Академический Документы
Профессиональный Документы
Культура Документы
The origin of the Middle Stone Age (MSA) denotes the transition from a highly persistent mode of stone
toolmaking, the Acheulean, to a period of increasing technological innovation and cultural indicators
The oldest development of the MSA has been obscure because sensu lato and the MSA in the middle Pleistocene (6, 7), the
stratigraphic records that span the Acheulean, which is char- earliest MSA predates current evidence for the first appear-
acterized by distinctive large cutting tools (LCTs), to the early ance of H. sapiens s.s. in eastern Africa and likely constitutes
MSA are rare and poorly constrained in time. Here we pre- the context in which the anatomical and behavioral charac-
sent precise geochronology based on 40Ar/39Ar and U-series teristics of our species evolved.
dating of middle Pleistocene archeological sites in the Olor- A summary of dates (Table 1) shows that Acheulean and
gesailie Basin, South Kenya rift (detailed site stratigraphy in transitional industries incorporating large tools (Sangoan,
Fig. 1; composite stratigraphy in fig. S8), that document the Fauresmith) disappear in Kenya before 285 ka or earlier, but
stone technology, fauna, and environments of the Acheulean persist in areas of Ethiopia, Sudan, and southern Africa into
to Middle Stone Age sequence (1, 2). the end of the middle Pleistocene, after the first appearance
As originally defined (3), the MSA lacked LCTs and of Homo sapiens s.s. While there are currently twelve pub-
bladelet technologies, although more recent work recognizes lished dated middle Pleistocene South African sites, only Flo-
the presence of LCTs in some early MSA assemblages and of risbad, Kathu Pan, and possibly Sterkfontein and
bladelets especially after 70 ka. For much of the 20th century Wonderwerk, all in northern South Africa, have dated MSA
the MSA was considered irrelevant to human evolution and horizons older than Marine Isotope Stage 6 (MIS6, ~191 ka)
dispersal largely because the chronometric techniques and (8), with six that lack LCTs. Coastal MSA cave horizons older
climate correlations used in African contexts placed it at the than MIS6 appear in northwestern Africa, but are absent
end of the Pleistocene, contemporary with the richer record from the rich later record of coastal South African caves.
of European cave art and elaborate lithic toolkits. Human be- Following investigations by our team on Acheulean sites,
havioral evolution during the middle Pleistocene (780-130 ka) fossil remains, and geology of the Olorgesailie Formation
and into the late Pleistocene prior to 30 ka was thus poorly dated ~1.2–0.5 Ma (9–12), research since 2002 has extended
understood (4). into the Oltulelei Formation (13) containing rich accumula-
Based on newer techniques, well-constrained oldest MSA tions of MSA artifacts. Here we set out the chronostrati-
sites in eastern and eastern/central Africa now have esti- graphic controls for the end of the Acheulean and its
mated ages ≥286 to ≤182 ka (Table 1), while the earliest fossils replacement by MSA sites in the Olorgesailie Basin, which
attributable to Homo sapiens sensu strictu in eastern Africa contains the longest stratigraphic sequence of Acheulean and
are ~160-200 ka (Table 1 and table S4; all K–Ar and 40Ar/39Ar overlying MSA sites in eastern Africa. Other recent and con-
ages are recalculated for consistency with K decay constants current manuscripts establish the stratigraphic and regional
and revised standard ages (5). While recent investigations in geological context of post-500-ka strata at Olorgesailie (13),
northern Africa suggest the coeval evolution of Homo sapiens
First release: 15 March 2018 www.sciencemag.org (Page numbers not final at time of first release) 1
characterize the MSA sites (1), and examine the paleoenviron- and the distinctive large-bodied grazing fauna associated
mental setting, faunal turnover, and landscape dynamics as- with these sites are located in Members 11, 12, and 14 (2). Pre-
sociated with the MSA in the Olorgesailie Basin (2). vious 40Ar/39Ar ages constrained Member 11 to a 62-ka inter-
The MSA sites of Olorgesailie are exposed within a ~65 val between underlying Member 10 at 671 ± 8 ka and
km2 early to late Pleistocene sedimentary basin in the south- overlying Member 12 at 609 ± 6 ka, with an age of 499 ± 2 ka
ern Kenya Rift, north of the deeply eroded late Pliocene–early in Member 14 (9). New 40Ar/39Ar age determinations for Mem-
Pleistocene central volcano of Mt. Olorgesailie (14) (Fig. 1, in- ber 11 tuffaceous deposits further constrain the archeological
set). The Pleistocene deposits lie upon older lavas dated at materials in Member 11' to the lower part of the 6 ka interval
2.66 ± 0.06 Ma (all chronologic uncertainties are reported from 615–609 ka. The 615-ka constraint is based on two ages
throughout at the 2σ level) (9) and the regionally extensive on pumice from a poorly sorted silty sandstone containing
Magadi Trachyte (1.4–0.8 ka) (15). Normal fault movement and underlying the artifacts (615.7 ± 6.9 from Locality A and
contemporaneous with sedimentation has resulted in separa- 614.7 ± 5.2 ka from Locality B; a third result of 626.6 ± 7.7 ka
tion of the Olorgesailie Basin into three sub-basins with shift- from Locality A was rejected as being too old (Text S3).
ing histories of terrigeneous and lacustrine sedimentary Archeological sites dated here in the Olkesiteti Member of
environments (13). the Oltulelei Formation in Locality B occur within a 4–6 m
First release: 15 March 2018 www.sciencemag.org (Page numbers not final at time of first release) 2
MSA sites in Locality G are distributed roughly north- markers such as long-distance obsidian transport and pig-
south over a distance of ~1.1 km, in strata of the Olkesiteti ment processing (1), are preserved in the Oltulelei Formation
Member of the Oltulelei Formation (Fig. 1C). Most of the tuff beginning most likely by 320 ka and no later than 305 ka.
samples that date these sites also are from this member. Six These ages also imply that a major shift in landscapes and
pumiceous tuff units were analyzed by the 40Ar/39Ar method, mammalian biota of the southern Kenya rift (2) had occurred
with additional age control from U-series dating. by 320 ka.
The U-series sample, consisting of silica and carbonate re- This newly calibrated archeological sequence provides a
placing roots or stems in the oldest tuff unit in Locality G, secure basis for comparison with the limited sample of other
yielded the precise age of 277.1 ± 1.8 ka (unit G-T1 in Fig. 1C, well-dated Acheulean and MSA occurrences in Kenya and
section G06-08). This tuff occurs in strata that overlie an ero- elsewhere in Africa. Evidence for an increased pace of envi-
sional unconformity above the MSA archeological material at ronmental change that now is well documented in the Olor-
GOK-1, which occurs in a thick (~2.5 m), well developed red- gesailie sequence suggests that eastern Africa was important
dish paleosol (Fig. 1C, section G08-01). Several factors suggest to the development of MSA technological, social, and cogni-
that site is considerably older than the U-series date: its oc- tive innovations in human behavior prior to 300 ka as a key
currence in a thick paleosol, presence of an erosional uncon- region fostering the development of a more widespread set of
First release: 15 March 2018 www.sciencemag.org (Page numbers not final at time of first release) 3
16. J. S. Bridge, Rivers and Floodplains: Forms, Processes, and the Sedimentary Record 38. G. Ll. Isaac, in Geological Background to Fossil Man, W. W. Bishop, Ed. (Scottish
(Blackwell Science, 2003). Academic Press, 1978), 173–206.
17. K. W. Min, R. Mundil, P. R. Renne, K. R. Ludwig, A test for systematic errors in 39. A. C. Parnell, J. Haslett, J. R. M. Allen, C. E. Buck, B. Huntley, A flexible approach to
40
Ar/39Ar geochronology through comparison with U/Pb analysis of a 1.1-Ga assessing synchroneity of past events using Bayesian reconstructions of
rhyolite. Geochim. Cosmochim. Acta 64, 73–98 (2000). doi:10.1016/S0016- sedimentation history. Quat. Sci. Rev. 27, 1872–1885 (2008).
7037(99)00204-5 doi:10.1016/j.quascirev.2008.07.009
18. K. F. Kuiper, A. Deino, F. J. Hilgen, W. Krijgsman, P. R. Renne, J. R. Wijbrans, 40. A. C. Parnell, C. E. Buck, T. K. Doan, A review of statistical chronology models for
Synchronizing rock clocks of Earth history. Science 320, 500–504 (2008). high-resolution, proxy-based Holocene palaeoenvironmental reconstruction.
doi:10.1126/science.1154339 Medline Quat. Sci. Rev. 30, 2948–2960 (2011). doi:10.1016/j.quascirev.2011.07.024
19. E. M. Niespolo, D. Rutte, A. L. Deino, P. R. Renne, Intercalibration and age of the 41. A. Deino, R. Potts, Age-probability spectra for examination of single-crystal
Alder Creek sanidine 40Ar/39Ar standard. Quat. Geochronol. 39, 205–213 (2017). 40
Ar/39Ar dating results: Examples from Olorgesailie, southern Kenya rift valley.
doi:10.1016/j.quageo.2016.09.004 Quat. Int. 7/8, 81–89 (1992).
20. J. J. Shea, The Middle Stone Age archaeology of the Lower Omo Valley Kibish 42. J. Ramos, D. Bernal, S. Dominguezbella, D. Calado, B. Ruiz, M. Gil, I. Clemente, J.
Formation: Excavations, lithic assemblages, and inferred patterns of early Homo Duran, E. Vijande, S. Chamorro, The Benzú rockshelter: A Middle Palaeolithic site
sapiens behavior. J. Hum. Evol. 55, 448–485 (2008). on the North African coast. Quat. Sci. Rev. 27, 2210–2218 (2008).
doi:10.1016/j.jhevol.2008.05.014 Medline doi:10.1016/j.quascirev.2008.08.030
21. I. McDougall, F. H. Brown, J. G. Fleagle, Stratigraphic placement and age of modern 43. J. Ramos-Muñoz, J. J. Cantillo-Duarte, D. Bernal-Casasola, A. Barrena-Tocino, S.
humans from Kibish, Ethiopia. Nature 433, 733–736 (2005). Domínguez-Bella, E. Vijande-Vila, I. Clemente-Conte, I. Gutiérrez-Zugasti, M.
First release: 15 March 2018 www.sciencemag.org (Page numbers not final at time of first release) 4
56. G. A. T. Duller, S. Tooth, L. Barham, S. Tsukamoto, New investigations at Kalambo
Falls, Zambia: Luminescence chronology, site formation, and archaeological
significance. J. Hum. Evol. 85, 111–125 (2015). doi:10.1016/j.jhevol.2015.05.003
Medline
57. R. Grün, P. Beaumont, Border Cave revisited: A revised ESR chronology. J. Hum.
Evol. 40, 467–482 (2001). doi:10.1006/jhev.2001.0471 Medline
58. R. Grün, J. S. Brink, N. A. Spooner, L. Taylor, C. B. Stringer, R. G. Franciscus, A. S.
Murray, Direct dating of Florisbad hominid. Nature 382, 500–501 (1996).
doi:10.1038/382500a0 Medline
59. K. Kuman, M. Inbar, R. J. Clarke, Palaeoenvironments and cultural sequence of the
Florisbad Middle Stone Age Hominid Site, South Africa. J. Archaeol. Sci. 26, 1409–
1425 (1999). doi:10.1006/jasc.1999.0439
60. P. Kiberd, Bundu Farm: A report on archaeological and palaeoenvironmental
assemblages from a pan site in Bushmanland, Northern Cape, South Africa. S. Afr.
Archaeol. Bull. 61, 189–201 (2006). doi:10.2307/20474927
61. A. I. R. Herries, A chronological perspective on the Acheulian and its transition to
the middle stone age in southern Africa: The question of the fauresmith. Int. J.
Evol. Biol. 2011, 961401 (2011). doi:10.4061/2011/961401 Medline
First release: 15 March 2018 www.sciencemag.org (Page numbers not final at time of first release) 5
Fig. 1. Locations, stratigraphic
relationships, and stratigraphic placement
of localities, archeological sites and dated
tuffaceous beds. (A) Olorgesailie Formation,
Member 11 in Localities A and B. M11-Tu
indices are samples 1) OL11/A-7p; 2)
A97/ALD-6p; 3) OL11/B-1p. (B) Oltulelei
Formation in Locality B and BOK sites. (C)
Oltulelei Formation in Locality G and GOK
sites. Inset maps show location of Olorgesailie
(top) and of stratigraphic locations for panels
(A), (B), and (C). Stratigraphic relationships
are based on field mapping of channel margins
and lithologies (13). Dashed lines indicate
inferred correlations between beds while solid
lines are more traceable and secure. Numbers
above columns are Lat-Long positions and
First release: 15 March 2018 www.sciencemag.org (Page numbers not final at time of first release) 6
Table 1. Eastern African MSA and Acheulean Middle Pleistocene sites with published K–Ar or 40Ar/39Ar age estimates from ~600 to ~180
ka, excluding new Olorgesailie Formation (Member 11) analyses reported in this paper. Ages are recalculated using revised K decay constants
(17), and astronomically calibrated standard ages (Fish Canyon Tuff sanidine = 28.201 Ma (18); Alder Creek Rhyolie sanidine = 1.1848 Ma (19).
First release: 15 March 2018 www.sciencemag.org (Page numbers not final at time of first release) 7
Chronology of the Acheulean to Middle Stone Age transition in eastern Africa
Alan L. Deino, Anna K. Behrensmeyer, Alison S. Brooks, John E. Yellen, Warren D. Sharp and Richard Potts
RELATED http://science.sciencemag.org/content/sci/early/2018/03/14/science.aao2200.full
CONTENT
http://science.sciencemag.org/content/sci/early/2018/03/14/science.aao2646.full
http://science.sciencemag.org/content/sci/359/6381/1200.full
REFERENCES This article cites 51 articles, 4 of which you can access for free
http://science.sciencemag.org/content/early/2018/03/14/science.aao2216#BIBL
PERMISSIONS http://www.sciencemag.org/help/reprints-and-permissions
Science (print ISSN 0036-8075; online ISSN 1095-9203) is published by the American Association for the Advancement of
Science, 1200 New York Avenue NW, Washington, DC 20005. 2017 © The Authors, some rights reserved; exclusive licensee
American Association for the Advancement of Science. No claim to original U.S. Government Works. The title Science is a
registered trademark of AAAS.