Cimbeb •asiaSTAATSMUSEUM - WINDHOEK - STATE - MUSEUM - WINDHOEK - STAATSMUSEUMCimbebasia Ser. B. - Vol. 3 - No.~ 2 December 1983

HUMAN BURIALS IN HUT CIRCLES ATSYLVIA HILL, S.W. AFRICA / NAMIBIA

Myra Shackley,Department of Archaeology,

University of Leicester,Leicester, LEI 7RH,

England.

(with 3 figures)(Manuscript received 20th July 1983)

ABSTRACT

Three human skeletons found in association with a coastal hut complex at Sylvia Hill (26°34'S, 14°36'E), are described. Theskeletons represent chance discoveries made during a ground survey project carried out in 1981. Two of the burials were ofKhoisan males, dated 510 ± 45 BP (Pta-3294) and 1070 ± 60 BP (Pta - 3295) respectively, and the third burial was of aneonate. All three individuals had been buried under cairns in collapsed hut circles with no associated grave goods. Descrip-tions and analyses by Professor Hertha de Villiers (University of the Witwatersrand, Johannesburg) are also included.

CONTENTS

I.

II.

III.

IV.

V.

Introduction .

The Burials .

Chronology .

Discussion .

Acknowledgments .

102

104

105

105

106

106VI. References .

Cimbebasia (B) 101 -106

Cimbebasia

I. INTRODUCTION

Hut circle complexes have been reported from severallocations in Namibia (Carr et al. 1978; Rudner 1957;Viereck 1968) including some from the southernNamib at Cape Cross (Wendt 1972) and atGrillenberg (2906'S, 14°34'E) near Conception Bay(2406'S, l4°34'E), (Vogel and Visser 1981). Theauthor has mapped a further series at Rooikamer(23°43'S, 15°26'E) on the Kuiseb river (Shackley, inprep.), and further numerous isolated examples oc-cur within the central Namib desert. The function ofthese circles seems likely to be complex, either long-term occupation sites (Carr et al. 1978) or occasional-ly mere hunting blinds, but they are usually ar-chaeologically sterile and located primarily inresponse to available food resources.

The extensive circle complex at Sylvia Hill is especial-ly interesting, firstly because of its location on a com-plex mass of marble outcrops interdigitated withcoastal dune sands sloping abruptly to the sea via asteep cliff (Fig. 1), and secondly because of a buriedspring of fresh water within walking distance. Thelocation of the circles appeared superficially to berelated to prevailing winds, and it was decided tomap a substantial part of the area to test thishypothesis. Individual circles were measured anddescribed, and associated shell midden depositssampled. The full results of this survey have yet to bepublished, but radiocarbon dates of 610± 50BP(Pta-3253) and 290± 45BP (Pta-3252) were obtainedfrom circles Nos. 20 and 29 (Fig. 2), together with ex-tensive molluscan remains and surface scatters ofpottery. Work on an isolated rock outcrop produced

o WINDHOEK

Atlantic Ocean

o 30m~

;"

/Ii • Circle complexes

i; ~ Shifting sand

r',..... oI

300kmI

Figure I. Location of Sylvia Hill on the coast of Namibia at 25°03'3., 14°54'E

102

Shackley - Sylvia Hill

58 hut circles (Fig. 2) in various constructional stylesand degrees of collapse; the presence of some smallstone mounds was noted. Similar conical piles ofstones O.3-0.8m high had been observed from theZerissene Mountain circle complex (Carr et al. 1978)and interpreted as marker cairns. It was thereforethought worthwhile to remove these cairns, afterplotting the circles, in the hope of obtaining dateableorganic material which would illustrate the temporalrange of the settlement. All three proved to concealburials - two of adult males and one of a neonate.The bodies had apparently been compressed into

spaces in the rock floor of the circles beneath thecairns and weighed down with stones before beingcovered by additional material derived either fromthe collapse of the circle or from deliberate dismant-ling of the circle to bury its former inhabitant. Therewere no associated grave goods. All three skeletonswere cleaned, drawn, planned, photographed,removed and packed in sandboxes for transport toJohannesburg, but it was felt that since the object ofthe expedition was to plot the surface distribution ofshelters no further clearance should be undertakensince the re-use of circles for burials was clearly com-

18

•o Grave Complex I

Atlantic Ocean

B

OQ

.'5

'3 ••44

.423

• 30 ••3

'

'''a-'2 6

5 ~9.28

7 8

32 33

•• 3' 38 .4'•• 35 ••

•• 39 ·4036 37

Circle Diameters;<1·5m1'5-1'9m2·O-2·5m

••••c:=J~~

>2·5m

Rocky Outcrop

50m

40m

Loose shattered30m rock piles

-=-

20m 110m

56

Section A,B

Steep cliffs

Figure 2. Circle distribution at Sylvia Hill and location of grave complex I.

103

Cimbebasia

monplace in this area, although not noted before.

II. 'T H E BUR I A L S'

Grave complex I. An apparently collapsed hut circlecomplex covering an area of 3m2 near the apex of therock outcrop (Fig. 2) contained a stone 'cairn' some0.70m high constructed from the remains of two col-lapsed shelters. The two graves revealed by theremoval of the cairn were spatially distinct, althoughfilled with blown sand and rock packing. Grave I(Fig. 3c) contained the remains of an immature maleaged 17-23 years, compressed into a depression in therock floor some 45x30cm in size. It was accompaniedby a much smaller grave (Fig. 3b) with the fragmen-tary remains of a neonate.

The cranium and mandible were well preserved andcomplete with the exception of some post-depositional erosion, and G I had his complete per-manent dentition with the exception of thepostmortem loss of II (left) and M, (right),A x i a I postcranial remains included vertebrae (7cervical, 10 thoracic, 3 lumbar), the sacrum, 22 ribs,

manubrium, and sternum. The appendicular skeletonwas represented by 2 clavicles, the right scapulatogether with fragments of the left, a complete set oflong bones with isolated epiphyses, right ischium,complete carpal and metacarpal bones, 24 phalanges,I calcaneum, 2 tali, 7 other tarsal bones, 10(complete) metatarsal, 14 phalanges and I patella.Features of the skull and ilium suggested that the re-mains were male, and the platymeric index(100FeDI/FeD2\ = 84.2%) indicated flattening ofthe femoral shaft which was not evident in the tibialshaft (100TiD2/TIDI '" 87070).Flattening of the shaftshas, in the past, been specifically associated with San(Bushman) populations. However, Lisowski (1968)showed that this flattening resulted from nutritionaldeficiencies which affected the structure of the bonesand so influenced the osseous resistance to thestresses of locomotion. The cranium was small withslight facial projection (100 GLlLB=98.91l,10), arelatively broad palate (100 x G2' /G l ' = 881l,10)and alow squat mandibular ramus (rameal index 100 rb'vrl= 77.5Il,1o)indicative 0:- a Khoisan individual. Thisidentification was supported by the small mastoidprocess, shallow digastric fossa, delicate tympanicplate and the contour of the dental arcade.

o 10 20Scale throughout;

30cm,

. k I G2 b Neonate, skeleton G3, c. Male Khoisan burial G I,Figure 3. a. Male Khoisan seeton ,.All burials are drawn to the same scale.

104

Shackley - Svlvia Hill

The remains of the neonate, skeleton G3, were veryfragile, and occupied a grave only 25cm long in closeproximity to the grave of G I and less than 7cm deep.The cranial remains (represented by the right frontal,fragments of the parietal and occipital bones, thebody and left greater wing of the sphenoid, and theright and left zygomatic bones) were too fragmentaryto permit any comment on population affinities.Half the mandible was also present together with thefollowing pieces of the postcranial skeleton: Axial -20 complete vertebrae, 42 lateral pieces. Complete setof 24 ribs, manubrium and 3 sternebrae. Appen-dicular - 2 clavicles, left scapula, 2 humeri, the prox-imal ends of the 2 ulnae, femora, left tibia, 2 fibulae,2 ilia, and only 8 metacarpal/tarsal bones which,with 8 phalanges, represented the entire remains ofthe hands and feet. Loss of bones was apparentlycaused by in situ weathering in the damp sandy fill ofthe grave.

Grave 2 complex. During the cleaning of another hutcircle (No. 297) situated on a small rock outcropsome 75m east of the main 'island' (Fig. 1) a secondgrave was revealed, also concealed by a stone pile fill-ing the circle (Fig. 3a). This contracted burial was ofa much larger adult Khoisan male, aged 25-35 years,which had also been buried in a contracted positionand was much damaged by erosion and crushingfrom the weight of rocks. The cranial vault showedpremature closure of the sagittal suture. The mandi-ble was nearly complete, as was the permanent denti-tion. All teeth except the third molars showed wearwith dentine exposure. The postcranial skeleton wasrepresented as follows: axial - 6 cervical vertebrae, 7thoracic, fragments of 6 others, 3 lumbar vertebraeplus 2 other fragments, fragments of the sacrum, 17ribs, manubrium and 2 sternebrae. Appendicular - 2clavicles, 2 fragmented scapulae, humeri, radii,ulnae, femora, tibiae, all the long bones being frac-tured, together with fragmentary handbones (9 car-pal, 7 metacarpal) and footbones (2 calcanei, 2 tali, 8other tarsal, 6 metatarsal bones, 26 phalanges). Thevertebral bodies showed bony lipping, there wasevidence of a healed fracture on the 2nd metacarpal.The remains suggested a fully adult individual aged23-35 at the time of death. No nutritional deficiencieswere apparent, and features of the skull andpost cranial skeleton suggest that G2 was a male. Theestimated living stature is 158cm. The cranial vaultwas small (L = 181ern) and narrow, and the cranialbase missing, giving an incomplete facial skeleton.However, the face appeared to be non-projectingwith a low mandible and relatively broad ramus(rameal index 100 rb+/rl 69.1070) indicative of aKhoisan skull.

To consider further the assignment of the more com-plete G 1 remains to the Khoisan population group,Penrose's (1954) distance statistic was applied to the

data obtained for Gland for five comparative maleseries, viz. San (Bushman), Natal Nguni, CapeNguni, Sotho, Shangana- Tonga (Southern AfricanNegro groups). The values obtained indicated thatGl has closer affinities with the San group than withanyone of the Negro groups, especially with respectto the shape component of the statistic.

III. CHRONOLOGY

A radiocarbon date of 510 ± 45BP (Pta-3294) wasobtained from collagen from G 1, but it was notpossible to date the associated neonate as this wouldhave required virtually the whole skeleton. The se-cond skeleton, G2, was dated at 1070 ± 60BP(Pta-3295) which showed it to be much older.

IV. DISCUSSION

The wide disparity in the dates (nearly 500 years) bet-ween these two major burials suggests that the SylviaHill hut circle complex was in use by Khoisan peoplefor a long period; a range of nearly 800 years is in-dicated by the dates obtained on charcoal samplesfrom circles 20 and 29.

It seems likely that the circle complex was located inresponse to the availability of fresh water and amplesupplies of molluscan food, together with shelterfrom the prevailing onshore winds. Circles seem! tohave been used on an opportunistic basis for burial ifthe need arose. The lack of grave goods and the con-tracted posture are common features of Khoisanburials, but in this case the contracted position is alsorelated to the necessity for compressing the bodies in-to the relatively small available spaces in the rockfloors. It seems likely that the bodies were thencovered with stones, in order to deter scavengers suchas the brown hyaena, and that the graves fillednaturally with blown sand. Although no other ar-chaeological material was associated with theseburials, surface artefact scatters including pottery areof a type widely recognised in south-west Africa anddated to 490 ± 50 (Pta-2295) at Hottentot Bay(26°09'S., 14°58'E). Kolb (1719) first reponed acomparable intact adult skeleton buried in sand in acrouched position which was recently dated to 71Oj;,50 (Pta-1863) (Vogel and Visser 1981) from Concep-tion Bay (Fig. I) 24°01 '5, 14°34'E, where fresh wateris also available at c. 3m depth, and where the sitealso seems to have been visited over many centuriesduring this millennium in connection with the ex-ploitation of marine food resources.

105

Cimbebasia

V. ACKNOWLEDGMENTS

I am most grateful to Professor Hertha de Villiers(Department of General Anatomy, University of theWitwatersrand, Johannesburg) for her report onthese three skeletons, and to Dr. J.e. Vogel(C.S.I.R. Pretoria) for undertaking the radiocarbondates. The- expedition to Sylvia Hill was undertakenwith colleagues from the Desert Ecological ResearchUnit at Gobabeb, under the direction of Dr. M.K.Seely, to whom the writer is much indebted. Mrs. P.de Villiers kindly transported the skeletal remains toJohannesburg; Leon Jacobson (State Museum,Windhoek) provided advice and assistance;fieldwork was financed by grants from the BoiseFund (University of Oxford), the British Academyand the L.S.B. Leakey Foundation; and permissionkindly granted by the Division of Nature Conserva-tion and Tourism (South West Africa) and Con-solidated Diamond Mines. The illustrations are thework of Andy Clark.

VI. REF ERE N C E S

CARR, M.J., CARR, A.e. and JACOBSON, L.(1978). Hut remains and related features fromthe Zerissene Mountain area: their dis-tribution, typology and ecology. Cimbe-basia Ser. B2 (11) 237-54.

DE VILLIERS, H. (1968). The skull of the SouthAfrican Negro. Witwatersrand UniversityPress.

106

KOLB, P. (1719). Caput Bonei Spei hidiernum, dasist vollstandige Beschreibung des Afri-kanischen Vorgeburges der Guten Hoffnung,pp. 156-7 in Reise zum Vorgeburge der GutenHoffnung, ed. P. Germann, Leipzig, Brock-haus.

LISOWSKI, F.P. (1968). Osseous characteristics ofthe femur and tibia indicating malnutrition orracial differences, in 8th International Con-gress of Anthropological and Ethno-logical Sciences, Tokyo and Kyoto, Vol. II'pp. 61. Published by Science Council,Japan.

PENROSE, L.S. (1954). Distance, size and shape.Ann. Eugen. 18,337-343.

RUDNER, J. (1957). The Brandberg and itsarchaeological remains. J. South West AfricaSci. Soc. 12.7-74.

VIERECK, A. (1968). Die Spuren der altenBrandbergbewohner. Scient. Res. S. W. Afr.6.1-80.

VOGEL, J.e. and VISSER, E. (1981). PretoriaRadiocarbon Dates II. Radiocarbon 23. (I)43-80.

WENDT, W.E. (1972). Preliminary report on anarchaeological research programme in south-west Africa. Cimbebasia Ser. B I. 1-61.