Date Published: July 30, 2018
Publisher: Springer US
Author(s): Jennifer M. Miller, Elizabeth A. Sawchuk, Amy L. R. Reedman, Pamela R. Willoughby.
Shell beads are well established in the archaeological record of sub-Saharan Africa and appear as early as 75,000 BP; however, most research has focused on ostrich eggshell (OES) and various marine mollusc species. Beads made from various land snails shells (LSS), frequently described as Achatina, also appear to be widespread. Yet tracking their appearance and distribution is difficult because LSS beads are often intentionally or unintentionally lumped with OES beads, there are no directly dated examples, and bead reporting in general is highly variable in the archaeological literature. Nevertheless, Achatina and other potential cases of LSS beads are present at over 80 archaeological sites in at least eight countries, spanning the early Holocene to recent past. Here, we collate published cases and report on several more. We also present a new case from Magubike Rockshelter in southern Tanzania with the first directly dated LSS beads, which we use to illustrate methods for identifying LSS as a raw material. Despite the long history of OES bead production on the continent and the abundance of land snails available throughout the Pleistocene, LSS beads appear only in the late Holocene and are almost exclusively found in Iron Age contexts. We consider possible explanations for the late adoption of land snails as a raw material for beadmaking within the larger context of environmental, economic, and social processes in Holocene Africa. By highlighting the existence of these artifacts, we hope to facilitate more in-depth research on the timing, production, and distribution of LSS beads in African prehistory.
Shell beads have a long history of production in sub-Saharan Africa and are one of the first indicators of early modern human symbolic behavior, appearing by 75,000 years ago alongside other forms material culture such as utilized ochre and portable/parietal art (d’Errico et al. 2005; Henshilwood and Marean 2003; McBrearty and Brooks 2000; Wadley 2001). Although the earliest examples were perforated whole marine shells, standardized production of shaped ostrich eggshell (OES) beads was established by at least 50,000 BP (Miller and Willoughby 2014). OES beads remain well represented in Later Stone Age (LSA) and Iron Age (IA) deposits, with the tradition continuing into the ethnographic present among linguistically and culturally diverse communities (Chittick 1975; Lee 1984; Marshall 1976; Silberbauer 1965, 1981; van der Sleen 1958). In contrast to more extensive research on glass beads, however, few studies have moved beyond quantification of OES to focus on chronology, distribution, and manufacture. Such studies are typically focused on metric analyses (e.g., Jacobson 1987; Kandel and Conard 2005; Orton 2008; Sadr et al. 2003; Smith et al. 1991, 2001; Wilmsen 2015). Other notable work has drawn on ethnographic data to explore the social contexts of these artifacts (Williams 1987; Wingfield 2009). This paper builds on the OES literature by focusing on a concurrent but even less studied phenomenon: the production of similar disc beads from the shells of terrestrial land snails.
Reviewing the available literature, it is apparent that LSS beads are not unusual finds. Unfortunately, it also becomes apparent that there is no consistent framework for identification or reporting them. Here, we present the geographic and chronologic distribution of published cases of LSS (and potential LSS) beads (Fig. 1, Table 1). We also include several sites which had not previously published the occurrence of LSS (Mumba, Mlambalasi, Border Cave, and White Paintings Shelter). These cases were ascertained by one of the authors (JM) during first hand observation of collections. It is highly possible that other collections likewise have LSS beads that were mistakenly attributed to OES and have not been further examined.Fig. 1Geographic distribution of sub-Saharan sites with LSS, and potential LSS beadsTable 1List of sub-Saharan archaeological sites with LSS, and potential LSS beads, by countrySiteBead OriginalAgeCalibrated age (BC/AD)ReferenceMaterialLSS?BotswanaBobonong RoadAchatinay810 ± 70 BP659–910 calBP (1,291–1,040 calAD)Kinahan et al. 1998BoutsweAchatinayIron Age–DuBroc 2010; Klehm 2013Khubu la DintšaShell690 ± 30 BP563–684 calBP (1,387–1,266 calAD)Klehm 2013MatangaAchatinay530 ± 75 BP338–668 calBP (1,612–1,282 calAD)Van Waarden 1987580 ± 75 BP508–670 calBP (1,422–1,280 calAD)Mmadipudi HillAchatinay–990–1,130 calBP (960–820 calAD)Klehm 2013; Klehm and Ernenwein 20161,390–1,460 calBP (630–490 calAD)TaukomeAchantina [sic]y995 ± 75 BP741–1,057 calBP (1,209–893 calAD)Denbow 19831,240 ± 80 BP982–1,295 calBP (968–655 calAD)1,265 ± 80 BP984–1,308 calBP (966–642 calAD)Toutswe (Toutswemogala)Achantina [sic]y755 ± 75 BP554–901 calBP (1,396–1,049 calAD)Denbow 1983990 ± 75 BP737–1,056 calBP (1,213–894 calAD)White Paintings Shelter*y––Robbins et al. 2000; *GhanaKintampo 6 Rock ShelterShell3,485 ± 100 BP3,482–4,068 calBP (1,532–2,118 calBC)Stahl 19853,550 ± 127 BP3,484–4,226 calBP (1,534–2,276 calBC)3,605 ± 100 BP3,640–4,225 calBP (1,690–2,275 calBC)KenyaGede (Gedi)Achatina/shellyIron Age–Flexner et al. 2008; Kirkman 1957; Mukhwana 1992KilepwaShellIron age–Kirkman 1952MgombaniShell1,300 ± 50 BP1,304–1,084 calBP (646–866 calAD)Helm et al. 2012North Horr IMollusc shell––Wandibba 1988Panga ya SaidiShellIron Age–Helm et al. 2012ShangaMollusc shellIron Age–Mukhwana 1992; Robertshaw 1984Wabukhe HillMollusc shellIron Age–Wandibba 1988MalawiFingiraShellLater Stone Age–Clark 1967Hora MountainAchatinayLater Stone Age–Clark 1956Linthipe/Changoni Sites (DZ40, DZ6A, DZ6B, DZ12, DZ126)Shell90 ± 40 BP12–269 calBP (1,938–1,681 calAD)Mgomezulu 19781,020 ± 80 BP739–1,172 calBP (1,211–778 calAD)1,580 ± 80 BP1,310–1688 calBP (640–262 calAD)Mazinga 1AchatinayIron Age/Later Stone Age–Zipkin and Thompson (personal comm.)Kadawonda 1AchatinayIron Age/Later Stone Age–Zipkin and Thompson (personal comm.)MozambiqueChibueneShellIron Age–Sinclair 1982ManekweniAchatinay340 ± 70 BP155–514 calBP (1,795–1,436 calAD)Garlake 1976; Otlet and Walker 1979590 ± 70 BP515–668 calBP (1,435–1,282 calAD)780 ± 80 BP563–905 calBP (1,387–1,170 calAD)South AfricaBalerno Shelter 3Achatinay1,650 ± 50 BP1,412–1,693 calBP (538–257 calAD)van Doornum 20142,270 ± 50 BP2,153–2,353 calBP (203–403 calBC)Border Cave*y––Beaumont 1978; *Bushman Rock ShelterAchatinay9,570 ± 55 BP10,717–11,131 calBP (8,767–9,181 calBC)Dayet et al. 2017; Plug 198212,950 ± 40 BP15,279–15,686 calBP (13,329–13,736 calBC)Castle RockAchatinayIron Age–Calabrese 2000; Wood 2006Ficus CaveAchatinay330 ± 50 BP302–495 calBP (1,648–1,455 calAD)Partridge 1966; Vogel and Marais 1971HarmonyAchatinay320 ± 25 BP306–462 calBP (1,644–1,488 calAD)Evers 1975, 1979iNkolimahashi Rock ShelterLand snaily550 ± 45 BP511–649 calBP (1,439–1,301 calAD)Mazel 1999K2 (Bambandyanalo)AchatinayIron Age–Hattingh and Hall 2009; Steyn and Nienaber 2000; Wood 2006Leokwe HillAchatinay880 ± 25 BP730–905 calBP (1,220–1,045 calAD)Calabrese 2000; Hall and Smith 20001,000 ± 60 BP782–1,055 calBP (1,168–895 calAD)MagogoA. immaculata, M. kraussi, Achatinidaey1,190 ± 50 BP981–1,258 calBP (969–692 calAD)Maggs and Ward 1984; Ward and Maggs 19881,360 ± 50 BP1,012–1,281 calBP (938–669 calAD)MapungubweSnailyIron Age–Meyer 2000MhlopeniMetachatinay1,410 ± 50 BP1,263–1,406 calBP (687–544 calAD)Maggs and Ward 1984MNR 74Achatinay729 ± 31 BP573–726 calBP (1,377–1,224 calAD)Antonites et al. 2016837 ± 35 BP683–895 calBP (1,267–1,055 calAD)MutambaAchatinayIron Age–Antonites 2012MpambanyaniAchatinidaey885 ± 50 BP705–919 calBP (1,245–1,031 calAD)Robey 1980; Ward and Maggs 1988980 ± 50 BP769–977 calBP (1,181–973 calAD)NdondondwaneA. immaculata, M. kraussi, Achatinidaey1,190 ± 50 BP981–1,258 calBP (969–692 calADFread 2007; Stoffberg and Loubser 1984; Ward and Maggs 19881,230 ± 50 BP1,012–1,281 calBP (938–699 calAD)NtshekaneA. immaculata, M. kraussi, Achatinidaey1,100 ± 50 BP928–1,173 calBP (1,022–777 calAD)Maggs and Michael 1976; Ward and Maggs 1988NtsitsanaShell1,180 ± 50 BP973–1,255 calBP (977–695 calAD)Prins and Granger 19931,290 ± 50 BP1,082–1,299 calBP (868–651 calAD)PengeAchatinay1,318 ± 26 BP1,184–1,295 calBP (766–655 calAD)Antonites et al. 2014Pont DriftAchatinay810 ± 50 BP666–899 calBP (1,284–1,051 calAD)Hanisch 19801,110 ± 50932–1,173 calBP (1,018–777 calAD)Princess HillShellIron Age–Antonites 2012; Loubser 1989SchrodaAchatinay780 ± 50 BP657–791 calBP (1,293–1,159 calAD)Hanisch 1980; Hall and Smith 2000840 ± 50 BP677–905 calBP (1,273–1,045 calAD)TloutleAchatinay375 ± 65 BP305–516 calBP (1,645–1,434 calAD)Mitchell 19935,080 ± 80 BP5,646–5,990 calBP (3,696–4,040 calBC)6,140 ± 100 BP6,759–7,261 calBP (4,809–5,311 calBC)6,910 ± 80 BP7,609–7,930 calBP (5,659–5,980 calBC)8,680 ± 70 BP9,534–9,889 calBP (7,584–7,939 calBC)VerulamAchatinay–160–80 calBP (1,790–1,870 calAD)Loubser 1991370–270 calBP (1,580–1,680 calAD)VhunyelaShellIron Age–Antonites 2012TanzaniaFukuchaniShellIron Age–Faulkner et al. 2017Gonja KalimaniLand snailyIron Age–Walz 2010aGonja MaoreAchatina/land snaily–342–468 calBP (1,608–1,482 calAD)Soper 1967; Walz 2010a, 2010b1,080 ± 115 BP769–1,265 calBP (1,181–685 calAD)–812–897 calBP (1,138–1,053 calAD)Jangwani ISNAILyPastoral Neolithic–Mehlman 1989KilwaSHELLIron Age–Soper 1967Kwa MgogoLAND snail (9 species)y–1,102–1,234 calBP (848–716 calAD)Walz 2010a, 2010b995–1,051 calBP (995–899 calAD)Magubike*y371 ± 23 BP319–501 calBP (1,631–1,449 calAD)Willoughby 2012; *397 ± 23 BP332–509 calBP (1,618–1,441 calAD)403 ± 23 BP334–510 calBP (1,625–1,455 calAD)1,732 ± 23 BP1,569–1,703 calBP (381–247 calAD)Mlambalasi*y151 ± 24 BP480–315 calBP (1,470–1,635 calAD)Biittner et al. 2017; *Mumba*y–454–304 calBP (1,496–1,646 calAD)Prendergast et al. 2007; *472–290 calBP (1,478–1,660 calAD)1,922–1,615 calBP (28–335 calAD)Unguja UkuuShellIron Age–Chittick 1974; Faulkner et al. 2017; Wynne-Jones 2016ZambiaChilimuliloShellIron Age–Musonda 1984Kalemba Rock ShelterAchatina/land snaily115 ± 70 BP0–283 calBP (1,950–1,667 calAD)Derricourt 1976; Phillipson 19764,480 ± 90 BP4,862–5,434 calBP (2,912–3,484 calBC)5,040 ± 110 BP5,491–6,095 calBP (3,541–4,145 calBC)7,030 ± 105 BP7,657–8,037 calBP (5,707–6,087 calBC)MakweAchatina/land snaily4,380 ± 130 BP4,585–5,440 calBP (2,635–3,490 calBC)Phillipson 1976; Yamasaki et al. 19724,920 ± 130 BP5,326–5,922 calBP (3,376–3,972 calBC)MufulweShellIron Age–Gutin and Musonda 1985; Musonda 1984MwambacimoShelly––Musonda 1984Nachikufu ShelterSnail/shelly1,060 ± 100 BP747–1,232 calBP (1,203–718 calAD)Miller 1969Nsalu Hill CaveSnailyLater Stone Age–Miller 1969Thandwe Rock ShelterAchatina/land snaily890 ± 110 BP656–1,050 calBP (1,294–900 calAD)Phillipson 1976; Yamasaki et al. 1972ZimbabweChiwona KopjeShellIron Age–Caton-Thompson 1931DanamombeAchatinayIron Age–Machiridza 2012Great ZimbabweAchatina/shellyIron Age–Beck 1931; Caton-Thompson 1931Hlamba Mlonga HillAchatinay720 ± 50 BP558–735 calBP (1,392–1,215 calAD)Thorp 2009780 ± 50 BP657–791 calBP (1,293–1,159 calAD)1,040 ± 40 BP832–1,057 calBP (1,118–893 calAD)Hubvumi RuinsShellIron Age–Caton-Thompson 1931Kadzi RiverAchatinay990 ± 50 BP785–1,045 calBP (1,165–905 calAD)Plug 19971,290 ± 50 BP1,082–1,299 calBP (868–651 calAD)KhamiShell290 ± 30 BP288–457 calBP (1,662–1,493 calAD)Mukwende 2016350 ± 30 BP315–492 calBP (1,635–1,458 calAD)430 ± 30 BP335–529 calBP (1,615–1,421 calAD)MalumbaAchatinay690 ± 95 BP517–792 calBP (1,433–1,158 calAD)Bvocho 2005; Manyanga 2006; Manyanga et al. 2000965 ± 80 BP707–1,053 calBP (1,243–897 calAD)MatendereShellIron Age–Caton-Thompson 1931ZimbabweMapelaShell770 ± 30BP669–733 calBP (1,281–1,217 calAD)Chirikure et al. 2014; House 2016900 ± 30 BP740–911 calBP (1,210–1,039 calAD)Mtao Village 16Achatina/shelly420 ± 45 BP319–532 calBP (1,631–1,418 calAD)Manyanga 2006MutshilachokweShell910 ± 60 BP705–931 calBP (1,245–1,019 calAD)Manyanga 2006670 ± 60 BP540–696 calBP (1,410–1,254 calAD)MweneziAchatinay1,250 ± 75 BP987–1,298 calBP (963–652 calAD)Bvocho 2005; Manyanga 2006; Manyanga et al. 2000800 ± 70 BP653–910 calBP (1,297–1,040 calAD)SamakandeAchatinay1370 ± 35 BP1,188–1,348 calBP (762–602 calAD)Manyanga and Shenjere 20121,580 ± 35 BP1,396–1,546 calBP (554–404 calAD)TshobwaneShell790 ± 60 BP572–902 calBP (1,378–1,048 calAD)Manyanga 2006890 ± 60 BP700–922 calBP (1,250–1,028 calAD)The asterisk refers to this publication as the original reporting
Investigating the phenomenon of LSS beads first requires some background on the raw material. Shelled gastropods, commonly referred to as snails, live in a variety of terrestrial and aquatic environments. Lunged, air-breathing gastropod species belong to the informal group Pulmonata (Cuvier in Blainville 1814) and are primarily terrestrial (Bouchet and Rocroi 2005). Three major tropical snail families within the achatinoid (Stylommatophora) clade are prevalent throughout sub-Saharan Africa (Rowson et al. 2011). These include the carnivorous hunter snail family Streptaxoidea (Gray 1860), the awl snails of Subulinidae (Fischer and Crosse 1877), and the giant African land snail family, Achatinidae (Swainson 1840). There are approximately 254 species within the Achatinidae family. Eight Achatinidae genera have taxa with the shell length, shapes, and thickness necessary to produce a dense, non-curved blank for beadmaking. Subulinidae and Streptaxoidea shell lengths range from < 1 to ~ 30 mm and have depressed trochiform shapes which are too thin and curved for disc bead production. Identifying small, worked shell beads to taxa can be challenging. Ostrich eggshell is one of the most recognizable materials because of its frequency at sub-Saharan archaeological sites and the regular dotted patterns of pores on the cuticle surface. Non-OES shell and worn beads of various materials can be more difficult to identify. In early production stages, mollusc beads are easily distinguished from OES by the shiny nacre on their inner surface and details on the outer surface such as distinctive ridges or colored patterns. However, on heavily manufactured or worn beads, these features tend to be replaced by a smooth, unremarkable surface. As a result, worn beads made from OES, LSS, marine shell, bone, or ivory may appear superficially similar. A review of published sources, combined with new data from Magubike Rockshelter and other previously unreported sources, presents a startlingly vast picture of LSS bead occurrence in sub-Saharan Africa. Furthermore, this is almost certainly an underestimate given variable reporting standards for land snail artifacts and disc beads respectively, as well as the challenges associated with identifying LSS as a bead raw material. We hope that by highlighting these artifacts, we can promote further discussion and the publication of new data that help refine understanding of the role LSS beads played in prehistory. At present, it appears that LSS beads are largely confined to eastern and southeastern Africa during the last 2,000 years, associated with Iron Age communities. Although this is partly biased by ongoing research foci within southeastern Africa, there is compelling evidence that this phenomenon is linked to population increase, elaboration of trade networks, and the growth of social complexity during the later Holocene. It is intriguing that LSS beads appear tens of thousands of years after the establishment of OES beadmaking, despite the much longer tenure of land snails in the same parts of the world. Although we offer some possible explanations as to why this might be, more data are required to begin formally testing hypotheses. Highlighting these artifacts is intended to promote discussion, and ultimately publication, of additional finds from scholars working in diverse regions across the continent. We hope that re-examination of many disc bead assemblages reveals some unexpected finds. Source: http://doi.org/10.1007/s10437-018-9305-3