- Email: [email protected]
Bone hoes from the Middle Iron Age, Limpopo Province, South Africa
Quaternary International 472 (2018) 126e134
Contents lists available at ScienceDirect
Quaternary International journal homepage: www.elsevier.com/locate/quaint
Bone hoes from the Middle Iron Age, Limpopo Province, South Africa Justin Bradfield a, b, *, Annie R. Antonites c a
Evolutionary Studies Institute, University of the Witwatersrand, Johannesburg 2050, South Africa School of Geography, Archaeology and Environmental Studies, University of the Witwatersrand, Johannesburg 2050, South Africa c Department of Anthropology and Archaeology, University of South Africa, Pretoria, South Africa b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 13 January 2017 Received in revised form 10 July 2017 Accepted 26 October 2017 Available online 4 November 2017
This paper presents the first recognised evidence of bone hoes in South Africa. Two bovine scapulae and a portion of a long bone show use-trace evidence that supports our interpretation as ground-working implements. The scapulae were probably hafted onto wooden handles using a combination of plant fibres and sinew, whereas the tool made from the long bone appears not to have been hafted. Bone hoes represent a short-lived technological innovation, although the reasons to account for this remain speculative. The recognition of these agricultural implements poses interesting questions about the extent and variety of bone working among Iron Age agriculturalists in the Limpopo Valley during the 10th e 13th centuries AD, and potentially also about the nature of women's work in these communities. © 2017 Elsevier Ltd and INQUA. All rights reserved.
Keywords: South Africa Iron Age Limpopo Valley Scapula hoes Agricultural implements Farming communities Worked bone Use-traces
1. Introduction Bone tools made from animal scapulae (shoulder blades) are found among pre-industrial societies in most parts of the world (see Griffitts, 2006; Xie, 2014). Various interpretations have been proffered for these tools, but most agree that they were used for moving earth. The precise function ranges from construction tools for digging trenches and pits (e.g. Curwin, 1926, 1937; Griffitts, 2006) to agricultural implements (e.g. Strong, 1933; Bell and Cross, 1980; Griffitts, 2006; Xie, 2014). Among the Plains Indians of the central United States for instance, bison scapulae were used in historic times for cultivating small garden crops like maize, beans and squash (Strong, 1933; Bell and Cross, 1980). Although perhaps less widely used for this purpose than wood, stone and metal implements (Xie, 2014), scapulae have a high tensile strength making them ideal for heavy-duty work such as digging, and thus more economical than their stone and metal counterparts in terms of manufacturing time (Griffitts, 2006; Xie, 2014). In southern Africa broad-blade, spatula-shaped bone tools have
* Corresponding author. Evolutionary Studies Institute, University of the Witwatersrand, Johannesburg 2050, South Africa. E-mail address: jbradfi[email protected] (J. Bradfield). https://doi.org/10.1016/j.quaint.2017.10.028 1040-6182/© 2017 Elsevier Ltd and INQUA. All rights reserved.
been recovered from numerous sites along the South Coast (Meiring, 1952). Where recovered in situ these tools have been placed within the Final Later Stone Age technocomplex; roughly the last 4000 years (Meiring, 1952; cf. Lombard et al., 2012). These tools are made from ribs or slivers of long bone and the edges ground into shape against an abrasive surface (Meiring, 1952). No proper use-trace studies of these artefacts have been conducted yet, and their precise function remains elusive. Proffered interpretations range from shell fish scoops, melon knives for shredding vegetables, meat mattocks, and skin scrapers (Clark, 1959) to hand adzes for wood working (Meiring, 1952). Further inland three similar tools, two of which are fashioned from bovine scapulae, have been recovered from Iron Age contexts dating between c. 1000 CE - 1250 CE in the Limpopo River Valley.1 Voigt (1983: 110) previously described one of these tools from K2 as a meat mattock for the preparation of skins, based on the extensive visible polish covering the artefact surface. The other two artefacts, one made from a scapula and the other from a piece of a large bovid
1 It is possible that more bone hoe-like objects exist but have not been recognised as tools. Voigt (1983) describes two other similarly-shaped tools from K2, one of which she interpreted as a chopping board. Neither tool could be located for analysis here.
J. Bradfield, A.R. Antonites / Quaternary International 472 (2018) 126e134
long bone, derive from the roughly contemporaneous site of Pont Drift (Fig. 1). Bone working is not unusual among Iron Age communities of this region (Hanisch, 1980; Voigt, 1983; Antonites et al., 2016). K2 and the nearby 13th century capital of Mapungubwe have well developed bone-working traditions, yet the function of spatulashaped tools remains poorly understood (Voigt, 1983). Understanding what these tools were used for may shed light on the diversity of bone tool technology among Bantu-speaking agriculturalists in the Limpopo Valley during this period. The aim of the present paper is to formally describe the use-trace features of three tools found at K2 and Pont Drift. The extensive use-wear on these artefacts, which are unique to the South African Iron Age, warrants closer scrutiny into their precise function.
2. Background
127
contemporaneous settlements in South Africa's middle Limpopo Valley e K2 and Pont Drift (Fig. 1). The area is best known for the emergence of class distinction and political centralisation in the 13th century AD (Huffman, 2000, 2007). K2 was a large ‘elite’ settlement that asserted socio-political and economic influence over a wide region (c. 1000 CE e 1200 CE, ‘K2 period’). Intermittent excavations between the 1930s and 1990s revealed several household areas and domestic middens surrounding a large central animal enclosure and communal midden. The K2 inhabitants kept livestock (cattle, sheep and goats), cultivated sorghum, millet and legumes in the floodplain, and occasionally hunted and gathered wild resources. They manufactured a range of crafts and traded ivory and other raw materials for glass beads with Arab traders on the East African coast (Huffman, 2007; Gardner, 1963; Meyer, 1998; Miller, 2001; Voigt, 1983). Sometime during the early ‘Transitional K2 period’ (1200 CE e 1250 CE), K2 was abandoned and its community relocated to the nearby site of Mapungubwe Hill e a move marked by the physical separation between elites and commoners
The bone implements described here come from two roughly
Fig. 1. Map showing the geographical region of the study area.
128
J. Bradfield, A.R. Antonites / Quaternary International 472 (2018) 126e134
and a shift in the political economy towards a more centralised system. Pont Drift is located about 25 km west of K2 and lies about 1.5 km south of the Limpopo River. Excavations in the late 1970s revealed an extensive vertical deposit, with the earliest occupation dating to the ‘Zhizo period’ (c. 850 CE e 1000 CE). A K2 period animal byre and a thick midden overlay the Zhizo layers, followed by a series of Transitional K2 hut floors and courtyard surfaces (Hanisch, 1980; Huffman, 2007). Although subsistence and material culture patterns are similar to those at K2, the assemblages are typical of ‘commoner’ settlements in each occupation phase (Hanisch, 1980; Huffman, 2007; Plug, 2000). In sub-Sharan Africa, iron hoes are common items among the various Bantu-speaking agriculturalists and were used for a range of excavation tasks. Apart from agriculture, iron hoes were used to hunt for subterranean prey, like rats and hares (Lupo and Schmitt, 2002), to excavate mineral ore (Chirikure, 2015), and as exchange items (Moffett et al., 2017). From about the 6th century AD tanged iron hoes were a common implement among the agriculturalist groups south of the Zambezi River (Gunda, 1976), although most extant examples are far younger (Moffett et al., 2017). Iron hoes appear to have replaced earlier chipped stone counterparts, which have been found in East and West Africa between 7 ka and 1.5 ka ago (Phillipson, 2005; Lane, 2013). Certainly by the 19th century hoes were being made almost exclusively from iron (e.g. Livingstone, 1851; Chapman, 1864; Gunda, 1976), although wooden digging sticks and stone hammers continued to be used (Childs and Herbert, 2005). In rare instances when, for whatever reason, iron was difficult to obtain, wooded hoes were used (Earthy, 1933; Shaw and van Warmelo, 1974; Moffett et al., 2017). An interesting feature about iron hoes is their symbolic male/ female dichotomy. Iron working was regarded as a male activity and was imbued with symbolism pertaining to procreation (Childs and Herbert, 2005; Chirikure, 2015). While only men were allowed to smelt and forge iron, fully-forged hoes were symbolically associated with women, as the ones responsible for the bulk of agricultural work (Chapman, 1864; Mumford, 1934; Gunda, 1976). Iron hoes also symbolise this link between women and production through its historical importance in marriage transactions in subSaharan Africa (e.g., Culwick and Culwick, 1934; Hunt, 1931; Mumford, 1934; see also Kuper, 1982: 21e25). Among some southern Bantu-speakers, iron hoes were incorporated into ancestor veneration rituals (e.g. Stayt, 1968: 245), and miniature iron hoes were worn by women during ceremonial dances to drive away demons (Gunda, 1976). In addition to the utilitarian, symbolic and ritual aspects of iron hoes, they also served as currency throughout Africa (e.g. Livingstone, 1851; Childs and Herbert, 2005; Chirikure, 2015). The trade in iron was usually controlled by men, again illustrating the fluidity of association attached to iron hoes (Schoeman, 2013). Iron hoes have been recovered from several Iron Age farmer settlements in southern Africa (e.g., Anderson, 2009; Garlake, 1973; Miller, 2002; Schoeman, 2013), and at least seven shape types have been recognised (Moffett et al., 2017). By contrast, the authors are aware of only one reference to bone being used as hoe material among the southern Bantu-speakers. Cattle scapulae apparently were used among the Zulu of the 19th century (Bryant, 1949), though none of these examples seem to have survived to posterity. Bone is regularly used to make hoe blades in other parts of the world (see Griffitts, 2006; Xie, 2014; Baron et al., 2016), and their frequency may even be under-represented as they may sometimes be mistaken for axe heads if no use-trace studies have been conducted (Luik, 2011). Among the Plains Indians of North America scapula hoes were hafted perpendicularly to wooden handles by binding them with sinew (Griffitts, 2006) and continued in use
after metal was easily available (Matthews, 1877). Among the Hemdu peoples of China scapula hoes were used to cultivate rice in soft swampy ground, but, unlike in North America, were gradually replaced by metal (Xie, 2014). 3. Use-wear associated with hoes Use-wear features that develop on bone implements associated with agricultural work have been described in two regions of the world, namely, the United States and China (Griffitts, 2006; Xie, 2014). In both regions large bovid scapulae were modified and used to till the soil. The spine of the scapula would be removed, spongy protuberances near the edges smoothed and the blade trimmed and sharpened to enhance efficacy (Bell and Cross, 1980; Griffitts, 2006; Xie, 2014). Hafted hoes usually penetrate the soil at an angle, which lead to certain concentrations of wear patterns (Semenov, 1964). Experimental studies have shown that microwear patterns vary depending on the conditions and composition of the soil (Xie, 2014). For instance, humic soils may produce striations associated with plant working due to the presence of vegetal fibres in the soil. Striation depth and width depend on the size of sand grains in the soil but are typically orientated diagonally to the working edge (Xie, 2014). Striations will mimic those produced by sediment abrasion (cf. d’Errico, 1993; Lyman, 1994; Fisher, 1995), but may also contain concentrations of smaller evenly distributed parallel striations (Griffitts, 2006; Xie, 2014). Polish will depend on the size of the substrate. A mix of bright polish, flattening and invasive polish has been noted on experimental bone hoes used to till soft, compact, clayey sediments (Griffitts, 2006; Xie, 2014). Xie (2014) considers polish to be the least helpful criteria for identifying earth working, as it is affected by too many variables and can resemble polish produced by dry hides and wood working, depending on the specific texture and consistency of the soil. Both Griffitts (2006) and Xie (2014) note the high incidence of reuse of bone scapula hoes in their respective study areas. Apart from cultivating domestic crops, bone hoes are also suitable for digging pits for post holes, cutting grasses for thatching, and processing skin and bark items (Campana, 1989; Wang, 1991; Griffitts, ra (2007) mention that wood or bone 2006). Legrand and Side scrapers will present with numerous notches and chips along the cutting edge of the tool. Some scapula ‘hoes’ were used by North American Plains Indians to break open squashes subsequent to digging them out of the ground (Griffitts, 2006). Another use for which scapula tools are suitable is for burnishing ceramics. Usewear features will differ depending on whether the clay is wet or dry (Meier, 2013), with dry clay producing numerous long parallel or oblique striations oriented perpendicularly to the edge of the ra, 2007; Meier, 2013). In this instance polish tool (Legrand and Side will be well developed, whereas wet clay produces moderate polish and asymmetric curving striations (Meier, 2013). 4. Methods and materials Three objects form part of our analysis (Table 1). The K2 scapula
Table 1 Provenience and morphometric data of the three artefacts mentioned in the text. Object
Provenience
Dimensions
W/289
K2, B.4.S.3, 12’.28’.30
TPD4612
TPD1/2, Block 2B, Layer 4, Feature 4
TPD4611
TPD1/2, Block 2AA, Layer 10, Feature 1.1
Length: 195 mm Max width: 69 mm Length: 147 mm Max width: 48.5 mm Length: 169 mm Max. width: 54 mm
J. Bradfield, A.R. Antonites / Quaternary International 472 (2018) 126e134
129
(W/289) is from a K2-period domestic context in the western settlement area of the site (Gardner, 1963; Meyer, 1998). The Pont Drift scapula (TPD4612) was excavated from a Transitional K2 household context, consisting of a living hut, smaller storage hut and household refuse (Hanisch, 1980: 237e245). The long bone (TPD4611) is from the preceding K2 period layers and came from an unusual mound-like feature located in an animal byre (or kraal). The small stone-and-soil mound lay on top of a gravel floor, interspersed with ceramic sherds, bones, ostrich eggshell beads and ash. The mound was cordoned off, which suggests some importance, perhaps forming part of a religious ceremony (Hanisch, 1980: 308e318). The bone artefacts were examined using low- and high-power microscopy. The methodological approach that we followed has been described in detail previously (e.g., Bradfield, 2015; Evora, 2015). Initial observation of manufacturing and use-wear features was performed using a Celestron USB microscope while advanced characterisation of features was achieved with Olympus SZX16 and BX51M light microscopes. Bright- and Dark-Field illumination was used, as well as the ultra-violet and polarising and cross-polarising capabilities. The artefacts were observed under magnification ranges of between 10 and 200, and use-wear features described under magnifications typically ranging from 50 to 100. Anthropogenic modifications were differentiated from natural ones on the basis of previously published criteria (e.g., Brain, 1981; Lyman, 1994; Fisher, 1995). 5. Results The K2 artefact is an extensively modified scapula, most likely Bos taurus (cattle) (Fig. 2). A substantial portion of this modification is the result of porcupine gnawing, which is responsible for the removal of some of the scapula protuberances. Interspersed among this damage are numerous chisel and snag marks, which were created by a metal blade (Fig. 2G). Striations consistent with scraping the surface of the bone with a metal blade (see Bradfield, 2015) are also apparent in places (Fig. 2C, E). The sides of the tool have been further modified by perpendicular grinding against a coarse abrasive surface e probably sandstone (Fig. 2B). The sharpened distal edge of the tool displays some polishing visible to the naked eye and the sides of the artefact display well-developed polish consistent with prolonged fricative contact against skin or hide (Fig. 2A; cf. Bradfield, 2015). Their placement on the tool suggests they derived from gripping the tool with a hand or from leather lashings to secure the bone to a haft. Patches of reddishbrown discolouration are present on the proximal half of the ventral surface of the tool and coincide with intensive flattening of the bone micro topography (Fig. 2O). Fine parallel and oblique striations are visible within the flattened areas consistent with contact against a hard, compact material such as wood or shell (Fig. 2N; cf. Griffitts and Bonsall, 2001; Bradfield, 2015). The distal section of the ventral surface of the tool, towards the edge, displays a more invasive type of polish and much evidence of sediment abrasion (Fig. 2HeK). The dorsal surface displays the same striations consistent with sediment abrasion, but less pronounced polish (Fig. 2F). Sediment abrasion wear is confined to the distal portion of the tool. The edge of the tool, on both dorsal and ventral surfaces, displays invasive rounding and smoothing, with numerous fine parallel striations interspersed with occasional long, fine, oblique striations (Fig. 2D). This wear is consistent with contact against a soft material but differs in some respects from that produced by contact with skin, namely in the number and orientation of the striations (cf. Bradfield, 2015). The first tool from Pont Drift is made from a medium-sized bovid (i.e., sheep/goat or impala size) scapula and has been modified in the same way as the K2 specimen (Fig. 3). Shaping was
Fig. 2. Modified scapula implement from K2 (W/289). A) invasive, homogenous polish with fine striations indicative of contact against skin or sinew; B) Striations produced via coarse grinding covered by skin-derived polish; C) grouped striations produced by scraping with a metal blade; D) ambiguous wear on the blade edge, consistent with contact against skin or starchy tubers; E) grouped striations produced by scraping with a metal blade; F) assorted striations consistent with sediment abrasion; G) chisel mark produced by a metal blade; H-K) assorted striations and bright polish consistent with fine-grained sediment abrasion; L) porcupine tooth gnawing; M) shallow parallel striations suggestive of contact with silica-rich plant material; N-O) flattened high surfaces and bright polish indicative of fricative contact with woody material. Scale bars represent 200 mm.
130
J. Bradfield, A.R. Antonites / Quaternary International 472 (2018) 126e134
Fig. 3. Modified scapula implement from Pont Drift (TPD 4612). A) homogenous smoothing along the bone ridges, indicative of contact with skin; B) multi-directional groups of coarse striations indicative of sediment abrasion with similarly-sized particles; C) grouped striations produced by scraping with a metal blade; D) edge damage, assorted striations and bright polish consistent with fine-grained sediment abrasion; E-F) tightly-grouped patches of short directional sharp striations, suggestive of wear left by fibrous plant bindings. Scale bars represent 200 mm.
achieved by grinding against an abrasive surface and scraping with a metal blade (Fig. 3C), although chisel and snag marks are absent on this specimen. There is evidence of heavy-duty utilisation along the distal edge of the tool, similar to the K2 specimen, which may have caused the damage on the dorsal surface. The micro-wear along the edge of the tool presents as homogenous invasive rounding and smoothing with numerous parallel striations, orientated parallel to the long axis of the tool, interspersed with occasional long, thin oblique striations (Fig. 3D). There is also some evidence of sediment abrasion, but less pronounced than on the K2 specimen. The degree of polishing along the sides of the tool is much less pronounced than the K2 specimen, but is also consistent with contact with skin or hide (Fig. 3A). In the middle of the ventral surface of the tool are several areas of tightly-grouped patches of short, thin and deep parallel striations (Fig. 3EeF). These traces are suggestive of wrapping or binding with fibrous vegetal lashings. The second implement from Pont Drift is made from a portion of a large bovid long bone (Fig. 4) and has been extensively modified on the dorsal and ventral surfaces. Like the previous two tools, the primary method of manufacture was whittling with a metal blade
as is evident by the presence of snag marks and metal-induced scraping striations (Fig. 4FeG). Both surfaces of the tool are discoloured, although the discolouration is more intense on the dorsal surface. It is clear from the manufacturing traces that the discolouration occurred prior to shaping. The general surface quality of the bone is not as well preserved as the two scapula implements, and it seems likely that this tool underwent some degree of weathering in the past. There are many chips and notches along the distal edge of the tool consistent with heavy-duty work. On both surfaces of the tool, near the distal edge, are many randomly orientated striations of varying widths (Fig. 4D, E, J), consistent with sediment abrasion. There is a stronger polish on the dorsal surface and the striations also have a diagonal slant (Fig. 4E). In the middle of the tool, on the dorsal surface, are patches of bright polish, which are limited to the high point bone topography, resembling prolonged contact with soft silica-rich plant material (Fig. 4AeB; cf. Bradfield, 2015). Traces resembling plant contact are also present on the ventral surface (Fig. 4I), some of which are grouped sets of fine parallel striations (Fig. 4K) probably caused by monocotyledonous vegetal fibres. Overlaying some of the flattened,
J. Bradfield, A.R. Antonites / Quaternary International 472 (2018) 126e134
131
Fig. 4. Heavily-modified long bone implement from Pont Drift (TPD 4611). A-B) patches of bright polish resembling prolonged contact with soft silica-rich plant material; C-E) sediment abrasion wear and polish; F-G) snag marks and striations caused by scraping with a metal blade; H) large invasive area of evenly-spaced parallel striations suggestive of fricative bone-on-bone contact; I) medium polish with shallow parallel striations resembling contact with a plant-based material; J) sediment abrasion wear; K) tightly-grouped patches of short directional sharp striations, suggestive of wear left by large groups of plant fibres. Scale bar represents 200 mm.
132
J. Bradfield, A.R. Antonites / Quaternary International 472 (2018) 126e134
high-point polish on the ventral surface of the tool are many long, deep, parallel striations similar to those produced by contact with another piece of bone (Fig. 4H; cf. Bradfield, 2015). 6. Discussion MacEachern (2005) has previously critiqued the notion that the replacement of lithic and organic tools by iron was the most important element in the transition from the Neolithic to the Iron Age. Indeed, the adoption of iron was more gradual and piecemeal and probably linked to changing crop cultivation (MacEachern, 2005). It has been shown that in some parts of the world changes in land-use and intensification are linked to changes in earthmoving implements (Xie, 2014). In particular, the soil properties, such as texture and compactness, will influence the durability and therefore the choice of raw materials selected to make earthmoving implements (Xie, 2014). For example, scapula hoes have been found to work best in soft soils, whereas compact soils require iron hoes to penetrate (Xie, 2014). At some Iron Age sites in South Africa, such as K2 and Mapungubwe, bone tools are found in abundance (Voigt, 1983; Antonites et al., 2016), and bone continued to be used as a raw material among Bantu-speaking agriculturalists well into the 20th century (Shaw and van Warmelo, 1974; Davison, 1975). The overall shape of the implements examined in this paper is suitable for a range of functions, and these tools could have been used to: process plant foods, prepare hides for leather working, till soil for agriculture, hunt subterranean prey, mine mineral ore, cut grasses, burnish pre-kilned ceramic vessels or to adze wood. Given the combination of use-trace evidence found on these tools we consider their most probable primary function to be agricultural hoes, although additional uses cannot be discounted. Despite the heavy presence of porcupine damage on the K2 artefact, the two scapula tools were manufactured in the same way and show almost identical modifications as scapula hoes from North America and China (cf. Griffitts, 2006; Xie, 2014). Porcupine-damaged tools are not unusual (e.g., Dart, 1958; Zhang et al., 2016), and may in the event even be preferred as pre-forms of the desired shape. The abundance and variety of use-traces on the bone surface, including evidence of a metal implement, supports the anthropic function of this tool. The implement from K2 displays convincing evidence of having been hafted via a wooden beam, which was fastened to the ventral surface of the scapula with sinew or leather thongs. This piece of wood extended from the proximal bulge to midway down the blade. The discolouration on the ventral surface may be due to organic acids leaching out of the wood (Stokes, 2016). A few microscopic traces of ochre were also observed in this region of the tool e possibly the remnants of a mastic paste. The Pont Drift
scapula appears to have been bound with fibrous vegetal cords across the middle of the blade, and there is evidence that sinew or skin wrappings were additionally used across the sides and top of the tool. These were presumably to hold a handle in place, although, unlike on the K2 scapula, direct evidence for this is lacking. Wood would have been the most likely choice for a handle and may have been used together with resin-based mastic, just as it was in the nineteenth century (e.g. Clark, 1959). The evidence on the Pont Drift long bone tool is less easy to interpret. The overall shape and manufacturing technique matches those of the scapula tools, but there is no clear evidence of hafting. There is evidence of bone-on-bone contact on the ventral surface, but this is isolated and insufficient to say whether this represents incidental contact, use or that the tool was hafted onto another piece of bone. The surface of this tool is less well preserved than the others, which might explain the absence of recognisable hafting traces, but given the abundance of other traces on the tool surface, we consider that this tool was probably never hafted. Wear traces suggestive of sediment abrasion are present on the distal edge of all three tools. Together with the chips along the edges and invasive polish we interpret this wear as signalling use in soft, loose, fine-grained sediments (cf. Xie, 2014: Fig. 3.19). Reynard (2013) has explored the effects of trampling on the formation of coarse striations and polish on bone tools. We discount the possibility that trampling was responsible for the sediment abrasion on our tools for two reasons: 1) the abrasion wear is confined to the distal portions of the tools; and 2) the sediment abrasion on both Pont Drift tools is uni-directional and slightly angled suggesting anthropic use (sensu Semenov, 1964). The combination of all the use-wear traces taken together supports our interpretation of anthropogenic implements. Along portions of the blade are patches of ambiguous wear. Experimental processing of African potatoes (Hypoxis hemerocallidea) and leather show that they produce almost indistinguishable traces (Fig. 5). Coupled with the sediment abrasion and inferred method of hafting, however, we consider leather working to be unlikely. It is probable that the three tools were used as hoes to harvest underground tubers and cultivate small garden crops, if not also to till soil for seed planting and weeding. Scapula hoes are known to have been used in ad hoc activities additional to tilling and cultivating (Griffitts, 2006; Xie, 2014), and it is possible that the same was done in South Africa. The use-wear associated with chopping squashes, for instance, is slow to develop and not very diagnostic (Griffitts, 2006), but may account for some of the larger chips along the distal edge of the implements. That iron was available and used during the period under study is well known (e.g., Calabrese, 2000; Miller, 2001), not least because the three bone tools themselves show evidence of having been fashioned, at least in part, by a metal blade. During the K2 and
Fig. 5. Some materials can produce almost identical use-wear features; distinguishing between them can be challenging. Here we show similar polish and striations produced experimentally by processing leather and an African potato (Hypoxis hemerocallidea) for approximately 30 min with a bone tool. Micrographs are taken at 100 magnification. Scale bars represent 200 mm.
J. Bradfield, A.R. Antonites / Quaternary International 472 (2018) 126e134
Transitional K2 periods, the production and distribution of metal objects was restricted and mainly concentrated at the elite settlements of K2 and Mapungubwe (Calabrese, 2000). The use of bone hoes may thus support the argument that metal objects were not widely accessible to all members of society (even at K2), or that iron was not generally used for utilitarian objects like hoes. On the other hand, the value of iron may imply more recycling and re-use of objects over time, resulting in the relatively rare occurrence of iron hoes in Limpopo Valley assemblages (Miller, 2002). The use of bone hoes introduces an interesting perspective on the dynamic between male and female associations related to tool use. Although hoe cultivation is historically associated with women, and iron production with men (e.g., Stayt, 1968; Shaw, 1974), this dichotomy may have been more fluid in the archaeological past (see Chirikure, 2007; Chirikure et al., 2015). Bone, on the other hand, does not appear to have had any specific gendered associations. Recent ethnographies, for example, document activities in which both women and men made and used bone tools (Du Toit, 1968; Davison, 1975; ; Shaw, 1974). The use-traces on the three bone tools examined here do not support the notion that they might have been used as apotropaic objects (sensu Gunda, 1976), but rather that they served a utilitarian function. The depositional context of the Pont Drift long bone hoe, however, suggests that it may have moved from the utilitarian to the ritual sphere (sensu Whitehouse, 1996; Gosden and Marshall, 1999). Whether its position in the mound-like feature was purely symbolic (cf. Du Toit, 1968), or it formed part of the ritual performance involved in the construction thereof, remains unknown. 7. Conclusion This paper presents the first recognised evidence of possible bone hoes in South Africa. The use-wear on these tools is consistent with that observed on bone hoes by other analysts (e.g. Griffitts, 2006; Xie, 2014). Our findings add to the growing understanding of the complexity of bone tool production among southern African Iron Age agriculturalists between the 10th and 13th centuries AD (Antonites et al., 2016), as well as the global distribution of bone hoe technology. The economy of the Limpopo Valley communities was based on mixed farming, with the focus being on cereals like sorghum, millets and legumes (Huffman, 2007). These crops, which grow well in warm, loose, well-drained soils, can be cultivated in small plots and require constant attendance in terms of weeding (Directorate of Plant Production, 2010). In other words, soil conditions conducive to producing the wear traces seen on the distal blades of the bone implements. In order to meet the consumption requirements of the community living at K2 during its zenith, it has been estimated that approximately 634 hectares of ground would need to have been under cultivation (Murimbika, 2006). Bone hoes may have been used during tillage of small plots or they may have been used to gather underground foods such as melons, tubers or rhizomes during foraging trips. Whatever their primary intention, bone hoes appear to have been a short-lived technological innovation, eventually giving way to their metal counterparts (Herbert, 1996). By extending the ethnography on gendered associations back to the Middle Iron Age, the recovery of the scapula hoes from domestic or household contexts links these objects to women and offers a glimpse into women's activities during the K2 and TK2 periods. Unfortunately, the use-wear does not allow us to say who made these implements, but given the cache of hafted bone arrows found at K2 (Antonites et al., 2016) it appears likely that both men and women worked and used bone during this period. The fact that only three such implements have been recovered may point to a
133
less intensive role in agricultural production, but it is possible that similar tools have not been recognised by faunal analysts. A worthwhile pursuit would be to revisit some of these collections. Acknowledgements For access to the artefacts we wish to thank Johnny van Schalkwyk and Frank Teichert at the Ditsong National Museum of Cultural History, as well as Thembi Russell and Faye Lander of the Archaeology Collections, University of the Witwatersrand. Marlize Lombard is thanked for permission to use the microscopes in the Micro-Tracks research facility, University of Johannesburg. Jan Boeyens kindly commented on a previous draft of the manuscript. JB's research is supported by grants from the Oppenheimer Memorial Trust and the National Research Foundation. References Anderson, M.S., 2009. Marothodi: the Historical Archaeology of an African Capital. Atikkam, Woodford. Antonites, A.R., Bradfield, J., Forssman, T., 2016. Technological, functional and contextual aspects of the K2 and Mapungubwe worked bone industries. Afr. Archaeol. Rev. 33, 437e463. Baron, J., Diakowski, M., Stolarczyk, T., 2016. Bone and antler artefacts from an 8-5th century BC settlement at Grzybiany, South-Western Poland. In: Vitezovi c, S. (Ed.), Close to the Bone: Current Studies in Bone Technologies. Institute of Archaeology, Belgrade, pp. 26e45. Bell, R., Cross, G., 1980. Oklahoma Indian Artefacts: Contribution from the Stovall Museum. University of Oklahoma Press, Oklahoma. Bradfield, J., 2015. Use-trace analysis of bone tools: a brief overview of four methodological approaches. South Afr. Archaeol. Bull. 70, 3e14. Brain, C.K., 1981. The Hunters or the Hunted? an Introduction to African Cave Taphonomy. University of Chicago Press, Chicago. Bryant, A., 1949. The Zulu People: as They Were before the White Man Came. Shuter & Shooter, Pietermaritzburg. Calabrese, J.A., 2000. Metals, ideology and power: the manufacture and control of materialised ideology in the area of the Limpopo-Shashe confluence, c. AD 900 to 1300. South Afr. Archaeol. Soc. Goodwin Ser. 8, 100e111. Campana, D., 1989. Natufian and Protoneolithic Bone Tools: the Manufacture and Use of Bone Implements in the Zagros and the Levant. BAR International Series 494, Oxford. Chapman, J., 1864. Travels in the Interior of South Africa. William Clowes and Sons., London. Childs, S.T., Herbert, E., 2005. Metallurgy and its Consequences. In: Stahl, A. (Ed.), African Archaeology: a Critical Introduction. Blackwell Publishing, Oxford, pp. 276e300. Chirikure, S., 2007. Iron production and its position in Iron Age communities of southern Africa. J. Soc. Archaeol. 5, 74e102. Chirikure, S., 2015. Metals in Past Societies: a Global Perspective on Indigenous African Metallurgy. SpringerBriefs in Archaeology. Springer, New York. Chirikure, S., Hall, S., Rehren, T., 2015. When ceramic sociology meets material science: sociological and technological aspects of crucibles and pottery from Mapungubwe, southern Africa. J. Anthropol. Archaeol. 40, 23e32. Clark, J.D., 1959. The Prehistory of Southern Africa. Penguin Books, London. Culwick, A.T., Culwick, G.M., 1934. The functions of bride-wealth in Ubena of the Rivers. Afr. J. Int. Afr. Inst. 7, 140e159. Curwin, E., 1926. On the use of scapulae as shovels. Sussex Archaeol. Collect. 67, 139e145. Curwin, E., 1937. The Archaeology of Sussex. Methuen & Co., London. d'Errico, F., 1993. Criteria for identifying utilised bone: the case of the Cantabrian “tensors”. Curr. Anthropol. 34, 299e309. Dart, R., 1958. Bone tools and porcupine gnawing. Am. Anthropol. 60, 715e724. Davison, P., 1975. Some Nguni Crafts: the uses of horn, bone and ivory. Ann. South Afr. Mus. 70, 79e155. Directorate of Plant Production, 2010. Sorghum Production Guidelines. Department of Agriculture. Forestry and Fisheries, Pretoria. €le skeppinge van die BaPhalaborwa in Du Toit, A.P., 1968. Die plek van die materie hul kultuur. PhD dissertation. University of South Africa, Pretoria. Earthy, E.D., 1933. Valenge Women: the Social and Economic Life of the Valenge Women of Portuguese East Africa. Oxford University Press, London. Evora, M., 2015. Use-wear methodology on the analysis of osseous industries. In: Marreiros, J.M., Gibaja Bao, J.F., Ferreira Bicho, N. (Eds.), Use-wear and Residue Analysis in Archaeology. Springer, Germany, pp. 159e170. Fisher, J., 1995. Bone surface modifications in zooarchaeology. J. Archaeol. Method Theory 2, 7e68. Gardner, G.A., 1963. Report on Excavations at Mapungubwe and Bambandyanalo in the Transvaal from 1935e1940, Vol. II. Van Schaik, Pretoria. Garlake, P.S., 1973. Great Zimbabwe. Thames and Hudson, London. Gosden, C., Marshall, Y., 1999. The cultural biography of objects. World Archaeol. 31
134
J. Bradfield, A.R. Antonites / Quaternary International 472 (2018) 126e134
(2), 169e178. Griffitts, J., Bonsall, C., 2001. Experimental determination of the function of antler and bone ‘bevel-ended tools’ from prehistoric shell middens in western Scotland. In: Choyke, A., Bartosiewicz, L. (Eds.), Crafting Bone: Skeletal Technologies through Time and Space. BAR International Series 937, Oxford, pp. 207e220. Griffitts, J., 2006. Bone Tools and Technological Choice: Change and Stability on the Northern Plains. Unpublished PhD thesis. University of Arizona, Texas. Gunda, B., 1976. More on iron hoes in Africa. Curr. Anthropol. 17, 172. Hanisch, E.O.M., 1980. An Archaeological Interpretation of Certain Iron Age Sites in the Limpopo/Shashi Valley. Unpublished MA dissertation. University of Pretoria, Pretoria. Herbert, H., 1996. Metals and power at Great Zimbabwe. In: Pwiti, G., Soper, R. (Eds.), Aspects of African Archaeology: Papers from the 10th Congress of the Pan African Association for Prehistory and Related Studies. University of Zimbabwe Publications, Harare, pp. 641e655. Huffman, T., 2000. Mapungubwe and the origins of the Zimbabwe culture. South Afr. Archaeol. Soc. Goodwin Ser. 8, 14e29. Huffman, T., 2007. Handbook to the Iron Age: the Archaeology of Pre-colonial Farming Societies in Southern Africa. University of KwaZulu-Natal Press, Pietermaritzburg. Hunt, D.R., 1931. An account of the Bapedi. Bantu. Stud. 5 (1), 275e326. Kuper, A., 1982. Wives for Cattle: Bridewealth and Marriage in Southern Africa. Routledge, London. Lane, P., 2013. Trajectories of pastoralism in northern and central Kenya: an overview of the archaeological and environmental evidence. In: Bollig, M., Schnegg, M., Wotzka, H.-P. (Eds.), Pastoralism in Africa: Past, Present, and Futures. Berghahn, New York, pp. 104e144. ra, I., 2007. Methods, means and results when studying European Legrand, A., Side bone industries. In: Gates St-Pierre, C., Walker, R. (Eds.), Bones as Tools: Current Methods and Interpretations in Worked Bone Studies. BAR International Series 1622, Oxford, pp. 67e78. Livingstone, D., 1851. Missionary Travels and Researches in South Africa. John Murray, London. Lombard, M., Wadley, L., Deacon, J., Wurz, S., Parsons, I., Mohapi, L., Swart, J., Mitchell, P., 2012. South African and Lesotho Stone Age sequence updated (i). South Afr. Archaeol. Bull. 67, 123e144. Luik, H., 2011. Material, technology and meaning: antler artefacts and antler working on the eastern shore of the Baltic Sea in the Late Bronze Age. Est. J. Archaeol. 15, 32e55. Lupo, K., Schmitt, D., 2002. Upper Paleolithic net-hunting, small prey exploitation and women's work effort: a view from the ethnographic and ethnoarchaeological record of the Congo Basin. J. Archaeol. Method Theory 9, 147e179. Lyman, R., 1994. Vertebrate Taphonomy. Cambridge University Press, Cambridge. MacEachern, S., 2005. Two thousand years of West African history. In: Stahl, A. (Ed.), African Archaeology: a Critical Introduction. Blackwell Publishing, Oxford, pp. 441e466. Matthews, W., 1877. Ethnography and philology of the Hidatsa Indians. Reissue 1969 Plains Anthropologist, 14, pp. 177e252. Meier, J., 2013. More fun than games? An experimental study of worked bone astragali from two Middle Bronze Age Hungarian sites. In: Choyke, A., O'Connor, S. (Eds.), From These Bare Bones: Raw Materials and the Study of Worked Osseous Objects. Oxbow Books, Oxford, pp. 166e173.
Meiring, A.J., 1952. Polished-edge bone implements. Navor. Nas. Mus. 1, 31e34. Meyer, A., 1998. The Archaeological Sites of Greefswald: Stratigraphy and Chronology of the Sites and a History of Investigations. University of Pretoria, Pretoria. Miller, D., 2001. Metal assemblages from Greefswald areas K2, Mapungubwe Hill and Mapungubwe southern terrace. South Afr. Archaeol. Bull. 56 (174), 83e103. Miller, D., 2002. Smelter and smith: iron Age metal fabrication technology in southern Africa. J. Archaeol. Sci. 29 (10), 1083e1131. Moffett, A., Maggs, T., Van Schalkwyk, J., 2017. Breaking ground: hoes in precolonial South Africadtypology, medium of exchange and symbolic value. Afr. Archaeol. Rev. https://doi.org/10.1007/s10437-017-9247-1. Mumford, W.B., 1934. The hehe-bena-sangu peoples of East Africa. Am. Anthropol. 36, 203e222. Murimbika, M., 2006. Sacred Powers and Rituals of Transformation: an Ethnoarchaeological Study of Rainmaking Rituals and Agricultural Productivity during the Evolution of the Mapungubwe State, AD 1000 to AD 1300. Unpublished PhD dissertation. University of the Witwatersrand, Johannesburg. Phillipson, D., 2005. African Archaeology, third ed. Cambridge University Press, Cambridge. Plug, I., 2000. Overview of Iron Age fauna from the Limpopo Valley. South Afr. Archaeol. Soc. Goodwin Ser. 8, 117e126. Reynard, J., 2013. Trampling in coastal sites: an experimental study on the effects of shell on bone in coastal sediment. Quat. Int. 330, 156e170. Schoeman, M., 2013. Excavating ephemeral remains of life in a time of witchcraft: new insights into the eighteenth and nineteenth-century occupations at Leokwe and Nyindi hills in the Shashe-Limpopo confluence area, South Africa. Int. J. Hist. Archaeol. 17, 613e631. Semenov, S., 1964. Prehistoric Technology. Adams and Dart, Bath. Shaw, M., 1974. Material culture. In: Hammond-Tooke, W.D. (Ed.), The Bantuspeaking Peoples of Southern Africa. Routledge & Keagan Paul, London, pp. 85e131. Shaw, M., van Warmelo, N., 1974. The material culture of the Cape Nguni. Ann. South Afr. Mus. 58, 103e214. Stayt, H.A., 1968. The Bavenda, second ed. Frank Cass, London. Stokes, P., 2016. A new interpretation of post-medieval bone scoops from the foreshore of the river Thames in London. In: Vitezovi c, S. (Ed.), Close to the Bone: Current Studies in Bone Technologies. Institute of Archaeology, Belgrade, pp. 324e337. Strong, D., 1933. The Plains culture area in the light of archaeology. Am. Anthropol. 35, 271e287. Voigt, E., 1983. Mapungubwe: an Archaeological Interpretation of an Iron Age Community. Transvaal Museum Monograph 1. Transvaal Museum, Pretoria. Wang, W., 1991. Bone si and related issues in the Hemudu culture. Southeast Cult. 1, 240e242. Whitehouse, R., 1996. Ritual objects. Archaeological joke or neglected evidence? In: Wilkins, J. (Ed.), Approaches to the Study of Ritual: Italy and the Ancient Mediterranean. Accordia Research Centre, London, pp. 9e30. Xie, L., 2014. Early To Middle Holocene Earth-Working Implements And Neolithic Land-Use Strategies On The Ningshao Plain, China. Unpublished doctoral thesis. University of Arizona. Zhang, S., d'Errico, F., Backwell, L., Zhang, Y., Chen, F., Gao, X., 2016. Ma'anshan cave and the origin of bone tool technology in China. J. Archaeol. Sci. 65, 57e69.
Contents lists available at ScienceDirect
Quaternary International journal homepage: www.elsevier.com/locate/quaint
Bone hoes from the Middle Iron Age, Limpopo Province, South Africa Justin Bradfield a, b, *, Annie R. Antonites c a
Evolutionary Studies Institute, University of the Witwatersrand, Johannesburg 2050, South Africa School of Geography, Archaeology and Environmental Studies, University of the Witwatersrand, Johannesburg 2050, South Africa c Department of Anthropology and Archaeology, University of South Africa, Pretoria, South Africa b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 13 January 2017 Received in revised form 10 July 2017 Accepted 26 October 2017 Available online 4 November 2017
This paper presents the first recognised evidence of bone hoes in South Africa. Two bovine scapulae and a portion of a long bone show use-trace evidence that supports our interpretation as ground-working implements. The scapulae were probably hafted onto wooden handles using a combination of plant fibres and sinew, whereas the tool made from the long bone appears not to have been hafted. Bone hoes represent a short-lived technological innovation, although the reasons to account for this remain speculative. The recognition of these agricultural implements poses interesting questions about the extent and variety of bone working among Iron Age agriculturalists in the Limpopo Valley during the 10th e 13th centuries AD, and potentially also about the nature of women's work in these communities. © 2017 Elsevier Ltd and INQUA. All rights reserved.
Keywords: South Africa Iron Age Limpopo Valley Scapula hoes Agricultural implements Farming communities Worked bone Use-traces
1. Introduction Bone tools made from animal scapulae (shoulder blades) are found among pre-industrial societies in most parts of the world (see Griffitts, 2006; Xie, 2014). Various interpretations have been proffered for these tools, but most agree that they were used for moving earth. The precise function ranges from construction tools for digging trenches and pits (e.g. Curwin, 1926, 1937; Griffitts, 2006) to agricultural implements (e.g. Strong, 1933; Bell and Cross, 1980; Griffitts, 2006; Xie, 2014). Among the Plains Indians of the central United States for instance, bison scapulae were used in historic times for cultivating small garden crops like maize, beans and squash (Strong, 1933; Bell and Cross, 1980). Although perhaps less widely used for this purpose than wood, stone and metal implements (Xie, 2014), scapulae have a high tensile strength making them ideal for heavy-duty work such as digging, and thus more economical than their stone and metal counterparts in terms of manufacturing time (Griffitts, 2006; Xie, 2014). In southern Africa broad-blade, spatula-shaped bone tools have
* Corresponding author. Evolutionary Studies Institute, University of the Witwatersrand, Johannesburg 2050, South Africa. E-mail address: jbradfi[email protected] (J. Bradfield). https://doi.org/10.1016/j.quaint.2017.10.028 1040-6182/© 2017 Elsevier Ltd and INQUA. All rights reserved.
been recovered from numerous sites along the South Coast (Meiring, 1952). Where recovered in situ these tools have been placed within the Final Later Stone Age technocomplex; roughly the last 4000 years (Meiring, 1952; cf. Lombard et al., 2012). These tools are made from ribs or slivers of long bone and the edges ground into shape against an abrasive surface (Meiring, 1952). No proper use-trace studies of these artefacts have been conducted yet, and their precise function remains elusive. Proffered interpretations range from shell fish scoops, melon knives for shredding vegetables, meat mattocks, and skin scrapers (Clark, 1959) to hand adzes for wood working (Meiring, 1952). Further inland three similar tools, two of which are fashioned from bovine scapulae, have been recovered from Iron Age contexts dating between c. 1000 CE - 1250 CE in the Limpopo River Valley.1 Voigt (1983: 110) previously described one of these tools from K2 as a meat mattock for the preparation of skins, based on the extensive visible polish covering the artefact surface. The other two artefacts, one made from a scapula and the other from a piece of a large bovid
1 It is possible that more bone hoe-like objects exist but have not been recognised as tools. Voigt (1983) describes two other similarly-shaped tools from K2, one of which she interpreted as a chopping board. Neither tool could be located for analysis here.
J. Bradfield, A.R. Antonites / Quaternary International 472 (2018) 126e134
long bone, derive from the roughly contemporaneous site of Pont Drift (Fig. 1). Bone working is not unusual among Iron Age communities of this region (Hanisch, 1980; Voigt, 1983; Antonites et al., 2016). K2 and the nearby 13th century capital of Mapungubwe have well developed bone-working traditions, yet the function of spatulashaped tools remains poorly understood (Voigt, 1983). Understanding what these tools were used for may shed light on the diversity of bone tool technology among Bantu-speaking agriculturalists in the Limpopo Valley during this period. The aim of the present paper is to formally describe the use-trace features of three tools found at K2 and Pont Drift. The extensive use-wear on these artefacts, which are unique to the South African Iron Age, warrants closer scrutiny into their precise function.
2. Background
127
contemporaneous settlements in South Africa's middle Limpopo Valley e K2 and Pont Drift (Fig. 1). The area is best known for the emergence of class distinction and political centralisation in the 13th century AD (Huffman, 2000, 2007). K2 was a large ‘elite’ settlement that asserted socio-political and economic influence over a wide region (c. 1000 CE e 1200 CE, ‘K2 period’). Intermittent excavations between the 1930s and 1990s revealed several household areas and domestic middens surrounding a large central animal enclosure and communal midden. The K2 inhabitants kept livestock (cattle, sheep and goats), cultivated sorghum, millet and legumes in the floodplain, and occasionally hunted and gathered wild resources. They manufactured a range of crafts and traded ivory and other raw materials for glass beads with Arab traders on the East African coast (Huffman, 2007; Gardner, 1963; Meyer, 1998; Miller, 2001; Voigt, 1983). Sometime during the early ‘Transitional K2 period’ (1200 CE e 1250 CE), K2 was abandoned and its community relocated to the nearby site of Mapungubwe Hill e a move marked by the physical separation between elites and commoners
The bone implements described here come from two roughly
Fig. 1. Map showing the geographical region of the study area.
128
J. Bradfield, A.R. Antonites / Quaternary International 472 (2018) 126e134
and a shift in the political economy towards a more centralised system. Pont Drift is located about 25 km west of K2 and lies about 1.5 km south of the Limpopo River. Excavations in the late 1970s revealed an extensive vertical deposit, with the earliest occupation dating to the ‘Zhizo period’ (c. 850 CE e 1000 CE). A K2 period animal byre and a thick midden overlay the Zhizo layers, followed by a series of Transitional K2 hut floors and courtyard surfaces (Hanisch, 1980; Huffman, 2007). Although subsistence and material culture patterns are similar to those at K2, the assemblages are typical of ‘commoner’ settlements in each occupation phase (Hanisch, 1980; Huffman, 2007; Plug, 2000). In sub-Sharan Africa, iron hoes are common items among the various Bantu-speaking agriculturalists and were used for a range of excavation tasks. Apart from agriculture, iron hoes were used to hunt for subterranean prey, like rats and hares (Lupo and Schmitt, 2002), to excavate mineral ore (Chirikure, 2015), and as exchange items (Moffett et al., 2017). From about the 6th century AD tanged iron hoes were a common implement among the agriculturalist groups south of the Zambezi River (Gunda, 1976), although most extant examples are far younger (Moffett et al., 2017). Iron hoes appear to have replaced earlier chipped stone counterparts, which have been found in East and West Africa between 7 ka and 1.5 ka ago (Phillipson, 2005; Lane, 2013). Certainly by the 19th century hoes were being made almost exclusively from iron (e.g. Livingstone, 1851; Chapman, 1864; Gunda, 1976), although wooden digging sticks and stone hammers continued to be used (Childs and Herbert, 2005). In rare instances when, for whatever reason, iron was difficult to obtain, wooded hoes were used (Earthy, 1933; Shaw and van Warmelo, 1974; Moffett et al., 2017). An interesting feature about iron hoes is their symbolic male/ female dichotomy. Iron working was regarded as a male activity and was imbued with symbolism pertaining to procreation (Childs and Herbert, 2005; Chirikure, 2015). While only men were allowed to smelt and forge iron, fully-forged hoes were symbolically associated with women, as the ones responsible for the bulk of agricultural work (Chapman, 1864; Mumford, 1934; Gunda, 1976). Iron hoes also symbolise this link between women and production through its historical importance in marriage transactions in subSaharan Africa (e.g., Culwick and Culwick, 1934; Hunt, 1931; Mumford, 1934; see also Kuper, 1982: 21e25). Among some southern Bantu-speakers, iron hoes were incorporated into ancestor veneration rituals (e.g. Stayt, 1968: 245), and miniature iron hoes were worn by women during ceremonial dances to drive away demons (Gunda, 1976). In addition to the utilitarian, symbolic and ritual aspects of iron hoes, they also served as currency throughout Africa (e.g. Livingstone, 1851; Childs and Herbert, 2005; Chirikure, 2015). The trade in iron was usually controlled by men, again illustrating the fluidity of association attached to iron hoes (Schoeman, 2013). Iron hoes have been recovered from several Iron Age farmer settlements in southern Africa (e.g., Anderson, 2009; Garlake, 1973; Miller, 2002; Schoeman, 2013), and at least seven shape types have been recognised (Moffett et al., 2017). By contrast, the authors are aware of only one reference to bone being used as hoe material among the southern Bantu-speakers. Cattle scapulae apparently were used among the Zulu of the 19th century (Bryant, 1949), though none of these examples seem to have survived to posterity. Bone is regularly used to make hoe blades in other parts of the world (see Griffitts, 2006; Xie, 2014; Baron et al., 2016), and their frequency may even be under-represented as they may sometimes be mistaken for axe heads if no use-trace studies have been conducted (Luik, 2011). Among the Plains Indians of North America scapula hoes were hafted perpendicularly to wooden handles by binding them with sinew (Griffitts, 2006) and continued in use
after metal was easily available (Matthews, 1877). Among the Hemdu peoples of China scapula hoes were used to cultivate rice in soft swampy ground, but, unlike in North America, were gradually replaced by metal (Xie, 2014). 3. Use-wear associated with hoes Use-wear features that develop on bone implements associated with agricultural work have been described in two regions of the world, namely, the United States and China (Griffitts, 2006; Xie, 2014). In both regions large bovid scapulae were modified and used to till the soil. The spine of the scapula would be removed, spongy protuberances near the edges smoothed and the blade trimmed and sharpened to enhance efficacy (Bell and Cross, 1980; Griffitts, 2006; Xie, 2014). Hafted hoes usually penetrate the soil at an angle, which lead to certain concentrations of wear patterns (Semenov, 1964). Experimental studies have shown that microwear patterns vary depending on the conditions and composition of the soil (Xie, 2014). For instance, humic soils may produce striations associated with plant working due to the presence of vegetal fibres in the soil. Striation depth and width depend on the size of sand grains in the soil but are typically orientated diagonally to the working edge (Xie, 2014). Striations will mimic those produced by sediment abrasion (cf. d’Errico, 1993; Lyman, 1994; Fisher, 1995), but may also contain concentrations of smaller evenly distributed parallel striations (Griffitts, 2006; Xie, 2014). Polish will depend on the size of the substrate. A mix of bright polish, flattening and invasive polish has been noted on experimental bone hoes used to till soft, compact, clayey sediments (Griffitts, 2006; Xie, 2014). Xie (2014) considers polish to be the least helpful criteria for identifying earth working, as it is affected by too many variables and can resemble polish produced by dry hides and wood working, depending on the specific texture and consistency of the soil. Both Griffitts (2006) and Xie (2014) note the high incidence of reuse of bone scapula hoes in their respective study areas. Apart from cultivating domestic crops, bone hoes are also suitable for digging pits for post holes, cutting grasses for thatching, and processing skin and bark items (Campana, 1989; Wang, 1991; Griffitts, ra (2007) mention that wood or bone 2006). Legrand and Side scrapers will present with numerous notches and chips along the cutting edge of the tool. Some scapula ‘hoes’ were used by North American Plains Indians to break open squashes subsequent to digging them out of the ground (Griffitts, 2006). Another use for which scapula tools are suitable is for burnishing ceramics. Usewear features will differ depending on whether the clay is wet or dry (Meier, 2013), with dry clay producing numerous long parallel or oblique striations oriented perpendicularly to the edge of the ra, 2007; Meier, 2013). In this instance polish tool (Legrand and Side will be well developed, whereas wet clay produces moderate polish and asymmetric curving striations (Meier, 2013). 4. Methods and materials Three objects form part of our analysis (Table 1). The K2 scapula
Table 1 Provenience and morphometric data of the three artefacts mentioned in the text. Object
Provenience
Dimensions
W/289
K2, B.4.S.3, 12’.28’.30
TPD4612
TPD1/2, Block 2B, Layer 4, Feature 4
TPD4611
TPD1/2, Block 2AA, Layer 10, Feature 1.1
Length: 195 mm Max width: 69 mm Length: 147 mm Max width: 48.5 mm Length: 169 mm Max. width: 54 mm
J. Bradfield, A.R. Antonites / Quaternary International 472 (2018) 126e134
129
(W/289) is from a K2-period domestic context in the western settlement area of the site (Gardner, 1963; Meyer, 1998). The Pont Drift scapula (TPD4612) was excavated from a Transitional K2 household context, consisting of a living hut, smaller storage hut and household refuse (Hanisch, 1980: 237e245). The long bone (TPD4611) is from the preceding K2 period layers and came from an unusual mound-like feature located in an animal byre (or kraal). The small stone-and-soil mound lay on top of a gravel floor, interspersed with ceramic sherds, bones, ostrich eggshell beads and ash. The mound was cordoned off, which suggests some importance, perhaps forming part of a religious ceremony (Hanisch, 1980: 308e318). The bone artefacts were examined using low- and high-power microscopy. The methodological approach that we followed has been described in detail previously (e.g., Bradfield, 2015; Evora, 2015). Initial observation of manufacturing and use-wear features was performed using a Celestron USB microscope while advanced characterisation of features was achieved with Olympus SZX16 and BX51M light microscopes. Bright- and Dark-Field illumination was used, as well as the ultra-violet and polarising and cross-polarising capabilities. The artefacts were observed under magnification ranges of between 10 and 200, and use-wear features described under magnifications typically ranging from 50 to 100. Anthropogenic modifications were differentiated from natural ones on the basis of previously published criteria (e.g., Brain, 1981; Lyman, 1994; Fisher, 1995). 5. Results The K2 artefact is an extensively modified scapula, most likely Bos taurus (cattle) (Fig. 2). A substantial portion of this modification is the result of porcupine gnawing, which is responsible for the removal of some of the scapula protuberances. Interspersed among this damage are numerous chisel and snag marks, which were created by a metal blade (Fig. 2G). Striations consistent with scraping the surface of the bone with a metal blade (see Bradfield, 2015) are also apparent in places (Fig. 2C, E). The sides of the tool have been further modified by perpendicular grinding against a coarse abrasive surface e probably sandstone (Fig. 2B). The sharpened distal edge of the tool displays some polishing visible to the naked eye and the sides of the artefact display well-developed polish consistent with prolonged fricative contact against skin or hide (Fig. 2A; cf. Bradfield, 2015). Their placement on the tool suggests they derived from gripping the tool with a hand or from leather lashings to secure the bone to a haft. Patches of reddishbrown discolouration are present on the proximal half of the ventral surface of the tool and coincide with intensive flattening of the bone micro topography (Fig. 2O). Fine parallel and oblique striations are visible within the flattened areas consistent with contact against a hard, compact material such as wood or shell (Fig. 2N; cf. Griffitts and Bonsall, 2001; Bradfield, 2015). The distal section of the ventral surface of the tool, towards the edge, displays a more invasive type of polish and much evidence of sediment abrasion (Fig. 2HeK). The dorsal surface displays the same striations consistent with sediment abrasion, but less pronounced polish (Fig. 2F). Sediment abrasion wear is confined to the distal portion of the tool. The edge of the tool, on both dorsal and ventral surfaces, displays invasive rounding and smoothing, with numerous fine parallel striations interspersed with occasional long, fine, oblique striations (Fig. 2D). This wear is consistent with contact against a soft material but differs in some respects from that produced by contact with skin, namely in the number and orientation of the striations (cf. Bradfield, 2015). The first tool from Pont Drift is made from a medium-sized bovid (i.e., sheep/goat or impala size) scapula and has been modified in the same way as the K2 specimen (Fig. 3). Shaping was
Fig. 2. Modified scapula implement from K2 (W/289). A) invasive, homogenous polish with fine striations indicative of contact against skin or sinew; B) Striations produced via coarse grinding covered by skin-derived polish; C) grouped striations produced by scraping with a metal blade; D) ambiguous wear on the blade edge, consistent with contact against skin or starchy tubers; E) grouped striations produced by scraping with a metal blade; F) assorted striations consistent with sediment abrasion; G) chisel mark produced by a metal blade; H-K) assorted striations and bright polish consistent with fine-grained sediment abrasion; L) porcupine tooth gnawing; M) shallow parallel striations suggestive of contact with silica-rich plant material; N-O) flattened high surfaces and bright polish indicative of fricative contact with woody material. Scale bars represent 200 mm.
130
J. Bradfield, A.R. Antonites / Quaternary International 472 (2018) 126e134
Fig. 3. Modified scapula implement from Pont Drift (TPD 4612). A) homogenous smoothing along the bone ridges, indicative of contact with skin; B) multi-directional groups of coarse striations indicative of sediment abrasion with similarly-sized particles; C) grouped striations produced by scraping with a metal blade; D) edge damage, assorted striations and bright polish consistent with fine-grained sediment abrasion; E-F) tightly-grouped patches of short directional sharp striations, suggestive of wear left by fibrous plant bindings. Scale bars represent 200 mm.
achieved by grinding against an abrasive surface and scraping with a metal blade (Fig. 3C), although chisel and snag marks are absent on this specimen. There is evidence of heavy-duty utilisation along the distal edge of the tool, similar to the K2 specimen, which may have caused the damage on the dorsal surface. The micro-wear along the edge of the tool presents as homogenous invasive rounding and smoothing with numerous parallel striations, orientated parallel to the long axis of the tool, interspersed with occasional long, thin oblique striations (Fig. 3D). There is also some evidence of sediment abrasion, but less pronounced than on the K2 specimen. The degree of polishing along the sides of the tool is much less pronounced than the K2 specimen, but is also consistent with contact with skin or hide (Fig. 3A). In the middle of the ventral surface of the tool are several areas of tightly-grouped patches of short, thin and deep parallel striations (Fig. 3EeF). These traces are suggestive of wrapping or binding with fibrous vegetal lashings. The second implement from Pont Drift is made from a portion of a large bovid long bone (Fig. 4) and has been extensively modified on the dorsal and ventral surfaces. Like the previous two tools, the primary method of manufacture was whittling with a metal blade
as is evident by the presence of snag marks and metal-induced scraping striations (Fig. 4FeG). Both surfaces of the tool are discoloured, although the discolouration is more intense on the dorsal surface. It is clear from the manufacturing traces that the discolouration occurred prior to shaping. The general surface quality of the bone is not as well preserved as the two scapula implements, and it seems likely that this tool underwent some degree of weathering in the past. There are many chips and notches along the distal edge of the tool consistent with heavy-duty work. On both surfaces of the tool, near the distal edge, are many randomly orientated striations of varying widths (Fig. 4D, E, J), consistent with sediment abrasion. There is a stronger polish on the dorsal surface and the striations also have a diagonal slant (Fig. 4E). In the middle of the tool, on the dorsal surface, are patches of bright polish, which are limited to the high point bone topography, resembling prolonged contact with soft silica-rich plant material (Fig. 4AeB; cf. Bradfield, 2015). Traces resembling plant contact are also present on the ventral surface (Fig. 4I), some of which are grouped sets of fine parallel striations (Fig. 4K) probably caused by monocotyledonous vegetal fibres. Overlaying some of the flattened,
J. Bradfield, A.R. Antonites / Quaternary International 472 (2018) 126e134
131
Fig. 4. Heavily-modified long bone implement from Pont Drift (TPD 4611). A-B) patches of bright polish resembling prolonged contact with soft silica-rich plant material; C-E) sediment abrasion wear and polish; F-G) snag marks and striations caused by scraping with a metal blade; H) large invasive area of evenly-spaced parallel striations suggestive of fricative bone-on-bone contact; I) medium polish with shallow parallel striations resembling contact with a plant-based material; J) sediment abrasion wear; K) tightly-grouped patches of short directional sharp striations, suggestive of wear left by large groups of plant fibres. Scale bar represents 200 mm.
132
J. Bradfield, A.R. Antonites / Quaternary International 472 (2018) 126e134
high-point polish on the ventral surface of the tool are many long, deep, parallel striations similar to those produced by contact with another piece of bone (Fig. 4H; cf. Bradfield, 2015). 6. Discussion MacEachern (2005) has previously critiqued the notion that the replacement of lithic and organic tools by iron was the most important element in the transition from the Neolithic to the Iron Age. Indeed, the adoption of iron was more gradual and piecemeal and probably linked to changing crop cultivation (MacEachern, 2005). It has been shown that in some parts of the world changes in land-use and intensification are linked to changes in earthmoving implements (Xie, 2014). In particular, the soil properties, such as texture and compactness, will influence the durability and therefore the choice of raw materials selected to make earthmoving implements (Xie, 2014). For example, scapula hoes have been found to work best in soft soils, whereas compact soils require iron hoes to penetrate (Xie, 2014). At some Iron Age sites in South Africa, such as K2 and Mapungubwe, bone tools are found in abundance (Voigt, 1983; Antonites et al., 2016), and bone continued to be used as a raw material among Bantu-speaking agriculturalists well into the 20th century (Shaw and van Warmelo, 1974; Davison, 1975). The overall shape of the implements examined in this paper is suitable for a range of functions, and these tools could have been used to: process plant foods, prepare hides for leather working, till soil for agriculture, hunt subterranean prey, mine mineral ore, cut grasses, burnish pre-kilned ceramic vessels or to adze wood. Given the combination of use-trace evidence found on these tools we consider their most probable primary function to be agricultural hoes, although additional uses cannot be discounted. Despite the heavy presence of porcupine damage on the K2 artefact, the two scapula tools were manufactured in the same way and show almost identical modifications as scapula hoes from North America and China (cf. Griffitts, 2006; Xie, 2014). Porcupine-damaged tools are not unusual (e.g., Dart, 1958; Zhang et al., 2016), and may in the event even be preferred as pre-forms of the desired shape. The abundance and variety of use-traces on the bone surface, including evidence of a metal implement, supports the anthropic function of this tool. The implement from K2 displays convincing evidence of having been hafted via a wooden beam, which was fastened to the ventral surface of the scapula with sinew or leather thongs. This piece of wood extended from the proximal bulge to midway down the blade. The discolouration on the ventral surface may be due to organic acids leaching out of the wood (Stokes, 2016). A few microscopic traces of ochre were also observed in this region of the tool e possibly the remnants of a mastic paste. The Pont Drift
scapula appears to have been bound with fibrous vegetal cords across the middle of the blade, and there is evidence that sinew or skin wrappings were additionally used across the sides and top of the tool. These were presumably to hold a handle in place, although, unlike on the K2 scapula, direct evidence for this is lacking. Wood would have been the most likely choice for a handle and may have been used together with resin-based mastic, just as it was in the nineteenth century (e.g. Clark, 1959). The evidence on the Pont Drift long bone tool is less easy to interpret. The overall shape and manufacturing technique matches those of the scapula tools, but there is no clear evidence of hafting. There is evidence of bone-on-bone contact on the ventral surface, but this is isolated and insufficient to say whether this represents incidental contact, use or that the tool was hafted onto another piece of bone. The surface of this tool is less well preserved than the others, which might explain the absence of recognisable hafting traces, but given the abundance of other traces on the tool surface, we consider that this tool was probably never hafted. Wear traces suggestive of sediment abrasion are present on the distal edge of all three tools. Together with the chips along the edges and invasive polish we interpret this wear as signalling use in soft, loose, fine-grained sediments (cf. Xie, 2014: Fig. 3.19). Reynard (2013) has explored the effects of trampling on the formation of coarse striations and polish on bone tools. We discount the possibility that trampling was responsible for the sediment abrasion on our tools for two reasons: 1) the abrasion wear is confined to the distal portions of the tools; and 2) the sediment abrasion on both Pont Drift tools is uni-directional and slightly angled suggesting anthropic use (sensu Semenov, 1964). The combination of all the use-wear traces taken together supports our interpretation of anthropogenic implements. Along portions of the blade are patches of ambiguous wear. Experimental processing of African potatoes (Hypoxis hemerocallidea) and leather show that they produce almost indistinguishable traces (Fig. 5). Coupled with the sediment abrasion and inferred method of hafting, however, we consider leather working to be unlikely. It is probable that the three tools were used as hoes to harvest underground tubers and cultivate small garden crops, if not also to till soil for seed planting and weeding. Scapula hoes are known to have been used in ad hoc activities additional to tilling and cultivating (Griffitts, 2006; Xie, 2014), and it is possible that the same was done in South Africa. The use-wear associated with chopping squashes, for instance, is slow to develop and not very diagnostic (Griffitts, 2006), but may account for some of the larger chips along the distal edge of the implements. That iron was available and used during the period under study is well known (e.g., Calabrese, 2000; Miller, 2001), not least because the three bone tools themselves show evidence of having been fashioned, at least in part, by a metal blade. During the K2 and
Fig. 5. Some materials can produce almost identical use-wear features; distinguishing between them can be challenging. Here we show similar polish and striations produced experimentally by processing leather and an African potato (Hypoxis hemerocallidea) for approximately 30 min with a bone tool. Micrographs are taken at 100 magnification. Scale bars represent 200 mm.
J. Bradfield, A.R. Antonites / Quaternary International 472 (2018) 126e134
Transitional K2 periods, the production and distribution of metal objects was restricted and mainly concentrated at the elite settlements of K2 and Mapungubwe (Calabrese, 2000). The use of bone hoes may thus support the argument that metal objects were not widely accessible to all members of society (even at K2), or that iron was not generally used for utilitarian objects like hoes. On the other hand, the value of iron may imply more recycling and re-use of objects over time, resulting in the relatively rare occurrence of iron hoes in Limpopo Valley assemblages (Miller, 2002). The use of bone hoes introduces an interesting perspective on the dynamic between male and female associations related to tool use. Although hoe cultivation is historically associated with women, and iron production with men (e.g., Stayt, 1968; Shaw, 1974), this dichotomy may have been more fluid in the archaeological past (see Chirikure, 2007; Chirikure et al., 2015). Bone, on the other hand, does not appear to have had any specific gendered associations. Recent ethnographies, for example, document activities in which both women and men made and used bone tools (Du Toit, 1968; Davison, 1975; ; Shaw, 1974). The use-traces on the three bone tools examined here do not support the notion that they might have been used as apotropaic objects (sensu Gunda, 1976), but rather that they served a utilitarian function. The depositional context of the Pont Drift long bone hoe, however, suggests that it may have moved from the utilitarian to the ritual sphere (sensu Whitehouse, 1996; Gosden and Marshall, 1999). Whether its position in the mound-like feature was purely symbolic (cf. Du Toit, 1968), or it formed part of the ritual performance involved in the construction thereof, remains unknown. 7. Conclusion This paper presents the first recognised evidence of possible bone hoes in South Africa. The use-wear on these tools is consistent with that observed on bone hoes by other analysts (e.g. Griffitts, 2006; Xie, 2014). Our findings add to the growing understanding of the complexity of bone tool production among southern African Iron Age agriculturalists between the 10th and 13th centuries AD (Antonites et al., 2016), as well as the global distribution of bone hoe technology. The economy of the Limpopo Valley communities was based on mixed farming, with the focus being on cereals like sorghum, millets and legumes (Huffman, 2007). These crops, which grow well in warm, loose, well-drained soils, can be cultivated in small plots and require constant attendance in terms of weeding (Directorate of Plant Production, 2010). In other words, soil conditions conducive to producing the wear traces seen on the distal blades of the bone implements. In order to meet the consumption requirements of the community living at K2 during its zenith, it has been estimated that approximately 634 hectares of ground would need to have been under cultivation (Murimbika, 2006). Bone hoes may have been used during tillage of small plots or they may have been used to gather underground foods such as melons, tubers or rhizomes during foraging trips. Whatever their primary intention, bone hoes appear to have been a short-lived technological innovation, eventually giving way to their metal counterparts (Herbert, 1996). By extending the ethnography on gendered associations back to the Middle Iron Age, the recovery of the scapula hoes from domestic or household contexts links these objects to women and offers a glimpse into women's activities during the K2 and TK2 periods. Unfortunately, the use-wear does not allow us to say who made these implements, but given the cache of hafted bone arrows found at K2 (Antonites et al., 2016) it appears likely that both men and women worked and used bone during this period. The fact that only three such implements have been recovered may point to a
133
less intensive role in agricultural production, but it is possible that similar tools have not been recognised by faunal analysts. A worthwhile pursuit would be to revisit some of these collections. Acknowledgements For access to the artefacts we wish to thank Johnny van Schalkwyk and Frank Teichert at the Ditsong National Museum of Cultural History, as well as Thembi Russell and Faye Lander of the Archaeology Collections, University of the Witwatersrand. Marlize Lombard is thanked for permission to use the microscopes in the Micro-Tracks research facility, University of Johannesburg. Jan Boeyens kindly commented on a previous draft of the manuscript. JB's research is supported by grants from the Oppenheimer Memorial Trust and the National Research Foundation. References Anderson, M.S., 2009. Marothodi: the Historical Archaeology of an African Capital. Atikkam, Woodford. Antonites, A.R., Bradfield, J., Forssman, T., 2016. Technological, functional and contextual aspects of the K2 and Mapungubwe worked bone industries. Afr. Archaeol. Rev. 33, 437e463. Baron, J., Diakowski, M., Stolarczyk, T., 2016. Bone and antler artefacts from an 8-5th century BC settlement at Grzybiany, South-Western Poland. In: Vitezovi c, S. (Ed.), Close to the Bone: Current Studies in Bone Technologies. Institute of Archaeology, Belgrade, pp. 26e45. Bell, R., Cross, G., 1980. Oklahoma Indian Artefacts: Contribution from the Stovall Museum. University of Oklahoma Press, Oklahoma. Bradfield, J., 2015. Use-trace analysis of bone tools: a brief overview of four methodological approaches. South Afr. Archaeol. Bull. 70, 3e14. Brain, C.K., 1981. The Hunters or the Hunted? an Introduction to African Cave Taphonomy. University of Chicago Press, Chicago. Bryant, A., 1949. The Zulu People: as They Were before the White Man Came. Shuter & Shooter, Pietermaritzburg. Calabrese, J.A., 2000. Metals, ideology and power: the manufacture and control of materialised ideology in the area of the Limpopo-Shashe confluence, c. AD 900 to 1300. South Afr. Archaeol. Soc. Goodwin Ser. 8, 100e111. Campana, D., 1989. Natufian and Protoneolithic Bone Tools: the Manufacture and Use of Bone Implements in the Zagros and the Levant. BAR International Series 494, Oxford. Chapman, J., 1864. Travels in the Interior of South Africa. William Clowes and Sons., London. Childs, S.T., Herbert, E., 2005. Metallurgy and its Consequences. In: Stahl, A. (Ed.), African Archaeology: a Critical Introduction. Blackwell Publishing, Oxford, pp. 276e300. Chirikure, S., 2007. Iron production and its position in Iron Age communities of southern Africa. J. Soc. Archaeol. 5, 74e102. Chirikure, S., 2015. Metals in Past Societies: a Global Perspective on Indigenous African Metallurgy. SpringerBriefs in Archaeology. Springer, New York. Chirikure, S., Hall, S., Rehren, T., 2015. When ceramic sociology meets material science: sociological and technological aspects of crucibles and pottery from Mapungubwe, southern Africa. J. Anthropol. Archaeol. 40, 23e32. Clark, J.D., 1959. The Prehistory of Southern Africa. Penguin Books, London. Culwick, A.T., Culwick, G.M., 1934. The functions of bride-wealth in Ubena of the Rivers. Afr. J. Int. Afr. Inst. 7, 140e159. Curwin, E., 1926. On the use of scapulae as shovels. Sussex Archaeol. Collect. 67, 139e145. Curwin, E., 1937. The Archaeology of Sussex. Methuen & Co., London. d'Errico, F., 1993. Criteria for identifying utilised bone: the case of the Cantabrian “tensors”. Curr. Anthropol. 34, 299e309. Dart, R., 1958. Bone tools and porcupine gnawing. Am. Anthropol. 60, 715e724. Davison, P., 1975. Some Nguni Crafts: the uses of horn, bone and ivory. Ann. South Afr. Mus. 70, 79e155. Directorate of Plant Production, 2010. Sorghum Production Guidelines. Department of Agriculture. Forestry and Fisheries, Pretoria. €le skeppinge van die BaPhalaborwa in Du Toit, A.P., 1968. Die plek van die materie hul kultuur. PhD dissertation. University of South Africa, Pretoria. Earthy, E.D., 1933. Valenge Women: the Social and Economic Life of the Valenge Women of Portuguese East Africa. Oxford University Press, London. Evora, M., 2015. Use-wear methodology on the analysis of osseous industries. In: Marreiros, J.M., Gibaja Bao, J.F., Ferreira Bicho, N. (Eds.), Use-wear and Residue Analysis in Archaeology. Springer, Germany, pp. 159e170. Fisher, J., 1995. Bone surface modifications in zooarchaeology. J. Archaeol. Method Theory 2, 7e68. Gardner, G.A., 1963. Report on Excavations at Mapungubwe and Bambandyanalo in the Transvaal from 1935e1940, Vol. II. Van Schaik, Pretoria. Garlake, P.S., 1973. Great Zimbabwe. Thames and Hudson, London. Gosden, C., Marshall, Y., 1999. The cultural biography of objects. World Archaeol. 31
134
J. Bradfield, A.R. Antonites / Quaternary International 472 (2018) 126e134
(2), 169e178. Griffitts, J., Bonsall, C., 2001. Experimental determination of the function of antler and bone ‘bevel-ended tools’ from prehistoric shell middens in western Scotland. In: Choyke, A., Bartosiewicz, L. (Eds.), Crafting Bone: Skeletal Technologies through Time and Space. BAR International Series 937, Oxford, pp. 207e220. Griffitts, J., 2006. Bone Tools and Technological Choice: Change and Stability on the Northern Plains. Unpublished PhD thesis. University of Arizona, Texas. Gunda, B., 1976. More on iron hoes in Africa. Curr. Anthropol. 17, 172. Hanisch, E.O.M., 1980. An Archaeological Interpretation of Certain Iron Age Sites in the Limpopo/Shashi Valley. Unpublished MA dissertation. University of Pretoria, Pretoria. Herbert, H., 1996. Metals and power at Great Zimbabwe. In: Pwiti, G., Soper, R. (Eds.), Aspects of African Archaeology: Papers from the 10th Congress of the Pan African Association for Prehistory and Related Studies. University of Zimbabwe Publications, Harare, pp. 641e655. Huffman, T., 2000. Mapungubwe and the origins of the Zimbabwe culture. South Afr. Archaeol. Soc. Goodwin Ser. 8, 14e29. Huffman, T., 2007. Handbook to the Iron Age: the Archaeology of Pre-colonial Farming Societies in Southern Africa. University of KwaZulu-Natal Press, Pietermaritzburg. Hunt, D.R., 1931. An account of the Bapedi. Bantu. Stud. 5 (1), 275e326. Kuper, A., 1982. Wives for Cattle: Bridewealth and Marriage in Southern Africa. Routledge, London. Lane, P., 2013. Trajectories of pastoralism in northern and central Kenya: an overview of the archaeological and environmental evidence. In: Bollig, M., Schnegg, M., Wotzka, H.-P. (Eds.), Pastoralism in Africa: Past, Present, and Futures. Berghahn, New York, pp. 104e144. ra, I., 2007. Methods, means and results when studying European Legrand, A., Side bone industries. In: Gates St-Pierre, C., Walker, R. (Eds.), Bones as Tools: Current Methods and Interpretations in Worked Bone Studies. BAR International Series 1622, Oxford, pp. 67e78. Livingstone, D., 1851. Missionary Travels and Researches in South Africa. John Murray, London. Lombard, M., Wadley, L., Deacon, J., Wurz, S., Parsons, I., Mohapi, L., Swart, J., Mitchell, P., 2012. South African and Lesotho Stone Age sequence updated (i). South Afr. Archaeol. Bull. 67, 123e144. Luik, H., 2011. Material, technology and meaning: antler artefacts and antler working on the eastern shore of the Baltic Sea in the Late Bronze Age. Est. J. Archaeol. 15, 32e55. Lupo, K., Schmitt, D., 2002. Upper Paleolithic net-hunting, small prey exploitation and women's work effort: a view from the ethnographic and ethnoarchaeological record of the Congo Basin. J. Archaeol. Method Theory 9, 147e179. Lyman, R., 1994. Vertebrate Taphonomy. Cambridge University Press, Cambridge. MacEachern, S., 2005. Two thousand years of West African history. In: Stahl, A. (Ed.), African Archaeology: a Critical Introduction. Blackwell Publishing, Oxford, pp. 441e466. Matthews, W., 1877. Ethnography and philology of the Hidatsa Indians. Reissue 1969 Plains Anthropologist, 14, pp. 177e252. Meier, J., 2013. More fun than games? An experimental study of worked bone astragali from two Middle Bronze Age Hungarian sites. In: Choyke, A., O'Connor, S. (Eds.), From These Bare Bones: Raw Materials and the Study of Worked Osseous Objects. Oxbow Books, Oxford, pp. 166e173.
Meiring, A.J., 1952. Polished-edge bone implements. Navor. Nas. Mus. 1, 31e34. Meyer, A., 1998. The Archaeological Sites of Greefswald: Stratigraphy and Chronology of the Sites and a History of Investigations. University of Pretoria, Pretoria. Miller, D., 2001. Metal assemblages from Greefswald areas K2, Mapungubwe Hill and Mapungubwe southern terrace. South Afr. Archaeol. Bull. 56 (174), 83e103. Miller, D., 2002. Smelter and smith: iron Age metal fabrication technology in southern Africa. J. Archaeol. Sci. 29 (10), 1083e1131. Moffett, A., Maggs, T., Van Schalkwyk, J., 2017. Breaking ground: hoes in precolonial South Africadtypology, medium of exchange and symbolic value. Afr. Archaeol. Rev. https://doi.org/10.1007/s10437-017-9247-1. Mumford, W.B., 1934. The hehe-bena-sangu peoples of East Africa. Am. Anthropol. 36, 203e222. Murimbika, M., 2006. Sacred Powers and Rituals of Transformation: an Ethnoarchaeological Study of Rainmaking Rituals and Agricultural Productivity during the Evolution of the Mapungubwe State, AD 1000 to AD 1300. Unpublished PhD dissertation. University of the Witwatersrand, Johannesburg. Phillipson, D., 2005. African Archaeology, third ed. Cambridge University Press, Cambridge. Plug, I., 2000. Overview of Iron Age fauna from the Limpopo Valley. South Afr. Archaeol. Soc. Goodwin Ser. 8, 117e126. Reynard, J., 2013. Trampling in coastal sites: an experimental study on the effects of shell on bone in coastal sediment. Quat. Int. 330, 156e170. Schoeman, M., 2013. Excavating ephemeral remains of life in a time of witchcraft: new insights into the eighteenth and nineteenth-century occupations at Leokwe and Nyindi hills in the Shashe-Limpopo confluence area, South Africa. Int. J. Hist. Archaeol. 17, 613e631. Semenov, S., 1964. Prehistoric Technology. Adams and Dart, Bath. Shaw, M., 1974. Material culture. In: Hammond-Tooke, W.D. (Ed.), The Bantuspeaking Peoples of Southern Africa. Routledge & Keagan Paul, London, pp. 85e131. Shaw, M., van Warmelo, N., 1974. The material culture of the Cape Nguni. Ann. South Afr. Mus. 58, 103e214. Stayt, H.A., 1968. The Bavenda, second ed. Frank Cass, London. Stokes, P., 2016. A new interpretation of post-medieval bone scoops from the foreshore of the river Thames in London. In: Vitezovi c, S. (Ed.), Close to the Bone: Current Studies in Bone Technologies. Institute of Archaeology, Belgrade, pp. 324e337. Strong, D., 1933. The Plains culture area in the light of archaeology. Am. Anthropol. 35, 271e287. Voigt, E., 1983. Mapungubwe: an Archaeological Interpretation of an Iron Age Community. Transvaal Museum Monograph 1. Transvaal Museum, Pretoria. Wang, W., 1991. Bone si and related issues in the Hemudu culture. Southeast Cult. 1, 240e242. Whitehouse, R., 1996. Ritual objects. Archaeological joke or neglected evidence? In: Wilkins, J. (Ed.), Approaches to the Study of Ritual: Italy and the Ancient Mediterranean. Accordia Research Centre, London, pp. 9e30. Xie, L., 2014. Early To Middle Holocene Earth-Working Implements And Neolithic Land-Use Strategies On The Ningshao Plain, China. Unpublished doctoral thesis. University of Arizona. Zhang, S., d'Errico, F., Backwell, L., Zhang, Y., Chen, F., Gao, X., 2016. Ma'anshan cave and the origin of bone tool technology in China. J. Archaeol. Sci. 65, 57e69.