapes is as yet available (cf.Van
Noten 1982; Vansina 1990; Clist 2004–2005). These regions are
key to understanding southward migrations during the first
millennium AD (Huffman 2007). However, on the fringes of
central-west Africa both circular and rectangular huts have
been found at archaeological sites. While sub-rectangular
dwellings have been found at Kumadzulo in the Victoria Falls
region of southern Zambia dating to between (calibrated)
AD 390 and AD 660 (Vogel 1971a: 18–26), at Kamangoza,
Simonga and Simbusenga, also in the same region dating to the
latterhalfofthefirstmillenniumAD,circularhutswerepresent
(Vogel 1971b: 87, 1975: 22,63). The Late Iron Age site of
Simulombwa Hill, also in southern Zambia, has circular huts
arranged around a central open space (Musonda 1987). At
Pouss in the Upper Benue Basin of Cameroon dating to the
second half of the first millennium AD, circular hut-shapes
havebeenfoundinassociationwithmounds(David1981:86).
Although it may seem as if these examples from Zambia
support the CCP, this is not necessarily the case. This is because
the CCP fails to consider, first, that there is evidence from
ethnographic and historical sources that rectangular huts
found in west-central Africa are recent adaptations, and,
second, that circular huts occur in the region, especially in the
DRC and Angola.
Considering first the issue of whether the settlement
patterns of Western Bantu-language speakers are recent
adaptations, there is evidence in support of this notion. Euro-
pean influences date back to the 15th century in west-central
and western Africa (e.g. Thornton 1984). This is particularly
important, as all the ethnographic examples of rectangular
houses under the Western Bantu-language speakers presented
by Huffman (1989: 158) in support of the ‘Western-Bantu street
pattern’, date to the 20th century. Observers have pointed out
that the rectangular houses in use among Western Bantu-
language speakers are recent adaptations. For example, Torday
(1910: 30) noted in the Kasai Basin in the west-central parts of
the DRC:
In the south the native form of hut is circular, but this type is
disappearing in favour of a rectangular pattern copied from
the houses of Europeans; those of the Olemba are also circular,
but further north the rectangular type of the forest is found.
A few examples of people living in different parts of
central-west, western and southern Africa illustrate that
rectangular huts do occur in the region as a result of external
influences. For example, photographer Herbert Lang noted
early in the 20th century in the DRC among the Azande how
the rectangular shape of some structures, such as houses of
chiefs, were a result of European influence (Schildkrout 1991:
84–85). In Déla of northern Cameroon, circular houses are also
disappearing in favour of rectangular houses. In this particular
case, both European and Islamic influences altered existing
settlement patterns (Lyons 1996). In the Banda-Wiae region of
Ghana, the shift from circular to rectangular houses occurred
some 400 years ago (McIntosh & McIntosh 1986: 434). A similar
process of adopting rectangular constructions in recent times
was noted among the Southern Sotho of the Free State in South
Africa and in Lesotho (Walton 1958: 133). Others have noted
similar changes elsewhere in South Africa (Schapera 1953b:
364; Hall 1997: 212–215). In fact, changes from circular to rectan-
gular homesteads could be accompanied by significant
changes in gender roles (Hall 1997).
In addition to adopting rectangular house-shapes due to
European and Islamic influences, circular huts occur in
west-central and western Africa. For example, Johnston
(1884: 468) noted in southwestern Angola amongst Western
Bantu-language speakers:
The general type of dwelling is a round hut, built of clay or
wattled, with a peaked thatch roof. The round house or hut
seems to go no further north than the southern bank of the
Quanza [River], where it is replaced by the rectilinear, oblong
building made of matting, interwoven palm-leaves, wooden
posts, and dried grass.
The Quanza River originates in central Angola, and flows
north and then west into the Atlantic Ocean just south of
modern-day Luanda in northern Angola, so Johnston (1884:
468) is presumably referring to large parts of southwestern,
central, and northwestern Angola. This region encompasses
the area considered to be the origin of the Kalundu Tradition
from the Early Iron Age (Huffman 2007).
To mention a few other examples of circular huts in
central-west Africa, although not all from Western Bantu-
language speakers, amongst the Azande people of the north-
eastern DRC “(t)heir huts are usually circular in form, with
conical thatched roof,…” (Burrows 1899: 41). Among the
Mangbetu, also from the northeastern DRC, huts are “…rect-
angularorcircularinshape…”(Burrows1899:44).Inthesouth-
western parts of the DRC, Torday and Joyce (1907: 139) noted
for a number of groups that both rectangular and semi-
cylindrical huts occurred in the region. In the village of Ganga
for example, most huts are rectangular, but that of the chief was
semi-cylindrical. In addition, in the region, huts were built
without any systematic arrangement (Torday & Joyce 1907:
139). Circular huts have also been noted in the Upper Congo
Basin of the Central African Republic (Caldwell & Sheldon
1930: 320) and different parts of Nigeria, including the eastern
(Fitzpatrick 1910: 47) and southwestern (Moloney 1890: 603)
portions (also Foyle 1953). Admittedly, not all these groups are
Western Bantu-language speakers, and not all live in central
and central-west Africa. However, due to the lack of archaeo-
logical evidence for either rectangular or circular huts in this
region, arguments in favour of applying the CCP before the
Late Iron Age in southern Africa need to take cognisance of the
presence of round huts in the region.
Experimental studies suggest that circular huts are more
durable and cost-effective than rectangular ones (McIntosh
1974: 159). In Déla of northern Cameroon, rectangular buildings
are considered to be more prestigious, and their building
requires the expertise of a skilled mason (Lyons 1996: 359).
South African Archaeological Bulletin 64 (190): 148–155, 2009 151
Among the Shuwa, an Arabic Muslim pastoral group in northern
Cameroon, the shift from circular to rectangular houses coin-
cided with a decrease in mobility and an increase in sedentary
lifeways (Holl & Levy 1993: 176), a process that was also noted
elsewhere in the world (Whiting & Ayres 1968; Flannery 2002).
A study of societies worldwide found a strong correlation
between large rectangular multi-roomed dwellings and large
extended families, or with high rank in status hierarchies, or
both (Willey 1968: 218; Whiting & Ayres 1968). These few
examples indicate that a range of social and practical issues
requirecarefulattentionwhenconsideringsettlementpatterns.
Huffman (1989: 159) acknowledged that circular huts and
circular villages occur in southern Zambia amongst the Ila and
Tonga (e.g. Thurnwald 1929: 360). In addition, the interior
layout of huts amongst both the Ila and Tonga reflect a pattern
found among Western Bantu-language speakers. According to
Huffman (1989: 159):
…even though these Ila and Tonga altered their settlement
patterns ‘externally’, so to speak (perhaps because of their
contactwithEasternBantu),theirWestern Bantuancestrywas
still reflected ‘internally’… (emphasis mine).
However, it is important to note two issues on this point.
First, it has not yet been demonstrated that the circular huts
amongst the Ila and Tonga are due to cultural influences of
Eastern Bantu-language speakers. Until this can be achieved,
such propositions by protagonists of the CCP cannot be
accepted. Unfortunately, with little archaeological excavations
in the region, it will take some time to determine beyond any
doubt when circular huts appeared in the region. Second, an
alternative and equally plausible explanation would be that
circular huts were common among Western Bantu-language
speakers, and not due to cultural contact with Eastern Bantu-
language speakers. This aspect too needs confirmation.
Notwithstanding a lack of archaeological evidence for hut
shapes in central and central-west Africa, preservation issues
are also a limiting factor. Central-west Africa receives more rain
than regions to the south. As a result, the shape of hut floors is
impossible to ascertain in many cases. For example, Fagan
(1969: 84) found it impossible to determine hut shapes at the
late first millennium AD site of Ingombe Ilede in the Lusitu area
ofsouthern Zambia. Fromthisexampleitisclearthatour ability
to recognize hut shapes is seriously impaired by preservation
issues.
In a similar vein, the cemeteries in use in modern Angola
and the DRC which Huffman (1989: 158) describes from 20th
century ethnographies, are likely, in many cases, to be recent
adaptations. For example, the cemetery of Mbanza Mbata in
the western DRC close to the border of northern Angola
contains high-ranking Bantu-language speakers. The graves,
and foundations of the church, date to between the 17th and
end of the 18th or beginning of the 19th centuries AD
(Vanderhoute 1973 in de Maret 1982: 82). In the south-central
DRC, local traditions have it that an old cemetery near the
modern town of Tshofa was used by the Kalebwe people living
in the area (de Maret 1982: 86).
Cemeteries predating European contact include the late
first millennium AD site of Sanga in the south-central parts of
the DRC (de Maret 1977), whilst the nearby sites of Kamilamba,
Kikulu and Malemba Nkulu yielded together 265 Iron Age
burials. Other Iron Age sites in the region also yielded graves
from cemeteries (de Maret 1982: 89). The Houlouf cemetery in
northern Cameroon dates to the second millennium AD (Holl
1994). Notwithstanding archaeological evidence for cemeteries,
some early eyewitness accounts in the Lower Congo region of
the DRC noted a lack thereof (Morgan 1884: 184). There is
ethnographic evidence which indicates that people, elites and
non-elites, were buried in and around villages in central
Cameroon for example (Malcolm 1923: 394–395). With regards
to cemeteries, no firm archaeological evidence has as yet been
presented to support opposing burial practices.
From the examples presented above, it is clear that the
evidence cited in support of the CCP that settlement patterns
found among speakers of Western Bantu-languages in
west-central and central Africa differ from those of Eastern
Bantu-language speakers (Huffman 1997: 151–152, 2001: 21), is
ambiguous. Firm archaeological evidence is still lacking from
central-west Africa, making it impossible to assess the validity
of the CCP model when applied to the Early and Middle Iron
Ages of southern Africa.
CENTRAL CATTLE KRAALS
According to the CCP, kraals should be found in the centre
of a village surrounded by huts (Huffman 2001: 20). The central
cattle kraals can be arranged in a variety of ways (Whitelaw
2005: 123). At KwaGandaganda for example, six cattle kraals are
located in the centre of the settlement in a loose cluster
(Whitelaw 1993, 2005: 123).
The first questionable aspect of the CCP contention of a
central cattle kraal is whether the dung found in these deposits
is actually that of cattle. Peter (2001) argued that many petrified
dung deposits assumed to be the result of cattle herding in
parts of southern Africa, are in fact those of sheep. I have
argued that cattle produce more dung per day than caprines
and, in addition, cattle dung decomposes more slowly than
that of small stock (Badenhorst 2008, 2009). This would result in
conspicuous cattle dung deposits. Zooarchaeological evidence
indicates that caprines often dominate faunal assemblages of
farming settlements predating the second millennium AD (for
a summary of sites, see Plug & Voigt 1985; Badenhorst 2008).
Therefore, we cannot accept that the kraals at settlements
predating the second millennium AD were necessarily used
exclusively to keep cattle.
Second, there are important functional reasons to keep
livestock in kraals close to or in the centre of a settlement. Any
village with livestock faces the burden of raids by other groups
in the area, and predators. Predators were a constant concern
(e.g. Livingstone 1857: 371–372; Beinart 1998), especially before
the spread of guns. For example, Walton (1958: 135) mentions
for the Southern Sotho of South Africa’s Free State and
Lesotho:
[m]any stories are told of lions and other wild animals being
gored to death in attempting to reach calves through the
cattle-kraal.
Lions in particular posed a serious threat to cattle and
sheep. However, a range of other predators such as hyaenas,
wild dogs, leopards (Skovlin 1971: 267–268; Beinart 1998)
and even baboons (Marais 1974: 74) cause livestock losses in
southern Africa. Current distribution maps of lions indicate
that they are absent from the Congo Basin in the DRC, the
Congo Republic, Gabon, and Equatorial Guinea continuing in
a wide band into West Africa (Skinner & Smithers 1990: 405).
These regions represent dense forest zones, which lions avoid
(Rautenbach 1982: 149; Smithers 1983: 375). If (some) farmers
from the Early Iron Age migrated from forested regions in
central and west-central Africa in the first millennium AD to
the savannas of southern Africa (see Mitchell 2002), then it is
very likely that settlement patterns had to be adapted to ensure
optimal protection from predators such as lions. The most
effective manner to provide optimal protection for livestock is
to build a kraal in the middle of a village.
Not only carnivores posed a serious threat to livestock.
Villages with livestock were targets of raids by other groups.
Numerous historical records documented raids among
farming communities in southern Africa before and after the
spread of guns (e.g. Atmore & Sanders 1971; Beach 1974;
Kingon 1919). Considering threats of raids and predators, it is
not surprising to find that other, non-Eastern Bantu-language
speakers in Africa, such as Khoekhoen pastoralists in the
Northern Cape, kept livestock in the centre of villages (e.g.
Laidler 1936). The Shuwa – Arabic Muslim nomads of northern
Cameroon who herd cattle – also keep their animals in the
centre of compounds (Lyons 1996: 360). At the Early Iron Age
village site Kumadzulo in southern Zambia (Vogel 1971a) it is
possible that cattle were herded in the central portion of the
site, which is surrounded by sub-rectangular dwellings. The
central portion of this site contained little cultural debris,
whereas the surrounding area yielded abundant evidence for
occupational activity (Vogel 1971b: 10). Even if the central
portion was not a kraal at Kumadzulo, but rather a central
meeting place within the village, it indicates that a circular
layout occurred amongst different people in Africa.
Not every farming village in southern Africa was necessarily
under threat from raids and predators, as human and carnivore
population densities, as well as relations with neighbouring
villages varied. Villages were constructed in a circular shape
not only to ensure protection for livestock. Social aspects were
also important. Small circular villages do not create social
barriers between people (see Draper 1976: 201–202 for !Kung
camps). It would not be difficult to imagine a situation where
the huts of an Early and Middle Iron Age village were arranged
around a central kraal, or central meeting place, where social
activities were performed. For example, among the Mangbetu
of the northeastern DRC, villages “…are sometimes rectangu-
lar, sometimes circular, occasionally some 200 yards in diame-
ter. The huts are built next to one another at fairly regular
intervals, the centre being kept quite clear and swept twice a
day by the women of the village” (Burrows 1899: 44). It is not
clear from this particular description what the use of the central
space was for, although no mention is made of livestock. In an-
other example, this time from northern Cameroon, Mura-Urza
compounds contain circular huts, arranged in a circle, with
male and female granaries in the centre. Animals are contained
inside huts at night (Lyons 1996: 360).
From a functional perspective, there are very good reasons
to construct kraals for livestock in or near the centre of a village
to provide optimal protection. The functional role of centrally
located kraals does not necessarily relate to a worldview, and
therefore central kraals do not provide direct evidence for the
CCP. We would expect to see this pattern in many parts of
Africa among farmers and pastoralists who keep livestock. In
addition, a circular village can be a social construct unrelated to
the CCP. The concentration of livestock kraals or open spaces in
the centre of villages is not a diagnostic feature of Eastern
Bantu-language speakers.
STORAGE PITS
According to protagonists of the CCP, the presence of
dung-lined storage pits in cattle kraals and huts are another
featureofthe model (Huffman 2001). I have already mentioned
that the kraals may not have been used exclusively for keeping
cattle.Twoissues are of relevance here. First, pits occur at settle-
ments in west-central and central Africa, and second, the func-
tion of pits varies. Pits are not just a feature of the villages of
Eastern Bantu-language speakers (although protagonists of
the CCP do not necessarily make this claim). For example, stor-
age and refuse pits from the first millennium BC have been
found in southern Cameroon (Mbida et al. 2000; Eggert et al.
2006) and Gabon-Equatorial Guinea (Clist 1987: 380). First
millennium AD sites in the region also yielded pits (Clist 1992).
An ethnographic account from Mura-Urza compounds in
northern Cameroon indicates that male and female granaries
are located in the centre (Lyons 1996: 360). If we find pits at
these settlements in central-west Africa well before the spread
of farmers to southern Africa, and people from this region
migrated to southern Africa during the first millennium AD,
then it is not strange to find pits on Early Iron Age farming
settlements in southern Africa. According to Huffman (1989:
158), storage pits are not found in village centres amongst
Western Bantu-language speakers. Again, no archaeological
evidence has as yet been presented to support this view, with-
out which this criterion for the CCP cannot be accepted.
Second, the location of pit features within villages relates to
their usage and function. Pits were often used to store grain,
although this was not always the case. Wild fruits, such as wild
loquat and spiny monkey-oranges, were also stored and
ripened in pits. Ethnographic information from Zimbabwe
suggests that pits in cattle kraals were often preferred since the
heatofdungassistsinthecuringofgrain,whilstthetramplingof
cattle prevents damp from entering the enclosed pit (Robinson
1963: 62–63). This example illustrates the point that the location
of pits within a village relates to their usage. In addition, once
people decided to construct storage pits within a village, the only
alternative locations would be within or close to kraals/village
centres,orassociatedwithhuts. It is unlikelythatpeoplewould
have constructed storage pits within or on refuse middens,
paths, or outside villages. Another reason to store food under-
ground or within villages may be to offer protection from
animals such as monkeys, baboons, rodents and insects. The
presence of storage pits within villages – be it associated with
huts or within kraals – is not unusual, and therefore is not a
good diagnostic trait of Eastern Bantu-language speakers. The
location of pits in modern-day Angola and the DRC behind
villages (Huffman 1989: 158) may be recent adaptations (cf.
Lyons1996).Functional reasons areofferedasanalternative ex-
planation for the spatial arrangements of pits at settlements
that do not relate to an Eastern Bantu-language speaker
worldview.
BURIALS
Burials are not common in first millennium AD archaeolog-
ical sites (Mitchell 2002). The little evidence that does exist
indicates that both men and women were buried in kraals
during the Early Iron Age (Huffman 1990; Whitelaw 1993;
Mitchell 2002). The position of the body in graves (upright or
horizontal) does not seem, as yet at least, to be particularly
important in recognizing the CCP during the Iron Age. As both
males and females were buried in kraals during the Early Iron
Age (Whitelaw 1993; Mitchell 2002), grave data provide no
conclusiveevidencefortheCCPduringthefirstmillenniumAD.
DISCUSSION AND CONCLUSION
Inthispaper, I haveindicatedthatnoneofthefeaturesused
toidentifytheCCPduringtheEarlyandMiddleIronAges–the
period before the arrival of Nguni- and Sotho-Tswana speakers
– are diagnostic, or necessarily restricted to patrilineal, Eastern
Bantu-language speakers. My purpose in this paper has been to
present alternative explanations for the archaeological evidence
used to distinguish the CCP. In short that, first, circular huts
152 South African Archaeological Bulletin 64 (190): 148–155, 2009
South African Archaeological Bulletin 64 (190): 148–155, 2009 153
occur among some Western Bantu-language speakers, such as
in Angola. In many instances, the usage of rectangular huts is a
recent adaptation. Second, positioning livestock kraals in the
centre of villages is to protect animals against predators and
potential raids. In addition, it is important to consider the social
reasons behind the construction of circular villages. Third, the
location of storage pits can relate to the various functions of
these features, and a lack of alternative places to store grain and
wild fruits. Storage pits also occur at villages from west-central
Africa. Fourth, burial data do not support the CCP model.
Based on these arguments, I do not believe the CCP can be
applied to Early and Middle Iron Age farmers in southern
Africa.
I presented ethnographic and historical information from
west-central Africa countering the CCP. It is very likely that
many more examples exist, although not all ethnographic
accounts noted hut shapes and village layouts. Portuguese,
German and French sources must contain a wealth of informa-
tion on these cultural aspects in west-central and western
Africa. In addition, the extended influence of European and
Islamic cultures (e.g. Thornton 1984) resulted in the disappear-
ance of many cultural practices during the last 500 years.
Of course, not all areas were equally affected by European
expansion, and some regions remained largely unknown to
Europeans for a long time (e.g. Price 1872). Unfortunately, any
discussion of the settlement patterns of Eastern and Western
Bantu-speakers is seriously hampered by a lack of archaeological
data from countries like the DRC and Angola. Further research
is necessary to investigate all these issues.
Huffman (2001: 20) has made the point that different
worldviews do not produce the same settlement pattern.
However, groups moving from forested regions into savanna
environments facing predators such as lions would have to
adapt their settlement patterns to ensure optimal protection
for their livestock. In such a case, a different worldview will
produce a different settlement pattern to the norm, as people
had to adapt to new surroundings. Other factors most likely
had an impact on settlement patterns as well, for example,
short- and long-term and short- and long-distance migration
(Anthony 1990), as well as mobility and sedentism (Kelly 1992).
People in Sub-Saharan Africa did not have to ascribe to a CCP
worldview to live in circular huts and construct kraals in the
centre of a village.
It can be argued that functional aspects such as a centrally
positioned kraal do not contradict the CCP, since it can have
both practical and ideological functions. While this is true for
Eastern Bantu-language speakersinrecenttimes(e.g.Schapera
1953a), the mere presence of a central kraal on an Early or
Middle Iron Age site is not enough evidence to support the
CCP. This is because there are practical and social reasons, all
unrelated to the CCP, for constructing a central kraal in a
village. Of course, it is possible that a spatial feature such as a
central kraal in villages could become part of a worldview. On
the other hand, this may not happen.
Archaeologists in southern Africa studying the Early and
MiddleIronAgearenotlikelytofindanyevidencecontradicting
the CCP. This is not because the CCP can be extended and
appliedto these time periods. Instead, this is because the model
relies on functional aspects which will inevitably be present at
farming villages. In addition, protagonists of the CCP have not
taken note of the fact that other cultural behaviours mimic the
CCP, and they have not provided thorough archaeological
support for their views. The presence of spatial features which
clearly have strong functionality on Early and Middle Iron Age
sites cannot be taken as evidence for the CCP even when
ethnographic sources suggest that such features were deeply
embedded in worldviews in recent times. I therefore contest
that there is no unambiguous evidence to indicate that the CCP
dates to the Early and Middle Iron Age.
ACKNOWLEDGEMENTS
I asked two colleagues to comment on earlier drafts of this
paper. I would like to thank Ina Plug and Jan Boeyens from the
University of South Africa for their comments and suggestions.
Two reviewers provided additional suggestions. None of the
opinionsandoversightsmustbeattributedtothem,butremain
my own responsibility.
REFERENCES
Anthony, D.W. 1990. Migration in archaeology: the baby and the
bathwater. American Anthropologist 92: 895–914.
Atmore, A. & Sanders, P. 1971. Sotho arms and ammunition in the
nineteenth century. Journal of African History 12: 535–544.
Badenhorst, S. 2008. Subsistence change among farming communities
in southern Africa during the last two millennia: a search for
potential causes. In: Badenhorst, S., Mitchell, P. & Driver, J.C. (eds)
Animals and People: Archaeozoological Papers in Honour of Ina Plug:
215–228. Oxford: British Archaeological Series 1849.
Badenhorst, S. 2009. Phytoliths and livestock dung at Early Iron Age
sitesinsouthernAfrica.South African Archaeological Bulletin 64:45–50.
Badenhorst, S. In press. Descent of Iron Age farmers in southern Africa
during the last two thousand years. African Archaeological Review.
Beach, D.N. 1974. Ndebele raiders and Shona power. Journal of African
History 15: 633–651.
Beinart, W. 1998. The night of the jackal: sheep, pastures and predators
in the Cape. Past and Present 158: 172–206.
Bruwer, J. 1956. Die Bantoe van Suid-Afrika. Johannesburg: Afrikaanse
Pers-Boekhandel.
Burrows, G. 1899. On the natives of the Upper Welle district of the
Belgian Congo. Journal of the Anthropological Institute of Great Britain
and Ireland 28: 35–47.
Caldwell,M.G.&Sheldon,H. 1930. The cultureoftheBaya tribe of West
Africa. The Scientific Monthly 30: 320–325.
Clist, B. 2004–2005. Des premiers villages aux premiers Europeens
autour de l’estauire du Gabon. Quatre millenaires d’interactions
entre l’homme et son milieu. Unpublished DPhil. dissertation.
Brussels: Universite Libre du Bruxelles.
Clist,B.1987.EarlyBantusettlements in west-central Africa: areviewof
recent research. Current Anthropology 28: 380–382.
Clist, B. 1992. Interim report of the Oveng archaeological site 1991
excavations, Estuaire Province, Gabon. NSI 10/11: 28–35.
David, N. 1981. The archaeological background to Cameroonian
history. In: Tardits, C. (ed.) Contribution de la Recherche Ethnologique à
l’Histoire des Civilisations du Cameroun: 79–99. Colloques Inter-
nationaux Du Centre Nationale De La Recherche Scientifique, 551.
Paris: C.N.R.S.
Davison, P. 1988. The social use of domestic space in a Mpondo home-
stead. South African Archaeological Bulletin 43: 100–108.
de Maret, P. 1977. Sanga: new excavations, more data, and some related
problems. Journal of African History 18: 321–337.
de Maret, P. 1982. The Iron Age in the west and south. In: Van Noten, F.
(ed.) The Archaeology of Central Africa: 77–96. Graz: Akademische
Druk- und Verlagsanstalt.
Draper, P. 1976. Social and economic constrains on child life among the
!Kung. In: Lee, R.B. & DeVore, I. (eds) Kalahari Hunter-gatherers.
Studies of the !Kung San and their Neighbors: 200–217. Cambridge:
Harvard University Press.
Eggert, M.K.H., Höhn, A., Kahlheber, S., Meister, C., Neumann, K. &
Schweizer, A. 2006. Pits, graves and grains: archaeological and
archaeobotanical research in southern Cameroun. Journal of African
Archaeology 4: 273–298.
Fagan, B.M. 1969. Excavations at Ingombe Ilede 1960–2. In: Fagan, B.M.
(ed.) Iron Age Cultures in Zambia. Volume Two: 57–161. London: Chatto
& Windus.
Fitzpatrick, J.F.J. 1910. Some notes on the Kwolla district and its tribes.
Journal of the Royal African Society 10: 16–52.
Flannery, K.V. 2002. The origins of the village revisited: from nuclear to
extended households. American Antiquity 67: 417–433.
154 South African Archaeological Bulletin 64 (190): 148–155, 2009
Foyle, A.M. 1953. Some aspects of Nigerian architecture. Man 53: 1–3.
Greenfield,H.J.,Van Schalkwyk, L.O.& Jongsma, T.L. 2000.Surfaceand
subsurface reconnaissance at Ndondondwane: preliminary results
of the 1995–1997 field season. Southern African Field Archaeology 9:
5–16.
Greenfield, H.J. & Van Schalkwyk, L. 2003. Intra-settlement social and
economic organization of Early Iron Age farming communities in
southern Africa: a view from Ndondondwane. Azania 38: 121–137.
Greenfield, H.J. & Miller, D. 2004. Spatial patterning of Early Iron Age
metal production at Ndondondwane, South Africa: the question of
cultural continuity between the Early and Late Iron Ages. Journal of
Archaeological Science 31: 1511–1532.
Hall, M. 1984. The myth of the Zulu homestead: archaeology and
ethnography. Africa 54: 65–79.
Hall, M. 1986. The role of cattle in southern African agropastoral societ-
ies: more than bones alone can tell. South African Archaeological Society
Goodwin Series 5: 83–87.
Hall, M. 1987a. Archaeology and modes of production in pre-colonial
southern Africa. Journal of Southern African Studies 14: 1–17.
Hall, M. 1987b. The Changing Past: Farmers, Kings and Traders in Southern
Africa, 200–1860. Cape Town: David Philip.
Hall, S. 1997. Material culture and gender correlations. The view from
Mabotse in the late nineteenth century. In: Wadley, L. (ed.) Our
gendered past. Archaeological studies of gender in southern Africa: 209–219.
Johannesburg: Witwatersrand University Press.
Herskovits, M.J. 1930. The culture areas of Africa. Africa 3: 59–77.
Holden, C.J. & Mace, R. 2003. Spread of cattle led to the loss of
matrilineal descent in Africa: a coevolutionary analysis. Proceedings:
Biological Sciences 270(1532): 2425–2433.
Holl, A.F.C. & Levy, T.E. 1993. From the Nile Valley to the Chad Basin:
ethnoarchaeology of Shuwa Arab settlements. Biblical Archaeologist
56(4): 166–179.
Holl, A.F.C. 1994. The cemetery of Houlouf in northern Cameroon
(AD1500–1600):fragmentsof a past social system. AfricanArchaeolog-
ical Review 12: 133–170.
Huffman, T.N. 1982. Archaeology and ethnohistory of the African Iron
Age. Annual Review of Anthropology 11: 133–150.
Huffman, T.N. 1986. Iron Age settlement patterns and the origins of
class distinction in southern Africa. Advances in World Archaeology 5:
291–338.
Huffman, T.N. 1989. Ceramics, settlements and Late Iron Age migra-
tions. African Archaeological Review 7: 155–182.
Huffman, T.N. 1990. Broederstroom and the origins of cattle-keeping in
southern Africa. African Studies 49: 1–12.
Huffman, T.N. 1993. Broederstroom and the central cattle pattern.
South African Journal of Science 89: 220–226.
Huffman, T.N. 1997. Architecture and settlement patterns. In: Vogel,
J.O. (ed.) Encyclopedia of Precolonial Africa. Archaeology, History,
Languages, Cultures, and Environments: 149–155. London: AltaMira
Press.
Huffman, T.N. 1998. The antiquity of lobola. South African Archaeological
Bulletin 53: 57–62.
Huffman, T.N. 2001. The central cattle pattern and interpreting the
past. Southern African Humanities 13: 19–35.
Huffman, T.N. 2004. Beyond data: the aim and practice of archaeology.
South African Archaeological Bulletin 59: 66–69.
Huffman, T.N. 2007. Handbook to the Iron Age: The Archaeology of
Pre-colonial Farming Societies in Southern Africa. Pietermaritzburg:
University of KwaZulu-Natal Press.
Johnston, H.H. 1884. On the races of the Congo and the Portuguese
colonies in Western Africa. Journal of the Anthropological Institute of
Great Britain and Ireland 13: 461–479.
Kelly, R.L. 1992. Mobility/sedentism: concepts, archaeological measures,
and effects. Annual Review of Anthropology 21: 43–66.
Kingon, J.R.L. 1919. Cattle as a factor in the economic development of
South Africa. South African Journal of Science 15: 425–440.
Kuper, A. 1982. Wives for Cattle. Bridewealth and marriage in South Africa.
London: Routledge & Kegan Paul.
Laidler, P.W. 1936. The kraal and the hut of the Nama Hottentot of Little
Namaqualand. Man 36: 109–110.
Lane, P. 1994/1995. The use and abuse of ethnography in the study of
the southern African Iron Age. Azania 29/30: 51–64.
Livingstone, D. 1857. Explorations into the interior of Africa. Journal of
the Royal Geographical Society of London 27: 349–387.
Lyons, D. 1996. The politics of house shape: round vs rectilinear
domestic structures in Déla compounds, northern Cameroon. Antiq-
uity 70: 351–367.
Maggs, T.M.O ’C. 1980a. Mzonjani and the beginning of the Iron Age in
Natal. Annals of the Natal Museum 24: 71–96.
Maggs, T.M.O’C. 1980b. Msuluzi confluence: a seventh century Early
Iron Age site on the Tugela River. Annals of the Natal Museum 24:
111–145.
Maggs, T.M.O’C. 1984. The Iron Age south of the Zambezi. In: Klein,
R.G. (ed.) Southern African Prehistory and Palaeoenvironments: 329–395.
Rotterdam: Balkema.
Maggs, T.M.O’C. 1994/95. The Early Iron Age in the extreme south:
some patterns and problems. Azania 29/30: 171–178.
Maggs, T.M.O’C. & Ward, V. 1984. Early Iron Age sites in the Muden
area of Natal. Annals of the Natal Museum 26: 105–140.
Malcolm, L.W.G. 1923. Notes on birth, marriage and death ceremonies
of the Etap tribe, central Cameroon. Journal of the Royal Anthropological
Institute of Great Britain and Ireland 53: 388–401.
Marais, E.N. 1974. The Soul of the Ape. Pretoria: Human & Rousseau.
Mbida, C.M., Van Neer, W., Doutrelepont, H. & Vrydaghs, L. 2000.
Evidence for banana cultivation and animal husbandry during the
first millennium BC in the forest of southern Cameroon. Journal of
Archaeological Science 27: 151–162.
McIntosh, R.J. 1974. Archaeology and mud wall decay in a WestAfrican
village. World Archaeology 6(2): 154–171.
McIntosh, S.K. & McIntosh, R.J. 1986. Recent archaeological research
and dates from West Africa. Journal of African History 27: 413–442.
Miller, D. & Whitelaw, G. 1994. Early Iron Age metal working from the
site of KwaGandaganda, South Africa. South African Archaeological
Bulletin 49: 79–89.
Mitchell, P.J. 2002. The Archaeology of Southern Africa. Cambridge:
University of Cambridge Press.
Mitchell, P.J. & Whitelaw, G. 2005. The archaeology of southernmost
Africa from c. 2000 BP to the early 1800s: a review of recent research.
Journal of African History 46: 1–33.
Moloney, A. 1890. Notes on Yoruba and the Colony and Protectorate of
Lagos, West Africa. Proceedings of the Royal Geographical Society and
Monthly Record of Geography 12(10): 596–614.
Mönnig, H.O. 1967. The Pedi. Pretoria: J. L. van Schaik.
Morgan, E.D. 1884. Notes on the Lower Congo, from its mouth to
Stanley Pool. Proceedings of the Royal Geographic Society and Monthly
Record of Geography 6(4): 183–196.
Morris, A.G. 1993. Human remains from the Early Iron Age sites of
Nanda and KwaGandaganda, Mngeni Valley, Natal. Natal Museum
Journal of Humanities 5: 83–89.
Musonda, F.B. 1987. Preliminary report on archaeological investiga-
tions in the Sikalongo area, Choma district, Zambia. NSI 2: 40–44.
Oliver, R. 1982. The Nilotic contribution to Bantu Africa. Journal of Afri-
can History 23: 433–442.
Peter, B. 2001. Vitrified dung in archaeological contexts: an experimental
study on the process of its formation in the Mosu and Bobirwa areas.
Pula: Botswana Journal of African Studies 15: 125–143.
Pistorius, J.C.C. 1992. Molokwane: an Iron Age Bakwena Village. Johannes-
burg: Perskor.
Plug, I. 1996. Domestic animals during the Early Iron Age in southern
Africa. In: Pwiti, G. & Soper, R. (eds) Aspects of African Archaeology:
515–520. Harare: University of Zimbabwe Press.
Plug, I. 2000. Overview of Iron Age fauna from the Limpopo Valley.
South African Archaeological Society Goodwin Series 8: 117–126.
Plug, I. & Voigt, E.A. 1985. Archaeozoological studies of Iron Age
communities in southern Africa. In: Wendorf, F. & Close, A. (eds)
Advances in World Archaeology 4: 189–238. London: Academic Press.
Plug, I. & Badenhorst, S. In press. Ethnography and southern African
archaeozoology. In: Peters, J. (ed.) Ethnoarchaeozoology. Documenta
Archaeobiologiae 7. Verlag Marie Leidorf GmbH: Rahden/Westf.
Price, F.G.H. 1872. A description of the Quissama tribe. Journal of the
Anthropological Institute of Great Britain and Ireland 1: 185–193.
Rautenbach, I.L. 1982. Mammals of the Transvaal. Ecoplan Monograph
No. 1. Pretoria: Ecoplan.
Robinson, K.R. 1963. A note on storage pits: Rhodesian Iron Age and
modern African. South African Archaeological Bulletin 18: 62–63.
Ruel, M.J. 1962. Banyang settlements: part I: pre-European settlement.
Man 62: 99–103.
Schapera, I. (ed.) 1953a. The Bantu-speaking tribes of South Africa. An
ethnographic survey. London: Routledge & Kegan Paul.
South African Archaeological Bulletin 64 (190): 148–155, 2009 155
Schapera, I. 1953b. Cultural changes in tribal life. In: Schapera, I. (ed.)
The Bantu-speaking tribes of South Africa. An ethnographic survey:
357–387. London: Routledge & Kegan Paul.
Schildkrout, E. 1991. The spectacle of Africa through the lens of Herbert
Lang:BelgiumCongophotographs1909–1915. African Arts 24: 70–100.
Segobye, A. 1998. Daughters of cattle: the significance of herding in the
growth of complex societies in southern Africa between the tenth
and fifteenth centuries A.D. In: S. Kent (ed.) Gender in African
Prehistory: 227–233.Walnut Creek: AltaMira Press.
Skinner, J.D. & Smithers, R.H.N. 1990. The Mammals of the Southern
African Subregion. Pretoria: University of Pretoria.
Skovlin, J.M. 1971. Ranching in East Africa: a case study. Jour nal of Range
Management 24: 263–270.
Smithers, R.H.N. 1983. Die Soogdiere van die Suider-Afrikaanse Substreek.
Pretoria: University of Pretoria.
Thornton, J. 1984. The development of an African Catholic Church in
the Kingdom of Kongo, 1491–1750. The Journal of African History 25:
147–167.
Thurnwald, R. 1929. Social systems of Africa (Conclusion). Africa 2:
352–380.
Torday, E. 1910. The land and peoples of the Kasai Basin. The Geographical
Journal 36: 26–53.
Torday, E. 1921. Culture and environment: cultural differences among
the various branches of the Batetela. Journal of the Royal Anthropological
Institute of Great Britain and Ireland 51: 370–384.
Torday, E. & Joyce, T.A. 1907. On the ethnology of the southwestern
Congo Free State. The Journal of the Royal Anthropological Institute of
Great Britain and Ireland 37: 133–156.
Van Noten, F. 1982. The Archaeology of Central Africa. Graz: Akademische
Druk- und Verlagsanstalt.
Van Reenen, J.F. 1987. Tooth mutilations amongst the peoples of
Kavango and Bushmanland, South West Africa (Namibia). Journal of
the Dental Association of South Africa 33: 205–218.
Van Schalkwyk, J.A. 1987. A Late Iron Age smelting furnace south-east
of Pietersburg in the Transvaal, Republic of South Africa. South
African Archaeological Bulletin 42: 131–135.
Vansina, J. 1990. Paths in the forest. Madison: University of Wisconsin
Press.
Vogel, J.O. 1971a. Kumadzulo. An Early Iron Age village site in southern
Zambia. Zambia Museum Papers 3. London: Oxford University
Press.mm
Vogel, J.O. 1971b. Kamangoza. An introduction to the Iron Age cultures of the
Victoria Falls region. Zambia Museum Papers 2. London: Oxford
University Press.
Vogel, J.O. 1975. Simbusenga. The archaeology of the Intermediate Period of
the southern Zambian Iron Age. Zambia Museum Papers 4. London:
Oxford University Press.
Walton, J. 1958. Sotho cattle-kraals. South African Archaeological Bulletin
13: 133–143.
Whitelaw, G. 1993. Customs and settlement patterns in the first
millennium AD: evidence from Nanda, an Early Iron Age site in the
Mngeni Valley, Natal. Natal Museum Journal of Humanities 5: 47–81.
Whitelaw, G. 1994. KwaGandaganda: settlement patterns in the Natal
Early Iron Age. Natal Museum Journal of Humanities 6: 1–64.
Whitelaw, G. 2005. Comment on Greenfield and Van Schalkwyk’s
article on Ndondondwane, Azania, 2003. Azania 60: 122–127.
Whiting, J.W.M. & Ayres, B. 1968. Inferences from the shape of
dwellings. In: Chang, K. C. (ed.) Settlement Archaeology: 117–133. Palo
Alto: National Press Books.
Willey, G.R. 1968. Settlement archaeology: an appraisal. In: Chang, K.
C. (ed.) Settlement Archaeology: 208–226. Palo Alto: National Press
Books .
