Content uploaded by Peter Mitchell
Author content
All content in this area was uploaded by Peter Mitchell on Jun 18, 2018
Content may be subject to copyright.
149
Southern African Humanities 22: 149–70 September 2010 Natal Museum
http://www.sahumanities.org.za
Making history at Sehonghong: Soai and the last Bushman
occupants of his shelter
Peter Mitchell
School of Archaeology, University of Oxford; School of Geography, Archaeology &
Environmental Studies, University of the Witwatersrand.
St Hugh’s College, Oxford, OX2 6LE, United Kingdom;
peter.mitchell@st-hughs.ox.ac.uk
ABSTRACT
Sehonghong Shelter is a site of considerable importance to southern African archaeology by reason
of its rock paintings, a few of which were interpreted by a Bushman informant with knowledge of the
art in 1873, and because of its long sequence of Later and Middle Stone Age assemblages with good
organic preservation reaching back to 57 000 years ago. This paper briey summarises the history of
archaeological research at the site, but focuses on its historical importance as a centre for the last generations
of Bushmen to live in the Lesotho highlands. Historical accounts relating them to the site are discussed,
the events surrounding the death there of the Bushman leader Soai described and the implications of
recently published oral histories indicating a post-Soai persistence of Bushman occupation considered.
The paper ends by underlining the urgent necessity of further investigation of the site’s remaining, and
threatened, rock art and the importance of extending archaeological research to include the history of
the local Basotho communities.
KEY WORDS: Bushman history, nineteenth century, Sehonghong, Lesotho, Soai, oral history.
Recent publications (Lewis-Williams 2003; Mitchell & Smith 2009; Wright & Mazel
2007) and the re-issue of Patricia Vinnicombe’s (1976, 2009a) groundbreaking but
long-unobtainable synthesis People of the Eland have once again drawn attention to the
archaeological richness of the Maloti-Drakensberg mountains. A key reference point
within this region is the large rock shelter near the village of Sehonghong in eastern
Lesotho. Known locally as Lehaha-la-Sehonghong or Lehaha-la-Soai (Sehonghong Cave
or Soai’s Cave), it is important for at least four reasons (Fig. 1).
First, it is one of only three known rock art sites in the whole of southern Africa
for which we possess interpretations of some images given by someone for whom
the production of rock art was still a living tradition (the others are Melikane, 40 km
to the south, and Pitsaneng, 1 km upstream; Jolly 2006a, 2008; Smits 1973). As such,
the comments given by the Bushman Qing to Joseph Orpen (1874) in December
1873 retain enormous value as a guide to understanding Bushman rock art (Lewis-
Williams 2003). Second, Sehonghong preserves a long sequence of deposits spanning
several episodes of the Holocene and late Upper Pleistocene. This sequence, which
extends back to 57 000 years ago (Jacobs et al. 2008), also contains well-preserved
palaeobotanical and faunal remains with the potential to inform on both human
activities and palaeoenvironmental change over much of the late Quaternary. Third,
Sehonghong lies within a rich concentration of other Middle and Later Stone Age
sites, two of which (Pitsaneng and Likoaeng) have been extensively excavated (Hobart
2004; Mitchell 2009), and all of which provide a detailed local context within which it
can be understood (Mitchell 1996a). Fourth and nally, Sehonghong is known from
multiple sources to have been inhabited by Bushmen (Baroa in Sesotho) as recently
150 SOUTHERN AFRICAN HUMANITIES, VOL. 22, 2010
as the late nineteenth century and events there are still remembered as a key moment
in the Basotho settlement of the Maloti landscape at that time. Though scattered
and disparate, these data, in Pieter Jolly’s words (pers. comm. 2010) “give a face” to
Sehonghong’s last Bushman inhabitants that is lacking for almost all other rock shelters
in southern Africa.
Following Pat Vinnicombe’s call in the original manuscript of her posthumously
published account of local oral traditions (Vinnicombe 2009b), it thus seems appropriate
to draw together all that we know now about the history of the site and of the last
Bushmen to occupy it. To do this I rst briey review the archaeological work that has
been undertaken there. I then use the available historical sources, some documentary,
others oral testimonies recorded at various times over the past hundred years, to attempt
a reconstruction of the site’s history and that of its last hunter-gatherer occupants from
the mid-1800s to the beginning of the twentieth century. A critical moment concerns the
attack on the Bushman group led by Soai that took place there early in the 1870s.
DOING ARCHAEOLOGY AT SEHONGHONG
Sehonghong Shelter (26°28′S, 28°47′E) lies approximately 3 km upstream of the Senqu
River on the south bank of its tributary, the Sehonghong, and at an elevation of about
1800 m above mean sea level (Fig. 2). Geopolitically, it falls within the territory of
Khomo-ea-Mollo village in the Sehonghong area of Thaba Tseka District. The site
has been known to the scholarly community since Orpen (1874) published his account
Fig. 1. Sehonghong Shelter. The small size of the gure to the right of the photograph emphasises the overall
size of the site, which has a width of some 86 m and a maximum depth behind the dripline of about
19 m, making Sehonghong one of the largest inhabited rock shelters in the Maloti-Drakensberg
region. Excavations by Carter (1978), Mitchell (1996b) and Stewart (ongoing; http://www.amemsa.
com) have sampled only a tiny fraction of its total area.
MITCHELL: SEHONGHONG IN HISTORY 151
of the mythology of the Maloti Bushmen well over one hundred years ago. As will
become apparent, it received occasional notice from subsequent writers. However,
realisation that this was indeed the location of the famous rain-animal scene copied by
Orpen (1874) and interpreted rst by Qing and then by the |Xam Bushman Diä!kwain
(Bleek 1874) came only with the work of Webb (1950), compiler of a gazetteer of
Lesotho privately published soon after the Second World War. Conrmation followed
with Vinnicombe’s visit to the site in 1957 and her subsequent tracing of the scene
in question, rst published by Smits (1973), who himself visited Sehonghong shortly
before excavations began there in July 1971.
Those excavations were undertaken by Pat Carter along with Pat Vinnicombe as part
of their joint exploration of the archaeology and history of the Maloti-Drakensberg
region, an exploration that culminated in the publication of People of the Eland
(Vinnicombe 1976) and Carter’s (1978) doctoral thesis. Carried out between 14 July and
30 August 1971, their excavations reached bedrock in an 8 m2 trench in the northern
half of the site. They revealed a sequence of Middle and Later Stone Age assemblages
Fig. 2. Map of Lesotho locating Sehonghong and other sites mentioned in the text. Site names are abbreviated
thus: LIK Likoaeng; MLK Melikane; PIT Pitsaneng; SEH Sehonghong.
152 SOUTHERN AFRICAN HUMANITIES, VOL. 22, 2010
that nine conventional radiocarbon dates later showed reached back beyond 32 000 years
ago (Carter 1971a, 1978). Only summary details of these excavations were published
at the time (Carter & Vogel 1974), along with the possible implications of successive
roof-fall episodes for understanding the dating and severity of late Pleistocene cold
climatic phases (Carter 1976). Over a decade later, I undertook a more detailed study
of the stone artefact assemblages (Mitchell 1987), focusing on a late Pleistocene, early
microlithic assemblage very similar to the Robberg Industry previously described
from sites in the southern Cape (Deacon 1984). Results of this analysis included
ve additional conventional radiocarbon dates run on samples obtained in the 1971
excavation (Carter et al. 1988). All nds from Carter and Vinnicombe’s excavation were
subsequently accessioned to the collections of the University Museum of Archaeology
and Anthropology, University of Cambridge.
Pat Vinnicombe traced the vast majority of Sehonghong’s surviving paintings in
1971 and these tracings now form part of the Vinnicombe archive at the Rock Art
Research Institute (RARI), University of the Witwatersrand. Only the rain-animal scene
originally copied by Orpen (1874) was published (Vinnicombe 1976: 337). Later, in
1985, Lucas Smits, then of the National University of Lesotho, arranged for the site to
be recorded by his Analysis of the Rock Art of Lesotho (ARAL) project. Taole Tesele
photographed all surviving images at Sehonghong itself (Fig. 3; recorded as ARAL
658), the shallow overhang immediately above it (ARAL 665) and a small shelter in
front (ARAL 666), as well as at several other sites in the Sehonghong, Matebeng and
Fig. 3. Sehonghong: an example of the many eland and human images, some (in the bottom centre of the image)
clearly carrying bows and arrows, painted on the site’s rear wall. In this particular case, the clarity of
the images was articially enhanced by spraying them with water during the ARAL project’s survey
of the site’s rock art in 1985.
MITCHELL: SEHONGHONG IN HISTORY 153
Melikane areas of the Senqu Valley. Accompanying sketches and notes were also made.
No further systematic attempt at recording the site’s art, or assessing its deterioration
over time, has taken place since, although the shelter and some of its paintings—and
grafti—were recently lmed and photographed by Dr David Pearce of RARI (pers.
comm. 2010).
Carter and Vinnicombe’s excavations demonstrated Sehonghong’s potential
for addressing several questions of interest in southern African hunter-gatherer
archaeology, including the opportunity of investigating cultural and environmental
change along a trajectory extending before, during and after the Last Glacial Maximum
(Mitchell 1996b). However, interpretation of their excavated data is constrained by
the choice as excavation technique of 10 cm thick horizontal spits that crosscut the
site’s complex natural stratigraphy. For these reasons, I undertook further excavations
at Sehonghong between 10 July and 4 September 1992, opening up a 13 m2 trench
to an approximate depth of 1.10 m below the modern surface (Fig. 4). Located 2 m
south of, and parallel to, the earlier excavation, this sampled the entirety of the site’s
Later Stone Age assemblages, including those attributable to the transition between
Middle and Later Stone Age lithic technologies. Publication of preliminary results
(Mitchell 1993, 1996b; Mitchell & Vogel 1994) was followed by more detailed reports
on the artefact assemblages found (Mitchell 1994, 1995, 1996c, 1996d). The faunal
samples have also recently been published (Plug & Mitchell 2008a, 2008b), while I
considered temporal patterning in the site’s ostrich eggshell bead and marine shell
ornaments within a wider regional context (Mitchell 1996e). Regrettably, the rich
charcoal assemblages recovered in 1992 remain unstudied. They are currently housed
at the Institute of Archaeology, University of Oxford, the fauna at the Department of
Archaeozoology, Transvaal Museum, Pretoria, and all the artefacts at the Department
Fig. 4. Sehonghong: the 1992 excavation trench at the close of eldwork exposed to the base of the RFS Layer,
which is dated to c. 26 000 BP. Excavations by Brian Stewart (University of Cambridge) in 2009 have
since extended these excavations down into the underlying Middle Stone Age horizons.
154 SOUTHERN AFRICAN HUMANITIES, VOL. 22, 2010
of Archaeology, University of Cape Town, at which I was employed at the time of
the 1992 excavation.
A partial photographic record of Sehonghong’s paintings was made during the
1992 excavation and on subsequent visits in 1995, 1998, 2006 and 2007. On the
latter occasion, Carter and Vinnicombe’s original excavation trench was opened
up and exposed to bedrock to allow samples to be taken for single-grain optically
stimulated luminescence (OSL) dating. Initial results establish that human occupation
of Sehonghong began some 57 000 years ago (Jacobs et al. 2008). Further OSL dates
from more recent phases of the site’s history and charcoal samples taken for potential
ABOX radiocarbon dating have yet to be published, but a preliminary report using
IRSL dating of sediment samples also taken in 2007 produced a broadly consistent
result (c. 63 ka) for the earliest deposits (Barré & Lamothe 2009). Demonstration of
the Sehonghong sequence’s true antiquity, which extends well beyond the limits of
the radiocarbon technique available to Carter (1978) nearly four decades ago, came
just as interest was renewed in the Pleistocene settlement history of the Lesotho
highlands. Following excavations at Melikane Shelter in 2008 and 2009, Dr Brian
Stewart (McDonald Institute for Archaeological Research, University of Cambridge)
and Dr Genevieve Dewar (Department of Anthropology, University of Toronto)
began work at Sehonghong in July 2009 (http://www.amemsa.com). Their excavations,
which continue within the trench I opened up in 1992, will, for the rst time, provide
stratigraphically contextualised artefact assemblages and palaeoenvironmental samples
from the Middle Stone Age part of the site’s sequence.
SOAI, SEHONGHONG AND THE BUSHMEN OF THE LESOTHO HIGHLANDS
People had thus inhabited Sehonghong for thousands of years before the rst written
record of the region appeared in the context of the accelerating competition for land
and resources that marked the nineteenth-century history of the Maloti-Drakensberg
region. That record forms part of an extensive archive of reports, some ofcial,
others not, that cast light on the relationships between the “Bushman raiders of the
Drakensberg”, as Wright (1971) called them, and the Bantu-speaking and European
farmers settled below the uKhahlamba-Drakensberg escarpment in what is now
KwaZulu-Natal. Drawing on this and on the extensive earlier work of Vinnicombe
(1976), Campbell (1987), Dowson (1994, 1995, 1999), Prins (1990), Jolly (1994, 1996)
and others, Blundell (2004) extends the scope of these connections, often peaceful but
sometimes violent, into the Barkly East/Maclear areas of the Eastern Cape Province,
while near-contemporary reports by Stow (1905) and others, as well as recent syntheses
like that of Gill (1993), show how they also characterised events along Lesotho’s western
border with the Orange Free State.
For the Senqu Valley, the earliest such record relates to a commando sent by Mdwebo,
leader of a Bushman band in the East Griqualand area, against the Thola (AmaTola),
a creolised group with signicant Bushman membership, living either side of the
uKhahlamba-Drakensberg escarpment (Challis 2008). Setting out in May 1850 from
Matatiele, they appear to have entered Lesotho via Qacha’s Nek, moving north by
“crossing and re-crossing the mountains till the river was no larger than the Keiskamma
at Line Drift” (Natal Mercury 23 August 1850; Wright 2007: 128). Before returning to
Natal, they attacked at least two AmaTola groups and noted numerous kraals stocked
MITCHELL: SEHONGHONG IN HISTORY 155
with the horses, cattle, sheep and goats that they were rearing and other animals that
they had captured on raids. Geographical details are necessarily imprecise, but the fact
that one of the groups attacked lay “somewhere opposite the source of the Mzimkhulu”
River (Vinnicombe 1976: 62–3; Wright 1971: 128) suggests they reached the upper
reaches of the Tsoelike River (Vinnicombe 1976: 75, note 40) or, if sticking to the main
Senqu Valley, perhaps even as far as the Sehonghong River. Of wider signicance, this
report underlines the degree to which by this time Bushmen, or groups partly composed
of Bushmen, had successfully integrated herding and raiding into a hunter-gatherer
way of life and suggests that at least some of the stone walling found in rock shelters
in highland Lesotho may be their work, rather than that of later Basotho herdsmen.
The same could hold for some of the domestic fauna found in the uppermost layers at
Sehonghong (Plug & Mitchell 2008b) and at nearby sites like Pitsaneng (Hobart 2004),
although the direct dating of sheep remains at Likoaeng to the late rst millennium
AD demonstrates that ‘hunters-with-livestock’ (sensu Sadr 2003) were by no means a
nineteenth-century innovation in the region (Mitchell et al. 2008).
In 1862 another punitive expedition entered the Lesotho highlands, this time from the
north, but though encountering Bushmen, taking one prisoner and capturing horses, it
does not seem to have reached as far south as Sehonghong (Vinnicombe 1976: 82–3).
Six years later, however, the Nguni chief Sakhayedwa led a party across the escarpment,
perhaps by way of the Mashai Pass, as far as the Senqu, and even a short way beyond,
undoubtedly coming much closer to the site (Vinnicombe 1976: 85). In response to a
further Bushman raid against farmers in Natal, the following year the British colonial
authorities organised a much more substantial expedition, commanded by Captain
Albert Allison. While covering an extensive stretch of highland Lesotho, neither Allison,
nor James Giles in command of a second military detachment, appears, however, to
have passed close to Sehonghong. Instead, they struck west from the Mokhotlong area
toward the Senqunyane (where they attacked and killed the majority of a Bushman
group numbering around thirty) and then advanced south along the Senqu, nally
descending the escarpment at Qacha’s Nek (Vinnicombe 1976: 87–92).
Quite erroneously, Allison’s raid has been implicated in the death of Soai, after whom
Sehonghong Shelter is named (e.g. Stow 1905: 229–30; cf. Wright 1971: 174, note 54).
However, as Vinnicombe (1976: 94) makes plain, not only is the location totally wrong,
but Allison was specically told that the leader of the Bushmen whom he attacked
near the Senqunyane was called Masharka, not Soai. The account provided by Basotho
historian Azariel Sekese to Victor Ellenberger (1953: 253–5), while matching many
of the details given by Allison, explicitly avoids mentioning Soai by name and thus
provides no justication for connecting the two.
Soai himself gained legendary status as stories of the last Bushmen of the Maloti-
Drakensberg mountains were told and retold. Vinnicombe (1976: 101) cites at least three
sites said to have been used by him on the South African side of the escarpment, but
only in the case of Mzimuti Shelter near Underberg is this based on a relatively early
source, information obtained by A.D. Whyte when surveying the Drakensberg for rock
paintings in 1910. Driven from there, Soai supposedly took refuge in Lesotho, with
Dornan (1909: 449) specically linking him to Sehonghong Shelter, a link supported by
others, including Webb (1950: 67, 282), who notes the site’s name as ‘Lehaha-la-Soai’
(Soai’s Cave) and ‘Lehaheng ha Soai’ (The Place of the Cave of the Village of Soai),
156 SOUTHERN AFRICAN HUMANITIES, VOL. 22, 2010
and Ellenberger (1953: facing 160). Soai’s afliation with other Bushman ‘chiefs’ is
variously recorded, by Dornan (1909: 439), citing unknown informants, as the son of
Melikane, but by one Lébousa Tsémané as reported by Ellenberger (1953: 258) as the
son of Khôbôkô, whom Dornan (1909: 439–40) perhaps locates in East Griqualand
under the name ‘Kholoboto’. More secure is the repeated connection between Soai and
the Phuthi leader Moorosi, whose chiefdom occupied the Quthing area of southern
Lesotho in the mid-1800s; indeed, Soai is said to have had a lover among the Phuthi
(Damane & Sanders 1974: 182).
Moorosi’s links with various Bushman groups are well attested, included extensive
intermarriage (Jolly 1996: 60) and certainly extended to sharing in the proceeds of raids,
as Allison himself suspected (Jolly 1996; Vinnicombe 1976: 92). They are conrmed
by the testimony of an elderly Basotho woman, Elisabetha ‘Malékètanyané Môhanoè,
interviewed by Ellenberger in the late 1920s/early 1930s. Born in 1856, she was part
of a group led by Moorosi that visited Soai and his followers at Sehonghong when she
was sixteen, which would thus place the visit in 1872. As I have discussed elsewhere
(Mitchell 2006–07), too much precision ought not to be attached to this date, but a visit
as late as the beginning of the 1870s is perfectly plausible. Among the many interesting
recollections that she recalled, ‘Me Môhanoè remembered seeing women make pottery
and men execute paintings—sadly undescribed—on the walls of the shelter (Ellenberger
1953: 86, 148–9). Much of the detail that she provides is supported by other sources
and there seems no reason to dispute the overall veracity of her account (Mitchell
2006–07), not least given the evident freshness of some of the paintings later observed
by James Murray Grant (Mitchell & Challis 2008: 434) and Sello Mokhoallo of the
nearby village of Khoma-ea-Mollo (Vinnicombe 2009b: 168).
One reason for supposing that ‘Me Môhanoè’s visit to Sehonghong took place a
little before 1872 is that both Sir Marshall Clarke (1888: 519), Resident Commissioner
in Basutoland 1884–93, and Rev. Edouard Jacottet (1893: 511) of the Lesotho
Evangelical Church, explicitly place the death of Soai in 1871 (see also reproductions
of their accounts in Germond 1967: 418, 425, 428). The likelihood that this is correct
is enhanced by the fact that Clarke (1888: 520) was accompanied on his trip by
“representatives of … Joel and Jonathan Molapo”, the chiefs responsible for Soai’s
demise. That Rev. Jobo Moteane also participated in Clarke’s expedition (Ambrose &
Sekoli 1991) provides at least one context in which he may have acquired the detailed
knowledge of Soai’s death that he provided to Dornan some time before November
1907 (1909: 449–50). Additional information comes from Azariel Sekese, Jonathan
Molapo’s one-time secretary, who published accounts of the relevant events in Sesotho
early in the twentieth century (Sekese 1912a, 1912b, 1924), another of Ellenberger’s
(1953: 255–7) informants, Filémone Ratèboho Matlénané, and praise songs (lithoko)
relating the exploits of both Molapo brothers. These praise poems were originally
published—rst in Leselinyana la Lesotho and then in book form—by Mangoaela (1921)
and, in the case of those relating to Jonathan, were almost certainly written down by
Sekese; the contents of the relevant poems were later explicated by close relatives of
Joel and Jonathan, including Chief ‘Mako Moliboea Molapo, who was brought up by
Jonathan, and one of Joel’s sons, Chief Lepokola Joel (Damane & Sanders 1974).
Jolly (1994: 67–8) has claried the events concerned, which were preceded by more
than one raid by Soai that took horses and cattle (Damane & Sanders 1974: 203) from
MITCHELL: SEHONGHONG IN HISTORY 157
followers of the Basotho king, Moshoeshoe I, and more than one attack on Soai by
Moshoeshoe’s grandsons, Joel and Jonathan Molapo. The respective roles played by the
two brothers are a little confused in the sources (Table 1). However, the overall sequence
is clear and Soai’s death itself is likely to have been the work of a party led by Joel Molapo,
given Sekese’s proximity to those principally involved, the relatively late date of Moteane’s
account (Dornan 1909) and the downgrading within it of the role played by Joel relative
to the much earlier comments—almost certainly acquired from Moteane—made by
Jacottet in 1893: “During the three days I spent with Mr. Job Moteane, at the station
of Sehonghong, I visited the famous cave … where Soai the last Bushman chief, was
massacred by Joel and Jonathan Molapo in 1871” (Germond 1967: 425).
Events after Azariel Sekese
Soai raids horses from the Basotho.
Jonathan Molapo asked to retrieve stolen horses.
Jonathan attacks Soai’s band, but Soai is away.
Jonathan takes prisoners back to Leribe.
Jonathan meets up with Joel Molapo.
Joel tracks Bushmen to Sehonghong.
Soai discovered and killed while hiding in a pool.
Soai’s belt is removed and given to Molapo.
Events after Jobo Moteane
Joel Molapo attacks Bushmen in response to raids.
Jonathan Molapo attacks Soai after further raids.
Soai and others chased up the Senqu into reeds.
Soai discovered and killed while hiding in a pool.
Soai’s belt is described as ‘beautiful’. His body is cut
up.
Women and children taken prisoner to Leribe, but
some escape.
TABLE 1
Comparison of the death of Soai as recounted by Sekese (1912a, 1912b, 1924) and Moteane (Dornan 1909).
First, one or more attacks on Soai’s band failed to kill him as he was away visiting
Moorosi at the time (Damane & Sanders 1974: 182, 185), although it is possible that
Soai himself was wounded when attacked by Joel Molapo (Damane & Sanders 1974:
206). Later, the pursuit was continued as far as Sehonghong Shelter (Fig. 5). There, the
approaches to the site were closed off and a small group of Bushmen chased along
the Senqu, upstream of its conuence with the Sehonghong (Fig. 6). Soai attempted to
conceal himself in a pool, breathing through a reed, but was detected by the shiny red
ochre on part of his body or when he came closer to the surface to breathe (Damane
& Sanders 1974: 207). Shot by an individual variously named as Mochebalele, son
of Tlaba (Damane & Sanders 1974: 207), or Lekosa Pokane (Sekese 1924), he was
dragged out. According to Moteane (Dornan 1909: 450), his body was then cut up for
‘medicine’. Most of the men with him were killed. The accounts given by Ratèboho
Matlénané (Ellenberger 1953: 255–6) and Rev. Moteane (Dornan 1909: 450), along
with information obtained by Damane & Sanders (1974: 183), all agree that some of
the women and children with Soai were captured and taken to Leribe, where they were
separated. Following two attempts, at least two women eventually escaped, a third being
killed and a fourth dying of illness, as did one of the children. The surviving girl (‘Qééa)
was given to a man named Nosi from Tsikoane, who later married her, while the two
158 SOUTHERN AFRICAN HUMANITIES, VOL. 22, 2010
Fig. 6. View of the Senqu Valley immediately upstream of its conuence with the Sehonghong River, presumably
close to the location at which Soai was killed according to the accounts transmitted by Sekese (1912a,
1912b, 1924) and Moteane (Dornan 1909).
Fig. 5. View of the Sehonghong Valley looking west toward the Senqu River and the Central Range of the Maloti.
Several surface scatters of Middle Stone Age artefacts, one perhaps linked to a collapsed rock shelter,
occur to the left of the Sehonghong River in the centre of the photograph. It was presumably along
this valley that Soai’s attackers came in 1871.
boys (P’hafôdi and ‘Qh’ôkh’ôôéa) worked as herdboys for Jonathan Molapo before
dying young; one of Jonathan’s praise poems conrms the identity and fate of Phafoli
and notes that one of the female captives taken by Jonathan during a raid on Soai and
his group was given to Jonathan’s mother (Damane & Sanders 1974: 182).
MITCHELL: SEHONGHONG IN HISTORY 159
To these details we can now add information obtained by Vinnicombe (2009b), who
interviewed two nonagenarian residents of Khomo-ea-Mollo, the village nearest to
Sehonghong Shelter, in 1971 (Fig. 7). Both informants were born during the Gun War of
1880–81 between the Basotho and Cape and imperial forces. Ntate Liselo Rankoli moved
with his family to Khomo-ea-Mollo about 1888, while Ntate Sello Mokoallo was born
there. Ntate Rankoli attributed Soai’s death to warriors of Jonathan Molapo, but stated
that Soai was killed while collecting honey in a crevice about a kilometre downstream
of Sehonghong Shelter, a location that he showed to Vinnicombe (2009b: 173–4, 179)
and at which bees and Bushman paintings were still visible in 1971. Though attributed
to a Bushman chief named Sehonghong, precisely the same story was recounted (and
dated to “about the beginning of 1873”) to a Captain James Smith of the Basutoland
Mounted Police by two of Jonathan’s councillors (How 1962: 16). Ellenberger (1953:
99–100) transmits a description of the same events by M. Makhetha, but with respect
to a chief named Melikane! As Vinnicombe (2009b: 179, note 4) comments, unravelling
precisely what happened is probably no longer possible, while an additional complication
may have been introduced into the oral accounts by the fact that highland Lesotho has
not one, but two, rivers named Sehonghong, the one passing by the rock shelter at the
centre of this paper and another, more northerly river, that rises a little to the north of
Sani Pass and eventually contributes to the Khubelu, one of the major sources of the
Senqu (Pieter Jolly pers. comm. 2010). However, the basic fact of the Basotho attack
on Soai and his people at Sehonghong Shelter around 1871 seems indisputable, with
the additional possibility of a follow-up attack—by Jonathan alone—shortly thereafter
against Sehonghong in the same or another area of the highlands.
Fig. 7. Khomo-ea-Mollo, the closest village to Sehonghong Shelter and home in 1971 to Ntate Liselo Rankoli
and Ntate Sello Mokoallo, both of whom recalled seeing Bushmen living in the shelter at the end of
the nineteenth/beginning of the twentieth centuries (Vinnicombe 2009b). Courtesy and copyright
Brian Stewart.
160 SOUTHERN AFRICAN HUMANITIES, VOL. 22, 2010
SEHONGHONG AFTER SOAI
Not all of Soai’s followers were shot or captured outright. If we assume an identity
between Soai and Smith’s ‘Sehonghong’, whose destruction had previously been urged
by Lesotho’s rst British administrator, Colonel James Bowker (How 1962: 58), then
his son and another man managed to reach the Leqoa River before being cornered
and killed (How 1962: 16). Echoes of this—perhaps indicating that not all met this
fate—may be found in Sello Mokoallo’s comment that some Basotho living along the
Leqoa married Bushman wives (Vinnicombe 2009b: 168) and other oral traditions that
Bushman survived there until 1886 (Vinnicombe 1976: 103). According to Elisabetha
Môhanoè, other survivors sought refuge with Moorosi’s Phuthi or in the Free State
(Ellenberger 1953: 258). The former possibility is supported by an oral account given
to Vinnicombe (1976: 101) by a former Qacha’s Nek trader, C.J. Laird, who recalled
meeting a Phuthi man who claimed to have adopted two children he had found in
Sehonghong Shelter after Jonathan’s attack. It is, of course, also tempting, if unprovable,
to follow How (1962: 22) in drawing a connection with Orpen’s (1874: 2) statement
that Qing, who showed him paintings at Sehonghong in December 1873, was the son
of the chief of a group of Bushmen otherwise largely exterminated in the Maloti “a
couple of years ago”.
Regardless of his parentage, Qing certainly knew the Upper Senqu Valley well
and was sought out by Grant and Orpen for that reason when they needed a guide
through the region as part of operations against the Hlubi chief Langalibalele in
December 1873. Grant’s account of their expedition makes clear, however, that at
least part of the area they traversed had already been scouted out by Basotho from
the western side of Lesotho before the Molapo brothers’ nal attack on Soai. He
mentions, for example, that his advance party “saw the spoor made by Molappo’s
[sic] cattle when driven into the mountains during the Boer war”, that is, the Seqiti
War between Lesotho and the Orange Free State of 1865–68 (Mitchell & Challis
2008: 431). More interestingly still, Nehemiah Moshoeshoe, a son of Moshoeshoe I,
who also helped guide the expedition, told Grant of “some large caves and a good
deal of pasturage” between two afuents of the Senqu known as the P(h)ofung
and the Bocheletsane to which his father had planned to retreat as a nal resort in
the event of defeat in the same war. Today a village named Bocheletsane lies on the
Litsoetse River, which enters the Senqu a few kilometres north of the Sehonghong.
Webb (1950: 8) conrms this identication, while the fact that the sandstones of
the Clarens formation run out a little to the north near Mashai implies that the ‘large
caves’ in question may well have included Sehonghong itself, by far the biggest rock
shelter in this section of the Senqu Valley (Mitchell & Challis 2008: 415, 456, note
44). My own observations of the lower reaches of the Litsoetse, like those of Pat
Vinnicombe who recorded rock art here in 1971 (Carter 1971b), certainly indicate
that nothing remotely describable as ‘large caves’ is likely to exist along the Litsoetse
itself, or north of Sehonghong Shelter.
That shelter emerges into the historical record for certain in the entry made by
Grant in his diary for 16 December 1873. Having camped very close to the site the
previous night in the valley of what he termed the Magwali River, Grant went to view
the paintings there and was suitably impressed by what he saw, writing:
MITCHELL: SEHONGHONG IN HISTORY 161
The paintings, many of them capitally done—a Hartebeeste, baboon, and Eland that I saw
were quite artistic. The colours were most brilliant, and stones which had been used as palettes
were lying about (Mitchell & Challis 2008: 434).
The last reference recalls ‘Me Môhanoè’s recollection of seeing paintings being made
during her visit to the site less than three years before and underlines the continuity of
the tradition on which Qing drew when explicating the Sehonghong rain-animal scene
to Orpen (Fig. 8), probably that very same day (Mitchell & Challis 2008: 403). Sadly,
Orpen either did not record, or decided not to publish, any further details of the site
that he called ‘Mangolong’, meaning, in Sesotho, ‘The place of letters’, presumably a
reference to the paintings on its walls (Eric Theko pers. comm. 1998). Neither he, nor
Fig. 8. The rain-animal scene at Sehonghong copied by Orpen (1874). The original (top) is now extremely
faded and difcult to see, but its details can be grasped from Pat Vinnicombe’s colour rendition
of the scene (bottom; courtesy of the Natal Museum).
162 SOUTHERN AFRICAN HUMANITIES, VOL. 22, 2010
Grant, however, made any comment at all about the attack on Soai, which suggests that
no evidence of this, or of the site’s earlier occupation, was visible at the time of their
visits. One wonders if survivors of the attack had removed any usable items, survivors
who clearly kept well out of the expedition’s way since Grant’s diary makes no mention
of any inhabitants in the region it traversed (Mitchell & Challis 2008: 404).
That situation changed markedly within a decade (Gill 1993). In 1878 the Basotho
paramount Letsie I sent one of his younger sons, Tlhakanelo, to the far side of the
Senqu to assert control over that part of Lesotho. Basing himself at the core of the
modern village of Sehonghong, Ha Tlhakanelo was the rst Basotho settlement east
of the main Senqu channel. Fuelled by a growing population now conned within
Lesotho’s present boundaries because of the earlier loss of the eastern Free State and
increased further by refugees from the Eastern Cape, Basotho presence in the Maloti
grew swiftly in response to the ghting of the Gun War (1880–81); the life history of
Liselo Rankoli, one of Vinnicombe’s (2009b) elderly informants, is a case in point.
By the time that Clarke (1888) reached Sehonghong in October 1887, villages were
numerous across Lesotho’s highlands. Like that of Grant and Orpen, his expedition,
which aimed at reconnoitering the territory for which he was responsible, found no
trace of any Bushmen, although he noted that the Molapo brothers’ expedition(s) had
found evidence of them growing tobacco, consistent with the origin of the name of
the Likoaeng stream (‘The place of the tobacco plants’) a few kilometres upstream of
Sehonghong and across the Senqu (Mitchell 2001; David Ambrose pers. comm.). Clarke
(1888: 523) does, however, provide a short, but appreciative description of Sehonghong
and its paintings, among which eland, hippopotamus (Orpen’s rain animals?), archers,
horses and cattle ‘raids’ (Fig. 9) are all still identiable today:
It is a simple overhanging rock, the wall in rear being covered with pictures of hunting scenes,
war dances, predatory expeditions, and various wild animals. Eland, hippopotamus, and the
smaller buck are all recognisable, while occasionally is depicted the uncouth form of the Rain-
god. In all the ghting pictures the Bushman is shown victorious. He is drawing his bow with
tiny hands, or balancing himself on shapely feet, throwing the assegai. His foes, on the other
hand, are exhibited with disproportionately big hands, eeing on caless legs stuck like broom
handles into the middle of their feet, and in the rear appear Bushwomen and boys driving herds
of horses and cattle, the spoils of victory.
Jobo Moteane, who accompanied Clarke, makes no mention of the site, merely
conrming the brevity of their stopover at Ha Tlhakanelo (Ambrose & Sekoli 1991:
5–6). Within ve years, however, he had returned to set up the rst mission station of
the Lesotho Evangelical Church east of the Senqu (Ambrose & Sekoli 1991: 3), and it
was there that Jacottet stayed with him in 1893 (Germond 1967: 425).
Neither Clarke nor Jacottet, who likewise visited Sehonghong Shelter, nor
Thomas Kennan (1959: 44), who rushed through Ha Tlhakanelo without doing so
on 8 September 1888, mentions any continued occupation of the site. However,
Vinnicombe’s (2009b) recently published oral histories make plain that the death of
Soai did not mark the end of its inhabitation by Bushmen. Both Sello Mokoallo and
Liselo Rankoli vividly recalled their presence there when they were children in the late
1880s/1890s and specically remembered several individuals: Ou Jan, described as an
old man; Rasethla (a medicine man); Baroli; two women named Serope and Qamoko,
the latter of whom had a son, Sekoqo; and another man, Mutsapi. Ntate Rankoli’s
admittedly vague reference to “women” and “their children” (Vinnicombe 2009b: 174)
MITCHELL: SEHONGHONG IN HISTORY 163
implies the presence of still others, and this is also likely from his description of an
eland hunt that he had personally witnessed in the immediate vicinity of the site and
that must have demanded the participation of several hunters (Vinnicombe 2009b:
176–7). Much of the other information these two elderly men provided, including
references to hunting, shing, gathering, shelter construction and tool use, likewise
seems to derive from their own observations. Their references to the Bushmen moving
about the landscape (Vinnicombe 2009b: 167, 172) further imply that Sehonghong was
not the only rock shelter then still in use.
Archaeological nds that might corroborate these data are effectively lacking,
although it is not implausible that some of the artefacts (including grindstones and an
iron arrowhead) recorded on the surface of Sehonghong Shelter by Smits (1973: 33)
could have been used as recently as this. Ntate Rankoli’s statement that the Bushmen
made pottery (Vinnicombe 2009b: 176) is also relevant as Collis (2010) has recently
OSL-dated a sherd from my Layer DC to AD 1910 ± 10; indistinguishable from the
other Later Stone Age ceramics at the site, there is no particular reason to suspect that
the date is false because of possible post-depositional ‘bleaching’ of its luminescence
signal. As previously indicated, some of the domestic livestock remains found in the
same layer could also relate to this phase of the site’s history, if not to more recent
Basotho activity there, which has included driving in livestock in poor weather. Later
nineteenth/early twentieth-century glass beads (Marilee Wood pers. comm.), caprine,
cattle and pig remains (Ina Plug pers. comm.), iron, and potsherds of likely Sotho
manufacture in deposits at two small shelters (2928DB33 and 2928DB34) across the
Fig. 9. Sehonghong: one of two panels showing armed men driving off cattle. In both, the men are shown
armed with spears and shields of an oval type resembling those associated historically with southern
Nguni-speaking warriors. Similarly armed and depicted individuals, also painted in white, occur at
a nearby rock art site (Challis et al. 2008), as well as at others in southeastern Lesotho and adjacent
areas of South Africa (Vinnicombe 1976).
164 SOUTHERN AFRICAN HUMANITIES, VOL. 22, 2010
Senqu from the Sehonghong River could also be connected to this post-Soai Bushman
presence; neither site has stone walling, neither seems large enough (44 m2 and 12 m2
respectively) to have been especially useful for keeping livestock and, today at least,
neither is immediately adjacent to cultivated elds (Fig. 10). The possibility that they
were occupied by residual hunter-gatherers at the very end of the 1800s/start of the
1900s should therefore be entertained.
How long did such communities persist? On a declining basis, and with increasing
assimilation into Bantu-speaking societies around them, the answer to this question
must now be an emphatic ‘several decades at least’. Information collated by Jolly
(1986), Lewis-Williams (1986), Prins (1990, 1994) and Blundell (2004) in the Eastern
Cape, Jolly (1994) in southern Lesotho and perhaps Prins (2009; but see Francis 2006)
in KwaZulu-Natal adds to the data already known to Vinnicombe (1976: 103–7) over
thirty years ago. Putting to one side the human tragedies of lives disrupted, uprooted
and prematurely ended, of languages and ways of life destroyed, the loss to history,
archaeology and anthropology is also to be regretted since, sadly, the elderly individuals
interviewed by Vinnicombe (2009b), Prins (2009) and Jolly (1986) must have been the tip
Fig. 10. View of the archaeological site 2928DB33 from the bottom of its talus. This is one of two small rock
shelters close to Sehonghong at which bones of domesticated animals and nds of iron objects and
Basotho pottery hint at use by Bushmen after Basotho settlement began in the area in 1878.
MITCHELL: SEHONGHONG IN HISTORY 165
of an iceberg of valuable information that ought to have been recorded decades earlier.
Instead, and as far as Sehonghong itself is concerned, we are left with a brief—and not
necessarily convincing—mention of its many paintings by Dornan (1909: 445–6) and
Ellenberger’s (1953: 247) reference to tracing “in the cave of Soai, precisely, paintings
of Bushmen representing galloping horses” (my translation); the horses remain visible
today and two panels of horsemen hunting eland were traced by Vinnicombe.
By the time of these visits no Bushmen lived at Sehonghong Shelter. Liselo Rankoli
mentioned that “they dispersed much later” than when he rst knew them as a young
boy, but that this had happened before he turned 22 (c. 1902) (Vinnicombe 2009b:
174, cf. 173). Until that time the man known as Ou Jan worked as a herdsman for
Chief Tlhakanelo, while comments made by Sello Mokoallo raise the possibility
that Tlhakanelo himself was responsible for “dispersing” the last Bushmen from
Sehonghong after they had wounded a Basotho boy in the eye (Vinnicombe 2009b: 167,
cf. 179). Given that neither Grant nor Orpen made any mention of human remains
lying around the surface of the site in 1873, perhaps it was this event that produced the
“Bushman skulls” collected from Sehonghong and taken to the Mashai trading store
as reported by Ntate Rankoli (Vinnicombe 2009b: 173) and Marion How (1962: 26).
Vinnicombe (2009b: 179–80) herself saw one such skull at this store in 1957, acquired
it from Mr Calder-Potts, the store owner, and had it identied by Hertha de Villiers at
the University of the Witwatersrand, where it was apparently catalogued and curated
in the Raymond Dart Palaeontological Museum. Regrettably, my own subsequent
enquiries failed to establish its presence in the catalogue of the university’s Department
of Anatomy (Vinnicombe 2009b: 186–7, note 56; Alan Morris pers. comm. 2007).
A likely terminus ante quem for all these events is set by Captain Dobson’s (1910) report
on highland Lesotho, which makes no mention of the shelter still being occupied, and
by the dated grafti that now mark much of its rear wall. The oldest noted during
a close examination in 2007 was left by a certain D.I.H. on 24 January 1901. It was
followed by others dated to 16 May 1909 (H.B. Gabaaliane), 1910 (SIASI) and New
Year’s Eve 1910 (a European with the surname Woodrooffe, accompanied by at least
nine Basotho, three with the surname Mofolo) (Fig. 11). Then, from November 1912
to October 1923 at least 31 people added their names to the walls, more than in the
almost nine decades since. Did Sekese’s (1912a, 1912b) Sesotho-language publications
of 7 March and 4 July 1912 regarding how Soai met his end at Sehonghong spark this
upsurge of interest, which is extraordinarily intense compared to any other rock shelter
with which I am familiar in highland Lesotho? Regrettably, such attention continues
sporadically today, as recently reected in the immortalisation of the ‘Standard 4 and
5 Group Excursion’ of 4 May 2006.
CONCLUSIONS AND REFLECTIONS
As this most recent grafto indicates, Sehonghong is still a place of signicance for
the local population. Despite the difculties that still attend the roads of this part of
highland Lesotho, it is also attractive to a small number of (mostly) South African
tourists, particularly those with an interest in rock art and/or four-wheel-drive
adventures (Fig. 12). Though no sign remains on the site’s surface of the archaeological
excavations, which have been comprehensively backlled, its paintings are, to say the
least, in a desperate state. Exposed everywhere to exfoliation, dust from livestock rolling
166 SOUTHERN AFRICAN HUMANITIES, VOL. 22, 2010
Fig. 12. Sehonghong Shelter viewed from the track descending from Sehonghong village into the Sehonghong
River. The track is only motorable, and then with difculty, using four-wheel drive. Sehonghong itself
is in the centre of the image behind the spur projecting right into the course of the river. Traces of
paint and a few artefacts survive on the sandstone ledge immediately above the main rock shelter
and there are traces of occupation at two smaller sites to the front of it.
Fig. 11. Sehonghong: examples of twentieth-century grafti on the rear wall of the site.
MITCHELL: SEHONGHONG IN HISTORY 167
inside the shelter and the danger of further grafti, damp is an additional problem in
places. There is no prospect of this deterioration being reversed and, in the absence
of a committed strategy from Lesotho’s inadequately funded Department of Culture,
no likelihood of it being stopped in its tracks. The long-term future of Sehonghong’s
art is thus bleak, and the fact that the much vaunted and massively funded Maloti-
Drakensberg Transfrontier Conservation Area/Peace Park achieved next to nothing
of lasting signicance for cultural heritage management in Lesotho (Cain 2009)
only reinforces the tragedy that this encompasses. The irony is deepened as it recalls
the location of James Ferguson’s (1990) landmark study of the failings of so many
‘development’ projects only kilometres to the north of Sehonghong at Mashai. A site
that might otherwise—along with Pitsaneng and Melikane—have had a chance at
World Heritage status because of its signicance in understanding one of the world’s
great rock art traditions teeters on the verge of disintegration.
If there is to be a future for Sehonghong’s rock art it has to lie in the active
participation and guardianship of local communities, grounded—should tourist
access become easier—in a serious level of nancial charge for outsiders visiting the
site and the transparent transfer of that money to those communities. Sensitisation
of local people (Jolly 2001, 2006b) and controlled access of this kind, rather than
misguided and unviable attempts to deny access at all, offer the only plausible strategy
for safeguarding the art. While this may remain a pipe-dream, perhaps ongoing
archaeological investigations at the site, and the experience currently being gained
by community involvement in archaeological work ahead of the Metolong Dam on
the other side of Lesotho, may lay the foundations for it. Sehonghong’s paintings, its
archaeological deposits and the information that they can yield, and its very history
demand nothing less. One useful step would be fuller publication and interpretation
of the site’s paintings, drawing on Vinnicombe’s tracings, the ARAL archive and what
remains visible today. What could also now protably be undertaken is a sustained study
of the site’s meaning to those living near it and a detailed investigation of the history
of those populations since the arrival of the rst Basotho settlers at Sehonghong in
1878.
In the meantime, I hope that this paper has shown how, in at least some instances,
combing the historical literature can provide a more thorough and nuanced
understanding of the last decades of hunter-gatherer use of one particular site. Oral
traditions in other parts of Lesotho and investigation of missionary and secular
archives might yet recover comparable information in respect of other rock shelters,
such as Melikane, while Blundell (2004) has shown that something similar can be done
at selected sites in the Eastern Cape. Janette Deacon’s (1986, 1996) studies of the
Bleek-Lloyd archive and her concomitant eldwork in the Northern Cape Province,
like Garth Sampson’s (e.g. 1995) work in the Seacow Valley, identify other locations
where the combination of documentary and archaeological sources may be able to
track the ending of the hunter-gatherer way of life and the assimilation of surviving
Bushmen into quite different social and economic formations. In all four areas, and
perhaps elsewhere, the history of Bushman communities in the late nineteenth and
early twentieth centuries is still to be made.
168 SOUTHERN AFRICAN HUMANITIES, VOL. 22, 2010
ACKNOWLEDGEMENTS
I am grateful to Lesotho’s Protection and Preservation Commission and the chiefs of
Sehonghong and Khomo-ea-Mollo for having granted me permission to undertake
eldwork at Sehonghong; to Prof. Lucas Smits and Taole Tesele for having arranged
my rst visit to the site in 1985; to the late Pat Carter for having introduced me to its
study; to the late Pat Vinnicombe and Dr Luiz Costa for having suggested, in different
ways, that I write something along these lines; to Jeremy Hollmann, Aron Mazel and
Pieter Jolly for their comments on an earlier draft; to everyone who worked with me
at Sehonghong and to all the funding agencies that contributed to my excavations
there.
REFERENCES
Ambrose, D. & Sekoli, P. 1991. JoboMoteane’s account of a journey through the Lesotho Highlands 100
years ago – Part 5. Mehloli 3 (1): 4–7.
Barre, M. & Lamothe, M. 2009. Luminescence dating of archaeosediments: a comparison of K-feldspar
and plagioclase IRSL ages. Quaternary Geoarchaeology 5: 324–8.
Bleek, W.H.I. 1874. Remarks by Dr Bleek. Cape Monthly Magazine 9: 10–13.
Blundell, G. 2004. Nqabayo’s Nomansland: San rock art and the somatic past. Uppsala: Department of
Archaeology & Ancient History, Uppsala University.
Cain, C.R. 2009. Cultural heritage survey of Lesotho for the Maloti-Drakensberg Transfrontier Project,
2005-2006: palaeontology, archaeolog y, history and heritage management. South African
Archaeological Bulletin 64: 33–44.
Campbell, C. 1987. Art in crisis: contact period rock art in the south-eastern mountains of southern Africa. MSc thesis,
University of the Witwatersrand.
Carter, P.L. 1971a. Sehonghong: unpublished eld notebook. University Museum of Archaeology and
Anthropology, University of Cambridge.
Carter, P.L. 1971b. Notes on archaeological sites in the vicinity of Sehonghong. Unpublished manuscript.
University Museum of Archaeology and Anthropology, University of Cambridge.
Carter, P.L. 1976. The effects of climatic change on settlement patterns in eastern Lesotho during the
Middle and Later Stone Ages. World Archaeology 8: 197–206.
Carter, P.L. 1978. The prehistory of eastern Lesotho. PhD thesis, University of Cambridge.
Carter, P.L., Mitchell, P.J. & Vinnicombe, P. 1988. Sehonghong: the Middle and Later Stone Age industrial sequence
at a Lesotho rock-shelter. BAR International Series 388. Oxford: British Archaeological Reports.
Carter, P.L. & Vogel, J.C. 1974. The dating of industrial assemblages from stratied sites in eastern
Lesotho. Man 9: 557–70.
Challis, W. 2008. The impact of the horse on the AmaTola ‘Bushmen’: new identity in the Maloti-Drakensberg mountains
of southern Africa. D.Phil. thesis, University of Oxford.
Challis, W., Mitchell, P.J. & Orton, J. 2008. Fishing in the rain: control of rain-making and aquatic resources
at a previously undescribed rock art site in highland Lesotho. Journal of African Archaeology 6:
203–18.
Clarke, M. 1888. Unexplored Basutoland. Proceedings of the Royal Geographical Society 10: 519–30.
Collis, B. 2010. Dating the introduction of pottery to the Maloti-Drakensberg region of southern Africa using Optically
Stimulated Luminescence. BA Hons thesis, University of Oxford.
Damane, M. & Sanders, P. 1974. Lithoko: Sotho praise-poems. Oxford: Clarendon Press.
Deacon, J. 1984. The Later Stone Age of southernmost Africa. BAR International Series 213. Oxford: British
Archaeological Reports.
Deacon, J. 1986. “My place is the Bitterputs”: the home territory of Bleek and Lloyd’s |Xam San informants.
African Studies 45: 135–56.
Deacon, J. 1996. Archaeology of the Flat and Grass Bushmen. In: J. Deacon & T.A. Dowson, eds,
Voices from the past:
|
Xam Bushmen and the Bleek and Lloyd collection. Johannesburg: Witwatersrand
University Press, pp. 245–70.
Dobson, M.C. 1910. Military report on Basutoland. Unpublished report. The War Ofce, London.
Dornan, S.S. 1909. Notes on the Bushmen of Basutoland. South African Journal of Philosophy 18: 437–50.
Dowson, T.A. 1994. Reading art, writing history: rock art and social change in southern Africa. World
Archaeology 20: 116–28.
MITCHELL: SEHONGHONG IN HISTORY 169
Dowson, T.A. 1995. Hunter-gatherers, traders and slaves: the ‘Mfecane’ impact on Bushmen, their ritual
and their art. In: C. Hamilton, ed., The Mfecane aftermath: reconstructive debates in southern African
history. Johannesburg: Witwatersrand University Press, pp 51–70.
Dowson, T.A. 1999. Rain in Bushman belief, politics and history: the rock art of rain-making in the south-
eastern mountains, southern Africa. In: C. Chippindale & P. Taçon, eds, Archaeology of rock art.
Cambridge: Cambridge University Press, pp. 73–89.
Ellenberger, V. 1953. La n tragique des Bushmen. Paris: Amiot Dumont.
Ferguson, J. 1990. The anti-politics machine: ‘development’, depoliticization and bureaucratic state power in Lesotho.
Cambridge: Cambridge University Press.
Francis, M. 2006. Explorations in ethnicity and social change among Zulu-speaking San descendents [sic] of the
Drakensberg Mountains, KwaZulu-Natal. PhD thesis, University of KwaZulu-Natal.
Germond, R.C. 1967. Chronicles of Basutoland. Morija: Sesuto Book Depot.
Gill, S.J. 1993. A short history of Lesotho. Morija: Morija Museum and Archives.
Hobart, J.H. 2004. Pitsaneng: evidence for a neolithic Lesotho? Before Farming 3: 261–70.
How, M.W. 1962. The mountain Bushmen of Basutoland. Pretoria: Van Schaik.
Jacobs, Z., Roberts, R.G., Galbraith, R.F., Barré, M., Deacon, H.J., Mackay, A., Mitchell, P.J., Vogelsang,
R. & Wadley, L. 2008. Ages for Middle Stone Age innovations in southern Africa: implications
for modern human behavior and dispersal. Science 322: 733–5.
Jacottet, E. 1893. L’oeuvre des Maloutis (2). Journal des Missions Evangéliques 68: 510–19.
Jolly, P. 1986. A rst generation descendant of the Transkei San. South African Archaeological Bulletin 41: 6–9.
Jolly, P. 1994. Strangers to brothers: interaction between south-eastern San and southern Nguni/Sotho communities. MA
thesis, University of Cape Town.
Jolly, P. 1996. Interaction between south-eastern San and southern Nguni and Sotho communities c. 1400
to c. 1880. South African Historical Journal 35: 30–61.
Jolly, P. 2001. Re kasireletsalitsoantsotseejoang? (What can we do to protect the paintings?). Poster produced
in partnership with the Archaeological Resource Development Project, GAES, University of
the Witwatersrand, for distribution to schools in Lesotho.
Jolly, P. 2006a. The San rock painting from “the upper cave at Mangolong”, Lesotho. South African
Archaeological Bulletin 61: 68–75.
Jolly, P. 2006b. Two related rock art conservation/education projects in Lesotho. The Digging Stick 23 (2):
13–14.
Jolly, P. 2008. An historical account of a rite depicted in a painting from the cave at Melikane, Lesotho.
South African Archaeological Bulletin 63: 125–9
Kennan, T.P. 1959. Notes on a journey round Basutoland. Basutoland Notes and Records 1: 38–48.
Lewis-Williams, J.D. 1986. The last testament of the southern San. South African Archaeological Bulletin 41:
10–11.
Lewis-Williams, J.D. 2003. Images of mystery: rock art of the Drakensberg. Cape Town: Double Storey.
Mangoaela, Z.D. 1921. Lithoko tsa ea marena ea Basotho. Morija: Sesuto Book Depot.
Mitchell, P.J. 1987. The Sehonghong bladelet industry in the context of the southern African Later Stone Age. DPhil
thesis, University of Oxford.
Mitchell, P.J. 1993. Late Pleistocene hunter-gatherers in Lesotho: report on eldwork at Sehonghong
Rockshelter, July to September 1992. Nyame Akuma 39: 43–9.
Mitchell, P.J. 1994. Understanding the MSA/LSA transition: the pre-20 000 BP assemblages from new
excavations at Sehonghong rock-shelter, Lesotho. Southern African Field Archaeology 3: 15–25.
Mitchell, P.J. 1995. Revisiting the Robberg: new results and a revision of old ideas at Sehonghong rock-
shelter, Lesotho. South African Archaeological Bulletin 50: 28–38.
Mitchell, P.J. 1996a. The archaeological landscape at Sehonghong, a hunter-gatherer site in the Lesotho
highlands, southern Africa. Antiquity 70: 623–38.
Mitchell, P.J. 1996b. The Late Quaternary of the Lesotho highlands, southern Africa: preliminary results and
future potential of recent research at Sehonghong Shelter. Quaternary International 33: 35–44.
Mitchell, P.J. 1996c. Filling a gap: the early and middle Holocene assemblages from new excavations at
Sehonghong Rock Shelter, Lesotho. Southern African Field Archaeology 5: 17–27.
Mitchell, P.J. 1996d. Sehonghong: the late Holocene assemblages with pottery. South African Archaeological
Bulletin 51: 17–25.
Mitchell, P.J. 1996e. Prehistoric exchange and interaction in south-eastern southern Africa: marine shells
and ostrich eggshell. African Archaeological Review 13: 35–76.
Mitchell, P.J. 2001. Recent archaeological work in Lesotho: an overview of eldwork 1988–2001. National
University of Lesotho Journal of Research 9: 1–24.
170 SOUTHERN AFRICAN HUMANITIES, VOL. 22, 2010
Mitchell, P.J. 2006–07. Remembering the Mountain Bushmen: observations of nineteenth century hunter-
gatherers in Lesotho as recorded by Victor Ellenberger. Southern African Field Archaeology 15–16:
3–11.
Mitchell, P.J. 2009. The aked stone artefact assemblages from Likoaeng: a late Holocene sequence in
highland Lesotho and its regional context. Southern African Humanities 21: 117–56.
Mitchell, P.J. & Challis, W. 2008. A ‘rst’ glimpse into the Maloti Mountains: the diary of James Murray
Grant’s expedition of 1873–74. Southern African Humanities 20 (2): 399–461.
Mitchell, P.J., Plug, I., Bailey, G.N. & Woodborne, S. 2008. Bringing the Kalahari debate to the mountains:
late rst millennium AD hunter-gatherer/farmer interaction in highland Lesotho. Before Farming
[online version] 2008/2 article 4.
Mitchell, P.J. & Smith, B.W., eds. 2009. The eland’s people: new perspectives on the rock art of the Maloti-Drakensberg
Bushmen. Essays in memory of Pat Vinnicombe. Johannesburg: Witwatersrand University Press.
Mitchell, P.J. & Vogel, J.C. 1994. New radiocarbon dates from Sehonghong rock-shelter, Lesotho. South
African Journal of Science 90: 380–8.
Orpen, J.M. 1874. A glimpse into the mythology of the Maluti Bushmen. Cape Monthly Magazine 9:
1–13.
Plug, I. & Mitchell, P.J. 2008a. Fishing in the Lesotho highlands: 26 000 years of sh exploitation, with
special reference to Sehonghong Shelter. Journal of African Archaeology 6: 33–55.
Plug, I. & Mitchell, P.J. 2008b. Sehonghong: hunter-gatherer utilization of animal resources in the highlands
of Lesotho. Annals of the Transvaal Museum 45: 1–23.
Prins, F.E. 1990. Southern Bushman descendants in the Transkei: rock art and rainmaking. South African
Journal of Ethnology 13: 110–16.
Prins, F.E. 1994. Living in two worlds: the manipulation of power relations: identity and ideology by the last
San rock artist in Tsolo, Transkei, South Africa. Natal Museum Journal of Humanities 6: 179–93.
Prins, F.E. 2009. Secret San of the Drakensberg and their rock art legacy. Critical Arts 23: 190–208.
Sadr, K. 2003. The Neolithic of southern Africa. Journal of African History 44: 195–209.
Sampson, C.G. 1995. Acquisition of European livestock by the Seacow River Bushmen between AD
1770–1890. Southern African Field Archaeology 4: 30–36.
Sekese, A. 1912a. Baroa. Leselinyana la Lesotho. 7th March.
Sekese, A. 1912b. Tsa Baroa. Leselinyana la Lesotho. 4th July.
Sekese, A. 1924. Lesotho fatsela bo ntat’a rona. Leselinyana la Lesotho. 17th October.
Smits, L.G.A. 1973. Rock-painting sites in the upper Senqu Valley, Lesotho. South African Archaeological
Bulletin 28: 32–8.
Stow, G.W. 1905. The native races of South Africa. London: Swan Sonnenschein.
Vinnicombe, P. 1976. People of the eland: rock paintings of the Drakensberg Bushmen as a reection of their life and
thought. Pietermaritzburg: University of Natal Press.
Vinnicombe, P. 2009a. People of the eland: rock paintings of the Drakensberg Bushmen as a reection of their life and
thought. Johannesburg: Witwatersrand University Press.
Vinnicombe, P. 2009b. Basotho oral knowledge: the last Bushman inhabitants of the Mashai District,
Lesotho. In: P.J. Mitchell & B.W. Smith, eds, The eland’s people: new perspectives on the rock art of the
Maloti-Drakensberg Bushmen. Essays in memory of Pat Vinnicombe. Johannesburg: Witwatersrand
University Press, pp: 165–91.
Webb, R.S. 1950. Gazetteer for Basutoland. Paarl: privately published.
Wright, J.B. 1971. Bushman raiders of the Drakensberg, 1840–1870: a study of their conict with stock-keeping peoples
in Natal. Pietermaritzburg: University of Natal Press.
Wright, J.B. 2007. Bushman raiders revisited. In: P. Skotnes, ed., Claim to the country: the archive of Wilhelm
Bleek and Lucy Lloyd. Cape Town: Jacana, pp. 118–29.
Wright, J.B. & Mazel, A.D. 2007. Tracks in a mountain range: exploring the history of the uKhahlamba-Drakensberg.
Johannesburg: Wits University Press.
