The genomic prehistory of peoples speaking Khoisan languages

Abstract

Peoples speaking so-called Khoisan languages—that is, indigenous languages of southern Africa that do not belong to the Bantu family—are culturally and linguistically diverse. They comprise herders, hunter-gatherers, as well as groups of mixed modes of subsistence and their languages are classified into three distinct language families. This cultural and linguistic variation is mirrored by extensive genetic diversity. We here review the recent genomics literature and discuss the genetic evidence for a formerly wider geographic spread of peoples with Khoisan-related ancestry, for the deep divergence among populations speaking Khoisan languages overlaid by more recent gene flow among these groups, and for the impact of admixture with immigrant food-producers in their prehistory.

Full text

UNCORRECTED MANUSCRIPT
The genomic prehistory of peoples speaking Khoisan languages
Brigitte Pakendorf1* & Mark Stoneking2
1CNRS & Université de Lyon
UMR5596, Dynamique du Langage
MSH-DDL, 14 avenue Berthelot, 69007 Lyon, France
brigitte.pakendorf@cnrs.fr
Tel : +33- 4 72 72 64 26
2Dept. of Evolutionary Genetics
MPI for Evolutionary Anthropology
Deutscher Platz 6, 04103 Leipzig, Germany
© The Author(s) 2020. Published by Oxford University Press.
Downloaded from https://academic.oup.com/hmg/advance-article/doi/10.1093/hmg/ddaa221/5930650 by guest on 24 November 2020

UNCORRECTED MANUSCRIPT
Abstract
Peoples speaking so-called Khoisan languages—that is, indigenous languages of southern
Africa that do not belong to the Bantu family—are culturally and linguistically diverse. They
comprise herders, hunter-gatherers, as well as groups of mixed modes of subsistence and their
languages are classified into three distinct language families. This cultural and linguistic
variation is mirrored by extensive genetic diversity. We here review the recent genomics
literature and discuss the genetic evidence for a formerly wider geographic spread of peoples
with Khoisan-related ancestry, for the deep divergence among populations speaking Khoisan
languages overlaid by more recent gene flow among these groups, and for the impact of
admixture with immigrant food-producers in their prehistory.
Downloaded from https://academic.oup.com/hmg/advance-article/doi/10.1093/hmg/ddaa221/5930650 by guest on 24 November 2020

UNCORRECTED MANUSCRIPT
Introduction
In this paper we use the term ‘Khoisan’ as a loose cover term to refer to the indigenous
languages of Southern Africa that do not belong to the Bantu family—and that are most
saliently characterized by their heavy use of click consonants—as well as by extension to the
genetic ancestry associated with the peoples who speak these languages. The term was coined
by the biological anthropologist Schulze in 1928 by combining the Khoekhoe herders’ term
for themselves with their term for foragers (1); variations encountered in the literature are
Khoe-San and KhoeSan. Given the fact that the peoples speaking Khoisan languages are
culturally and linguistically distinct and each has their own particular history, all umbrella
terms are flawed; it is thus of crucial importance to keep in mind that use of a single term does
not signify a unified entity.
Nowadays, peoples speaking Khoisan languages live mainly in Namibia, Botswana, and
Angola, with smaller numbers in Zambia, Zimbabwe, South Africa, Lesotho, and Eswatini (2)
(Figure 1). In South Africa, most historically known Khoisan groups have given up their
original languages and have merged with so-called Coloured populations (3). Two groups
often included in recent genetic studies are the Karretjie people, itinerant sheepshearers of the
Karoo who are probably partly descendants of ǀXam hunter-gatherers (4), and the ǂKhomani.
This latter is a group of people with diverse Khoisan-related ancestries tracing back to the
southern Kalahari, who joined together in the 1990s in order to file a land rights claim (5).
Although an initial broad classification of the languages of Africa identified a single Khoisan
phylum comprising three branches in southern Africa plus two languages—Sandawe and
Hadza—spoken in East Africa (6), nowadays specialists of these languages agree that there
are three distinct language families in southern Africa, namely Kx’a, Tuu, and Khoe-Kwadi
(1). Of these, Kx’a and Tuu might ultimately descend from a shared ancestor, but that has not
yet been conclusively demonstrated (7). The Khoe-Kwadi languages are not related to either
Downloaded from https://academic.oup.com/hmg/advance-article/doi/10.1093/hmg/ddaa221/5930650 by guest on 24 November 2020

UNCORRECTED MANUSCRIPT
the Kx’a or the Tuu languages (1,8). As for the East African languages, whereas there is no
demonstrable relationship between Hadza and any of the southern African Khoisan languages,
there is some indication that Sandawe might be related to the Khoe-Kwadi family; however
this, too, needs further corroboration (7).
Culturally, too, there is considerable heterogeneity among the Khoisan-speaking peoples of
southern Africa (9): herding groups are known historically from coastal and interior regions of
the Cape, the descendants of whom are the Nama (nowadays settled mainly in southern
Namibia) as well as several Coloured groups in South Africa (9,10). Furthermore, the Kwepe,
small-stock pastoralists from southwestern Angola, are known to have spoken Kwadi, a
language of the Khoe-Kwadi family, although this is nowadays practically extinct (9,11).
Hunter-gatherers roamed the Cape interior of South Africa in historic times and are still found
in the Kalahari region spanning Namibia, Botswana, and parts of South Africa. But there are
also groups that do not neatly fit into the herder-forager dichotomy. Foremost among these are
the Damara, a peripatetic group (12) who traditionally practiced foraging, small-scale herding
of goats, and blacksmithing in a client relationship to the Nama and the Bantu-speaking
pastoralist Herero. Along the Kavango river the Khwe rely on fishing as well as hunting and
gathering, while in the eastern Kalahari the Shua and Tshwa are transitioning to food
production and are in a client relationship to their Bantu neighbours in addition to their
foraging subsistence.
Given this linguistic and cultural diversity, it is clear that the prehistory of the peoples
speaking Khoisan languages must have been highly complex. Numerous studies over the past
decade have highlighted the considerable genetic diversity found in these groups, the deep
divergence among them as well as between them and other African groups, and the impact of
successive waves of migration of food-producing peoples from East and West-Central Africa
as well as in historical times of European colonizers (see (13) for a recent review). We here
Downloaded from https://academic.oup.com/hmg/advance-article/doi/10.1093/hmg/ddaa221/5930650 by guest on 24 November 2020

UNCORRECTED MANUSCRIPT
survey the recent literature and discuss the genetic evidence for an erstwhile wider geographic
spread of peoples with Khoisan-related ancestry, for the deep divergence among populations
speaking Khoisan languages overlaid by more recent gene flow among these groups, and for
the impact of admixture with immigrant food-producers in their prehistory. For convenience
we will refer to peoples speaking a Khoisan language as a Khoisan-speaking group, and to
peoples speaking a Bantu language as a Bantu-speaking group (although of course each group
speaks a particular language belonging to the Kx’a, Tuu, or Khoe-Kwadi families for
Khoisan-speaking groups, or to the Bantu family). Throughout the paper we follow the
nomenclature of Güldemann (1), irrespective of the spelling of group names found in
individual articles.
A wider geographic spread in prehistoric times
Recent genome-wide analyses of DNA from ancient human remains in East Africa have
demonstrated the presence in the past of Khoisan-related ancestry in regions as distant from
the Kalahari as Tanzania and Kenya (throughout this review, for convenience we use present-
day countries to refer to the location of ancient remains that predate country formation). Thus,
approximately 60% of the ancestry of ancient remains from Malawi dated to between 2500
and 8100 BP and ~30% of the ancestry of a 1400-year-old individual from Tanzania is related
to ancestry detectable both in 2000-year-old hunter-gatherer remains from South Africa and
modern Juǀ’hoan (14). Similarly, an ancient individual from Kenya dated to 3500 BP shows
evidence of low levels of Khoisan-related ancestry (15). In addition, there is evidence from
whole genome sequences from modern populations for potential long-distance migrations
involving Khoisan-speaking groups, e.g. in the sharing of private alleles between the Juǀ’hoan
and Mbuti central African rain forest foragers (16).
Downloaded from https://academic.oup.com/hmg/advance-article/doi/10.1093/hmg/ddaa221/5930650 by guest on 24 November 2020

UNCORRECTED MANUSCRIPT
Interestingly, the Khoisan-related ancestry in eastern Africa is related in equal degrees to the
deeply diverging lineages identified in modern-day Khoisan-speaking populations (see
below), implying that the Khoisan-related groups settled in eastern Africa were genetically
distinct from those living in southern Africa (14). These results mirror the results of mtDNA
analyses that found a complementary distribution of one of the Khoisan-specific haplogroups,
L0k. Of three deeply divergent branches (L0k1a, L0k1b, and L0k2), only L0k1a is found
among extant Khoisan-speaking groups of Namibia and Botswana, while L0k1b and L0k2 are
found practically exclusively in Bantu-speaking populations settled in Zambia. This implies
that peoples genetically related to currently known Khoisan-speaking groups, yet carrying
distinct lineages, were resident in regions beyond those previously attested (17). There are no
historically known Khoisan-speaking groups in either Zambia or Malawi or further northeast,
and modern-day populations of Malawi show no traces of Khoisan-related ancestry. It is thus
clear that the incoming Bantu-speaking populations must have replaced the Khoisan-related
autochthonous populations with hardly any admixture. Linguistic analyses, too, show that
some Bantu languages of the Kavango-Zambezi transfrontier area borrowed words with click
consonants from Khoisan languages that are nowadays extinct, in addition to borrowing
words from Khwe and Ju languages (18).
High levels of genetic diversity in Khoisan-speaking peoples
Khoisan-speaking groups are consistently found to harbour high levels of genetic diversity.
Several recent studies of whole genome sequences found highest levels of genetic diversity in
Khoisan-speaking individuals (16,19–21), and these individuals also have the highest
frequencies, on average, of population-specific CNV variants worldwide (22). However,
sample sizes and ethnolinguistic diversity of the groups analyzed remain quite limited (Table
Downloaded from https://academic.oup.com/hmg/advance-article/doi/10.1093/hmg/ddaa221/5930650 by guest on 24 November 2020

UNCORRECTED MANUSCRIPT
1); there is a clear need for additional whole genome sequence studies of further Khoisan-
speaking groups.
Khoisan-speaking groups are also the first to branch off in genomic studies of African or
world-wide populations (16,19,20,23), with their divergence from other populations dated to
160-300 kya (Table 1). The fact that Khoisan-related lineages are the first to diverge has
sometimes been erroneously interpreted as strong evidence for an origin of modern humans in
southern Africa (e.g. (24), which is based solely on mtDNA lineages; see (25) and (26) for
substantial critiques of this paper). However, as noted above, Khoisan-related groups were
formerly more widespread, and moreover the divergence between Khoisan-speaking groups
and other African groups could, in principle, have occurred anywhere in Africa.
Khoisan-speaking peoples also show evidence of a larger effective population size over time
than other African populations (16,19,20,27). All human populations show a signal of
decreasing effective population size beginning around the time of the divergence of African
from non-African populations, ~50-100 kya; however, Khoisan-speaking groups show less of
a reduction in effective population size than do other populations (16,19,20). Some of this
diversity might be due to archaic admixture from an as yet undiscovered population (28). For
example, whole-genome sequencing (20) suggests approximately 4% ancestry from an
archaic ‘ghost’ population in the four Khoisan-speaking individuals analyzed.
In addition to carrying considerable amounts of genetic diversity, Khoisan-speaking
populations are also quite diverged from one another, as shown by a deep split between
populations residing in the northwestern Kalahari and those from the southeastern Kalahari or
South Africa (29,30), respectively). Recent reanalyses of these data together with some new
data (31)—providing the most complete geographical coverage of extant Khoisan-speaking
groups—that focus solely on genomic segments of Khoisan-related ancestry have
demonstrated a tripartite split into northern, central, and southern Khoisan-speaking groups
Downloaded from https://academic.oup.com/hmg/advance-article/doi/10.1093/hmg/ddaa221/5930650 by guest on 24 November 2020

UNCORRECTED MANUSCRIPT
(roughly corresponding to Pickrell et al’s (29) northwestern and southwestern and Schlebusch
et al’s (30) southern groups, respectively, and also corresponding to groups defined by
ecogeographic boundaries in (32); Figure 2). It should be noted that these northern, central,
and southern genetic groupings do not correspond to a previous linguistic classification of
Southern African Khoisan languages into Northern, Central, and Southern branches (6): the
northwestern/northern grouping includes not only the !Xuun and Juǀ’hoan, whose languages
belong to the Kx’a family, but also the Haiǁom whose language belongs to the Khoe family;
the southeastern/central grouping includes populations speaking languages belonging to all
three families, and the southern grouping includes both the pastoralist Nama, whose language
belongs to the Khoe family, and the descendants of foragers, the Karretjie and ǂKhomani,
whose heritage languages belonged to the Tuu family.
However, there is uncertainty as to when this deep split occurred. Initial studies based on
genome-wide SNP array data dated the divergence to ~25-35 kya (29–31). The inclusion of
genome sequence data from a 2000-year-old hunter-gatherer individual from South Africa
pushed back the date of divergence between ‘northern’ and ‘southern’ Khoisan to 156-185
kya (23), which might suggest that the SNP array data under-estimate divergence times due to
ascertainment bias. Nevertheless, subsequent estimates based on whole genome sequences
range from ~30 kya to ~160 kya (Table 1); different mutation rates can account for some
differences in these estimates, but not all, leaving this an open question.
In any event, while this deep divergence between northern, central, and southern Khoisan-
speaking populations suggests that they must have been isolated from each other for a
considerable period of time, there is also evidence for gene flow among Khoisan-speaking
groups taking place at a more recent time-scale. This is shown by analyses focusing on
genome segments of Khoisan-specific ancestry that show a high correlation of genetic with
geographic distances, and a clear signal of isolation by distance (31,33). It is therefore
Downloaded from https://academic.oup.com/hmg/advance-article/doi/10.1093/hmg/ddaa221/5930650 by guest on 24 November 2020

UNCORRECTED MANUSCRIPT
possible that the deep divergence times arise purely as a consequence of long-distance
separation in what is actually a gradient of relatedness. However, it is also possible that the
signals of isolation by distance reflect more recent processes after initial older divergence
events. In particular, it has been suggested that Khoisan-speaking groups were initially split
by the prehistoric lake Makgadigadi, with gene flow being reinitiated when the lake dried up
around 10 kya (34). One group in particular that shows evidence for admixture are the Naro,
who are both geographically (Figure 1) and genetically (Figure 2) intermediate between the
northwestern/northern and southeastern/central groupings (29,31) and who also show
evidence for gene flow from the Gǀui and an ethnolinguistically undefined group from Xade in
the Central Kalahari Game Reserve (33). Additionally, the ǂHoan, who speak a divergent
language of the Kx’a family nowadays called ǂ’Amkoe, show only 5% shared ancestry with
their linguistic relatives the !Xuun and the Juǀ’hoan (33), while they are genetically close to
the neighbouring Taa (who speak a Tuu language) and the Gǀui, whose language belongs to
the Khoe family (31; cf. 34,35). Distinguishing between long-term isolation by distance, vs.
deep divergence followed by more recent contact, may be possible when more whole-genome
sequence data become available.
Admixture with immigrating food-producing populations
In addition to gene flow among Khoisan-speaking groups, these have also undergone variable
amounts of admixture from immigrating food-producers (23,36) (Figure 3). Sheep and goat
pastoralists are thought to have immigrated to southern Africa from East Africa a few
centuries before the immigration of Iron Age agropastoralists commonly associated with the
expansion of Bantu-speaking peoples into large parts of sub-Saharan Africa (37). The
presence among southern African populations of the Lactase Persistence variant C-14010
(30,38,39), which is of probable East African origin (40), points towards a demic diffusion of
Downloaded from https://academic.oup.com/hmg/advance-article/doi/10.1093/hmg/ddaa221/5930650 by guest on 24 November 2020

UNCORRECTED MANUSCRIPT
pastoralism into southern Africa. Significantly higher frequencies of this variant in pastoralist
populations than in foragers, and in groups speaking languages of the Khoe family than in
Tuu- or Kx’a-speaking populations (39), support the hypothesis that the Khoe-Kwadi
languages were brought to southern Africa by a migration of pastoralists from East Africa (8).
Unexpectedly, however, the formerly Khoe-Kwadi-speaking pastoralist Kwepe from
southwestern Angola have only low frequencies of this allele (41), in accordance with their
low frequency of the East African Y-chromosome haplogroup E-M293. The spread of
pastoralism and the Khoe-Kwadi languages is therefore likely to been a complex process,
which might also have involved shift of Bantu-speaking groups to Khoe-Kwadi languages
(42). Interestingly, two modern-day forager groups, the Gǀui and the Tshwa, show evidence
for ongoing positive selection for the C-14010 allele, indicating a possibly recent reversion
from a herding way of life to foraging (39). Ancient DNA analyses have provided further
direct evidence for admixture with East African pastoralists: a 1200-year-old specimen found
in a herder context in the western Cape was shown to have approximately 40% ancestry
related to an early pastoralist from Tanzania and about 60% ancestry related to 2000-year-old
South African foragers (14). Two Early Iron Age individuals from Botswana—who are likely
to have spoken Bantu languages—confirm the earlier presence of East African pastoralists
than Iron Age agropastoralists in the region, since they carry ancestry related to the 1200-
year-old admixed herder from the western Cape (15).
The admixture with food producing populations did not take place at the same time or to the
same extent across southern Africa (29,36). Analyses of uniparental data show a strongly sex-
biased signal of gene flow in southern Africa, with Khoisan-speaking populations receiving
paternal lineages from food-producers, while Bantu-speaking groups incorporated mainly
Khoisan-related maternal lineages. The intensity of this sex bias increases from North to
South, possibly indicating changes in social interactions between immigrating groups and
Downloaded from https://academic.oup.com/hmg/advance-article/doi/10.1093/hmg/ddaa221/5930650 by guest on 24 November 2020

UNCORRECTED MANUSCRIPT
autochthonous peoples over time (35). Such changes in interactions are also implied by the
varying levels of Khoisan-related ancestry detectable in modern-day Bantu-speaking
populations of southern Africa: populations from Malawi do not show any evidence for
Khoisan-related ancestry (14), and populations from southern Mozambique show only low
levels of such ancestry (4-5% maximum (43)). This is in contrast to populations such as the
Kgalagadi and Tswana from Botswana with 33-39% and 22-24% Khoisan-related ancestry,
respectively (29,36), or the Sotho, Xhosa and Zulu from South Africa with between ~10-24%
Khoisan-related ancestry (43,44). Such changes in social interactions between immigrating
Iron Age agropastoralists and resident Khoisan-speaking populations might also explain
variable patterns of click borrowing in Bantu languages (18,45).
Ethical considerations
Indigenous communities are playing an increasingly prominent role in genomics research,
going beyond merely providing samples to being fully informed about the results and how
they are presented (46,47). Even well-meaning scientists engaged in research on indigenous
peoples can fail to appreciate how their scientific statements about their results may be
viewed and interpreted by the individuals and communities studied—a prominent example
involved a study that sequenced the genomes of four Khoisan-speaking individuals (cf. (48)).
One outcome of such misunderstandings was the establishment of the San Code of Research
Ethics in 2017 (https://www.globalcodeofconduct.org/affiliated-codes/), the first such ethics
code by an indigenous African group, and a model for research involving Khoisan-speaking
groups. Nonetheless, ethical difficulties continue to arise (e.g., (49)).
Downloaded from https://academic.oup.com/hmg/advance-article/doi/10.1093/hmg/ddaa221/5930650 by guest on 24 November 2020

UNCORRECTED MANUSCRIPT
Conclusion
The stereotypical image of Khoisan-speaking peoples as Stone Age hunter-gatherers who
have lived in splendid isolation since the dawn of humankind can, without any doubt, be laid
to rest. These groups exhibit extensive cultural, linguistic, and biological diversity. They
harbor more genetic diversity, the earliest divergences, and larger effective population sizes
than other human populations. They used to be more widespread in former times, are likely to
have engaged in long-distance migrations, and they have both influenced and been influenced
by at least two migrations, an earlier migration of pastoralists from eastern Africa, and a later
migration of agropastoralists associated with the spread of Bantu languages. Understanding
the complex genomic history and structure of Khoisan-speaking populations has important
implications not only for their individual histories and the history of humans in general, but
also for potential variation in disease susceptibility (cf. (50,51)). There is a clear need for
further whole genome sequence studies of Khoisan-speaking groups, in order to achieve these
goals.
Downloaded from https://academic.oup.com/hmg/advance-article/doi/10.1093/hmg/ddaa221/5930650 by guest on 24 November 2020

UNCORRECTED MANUSCRIPT
Acknowledgements:
BP is grateful to the LABEX ASLAN (ANR-10-LABX-0081) of Université de Lyon for its
financial support within the program "Investissements d'Avenir" (ANR-11-IDEX-0007) of the
French government operated by the National Research Agency (ANR). MS acknowledges
support from the Max Planck Society. We thank Linda Schymanski for help with the figures.
Conflict of interest:
The authors declare no conflict of interest.
Downloaded from https://academic.oup.com/hmg/advance-article/doi/10.1093/hmg/ddaa221/5930650 by guest on 24 November 2020

UNCORRECTED MANUSCRIPT
References:
1. Güldemann, T. (2014) ‘Khoisan’ linguistic classification today. In Güldemann, T., Fehn, A.-M.
(eds.), Beyond ‘Khoisan’. Historical relations in the Kalahari Basin, CILT, John Benjamins,
Amsterdam, pp. 1–40.
2. Hitchcock, R. K. (2020) The Plight of the Kalahari San: Hunter-Gatherers in a Globalized World.
Journal of Anthropological Research, 76, 164–184.
3. Schenck, M. (2008) Land, Water, Truth, and Love - Visions of Identity and Land Access: From
Bain’s Bushmen to Khomani San. Thesis.
4. Traill, A. (2007) !Khwa-ka Hhouiten Hhouiten ‘The rush of the storm’: The linguistic death of
|Xam. In Skotnes, P. (ed.), Claim to the Country: the Archive of Lucy Lloyd and Wilhelm Bleek,
Ohio University Press, Athens, OH, pp. 130–147.
5. South African Human Rights Commission (2004) Report on the inquiry into human rights
violations in the Khomani San community. Andriesvale-Askham area, Kalahari (2004) .
6. Greenberg, J. H. (1963) The languages of Africa; Indiana University Center in Anthropology,
Folklore and Linguistics, and The Hague: Mouton, Bloomington, (1963) .
7. Güldemann, T. (2018) Historical linguistics and genealogical language classification in Africa. In
Güldemann, T. (ed.), The languages and linguistics of Africa, The World of Linguistics, De
Gruyter Mouton, Berlin, Boston, pp. 58–444.
8. Güldemann, T. (2008) A linguist’s view: Khoe-Kwadi speakers as the earliest food-producers of
southern Africa. South. Afr. Humanit., 20, 93–132.
9. Barnard, A. (1992) Hunters and Herders of Southern Africa. A Comparative Ethnography of the
Khoisan Peoples.; Cambridge Studies in Social and Cultural Anthropology; Cambridge University
Press, Cambridge, (1992) .
10. Robins, S. (2000) Land struggles and the politics and ethics of representing ‘bushman’ history and
identity. Kronos, 56–75.
Downloaded from https://academic.oup.com/hmg/advance-article/doi/10.1093/hmg/ddaa221/5930650 by guest on 24 November 2020

UNCORRECTED MANUSCRIPT
11. Oliveira, S., Fehn, A.-M., Aço, T., Lages, F., Gayà‐Vidal, M., Pakendorf, B., Stoneking, M. and
Rocha, J. (2018) Matriclans shape populations: Insights from the Angolan Namib Desert into the
maternal genetic history of southern Africa. Am. J.Phys. Anthropol., 165, 518–535.
12. Bollig, M. (2005) Singing smiths and hunting ritual entrepreneurs: Transitions between forager
and peripatetic communities in Africa. In Berland, J. C., Rao, A. (eds.), Customary Strangers.
New Perspectives on Peripatetic Peoples in the Middle East, Africa and Asia, Praeger, New York,
pp. 195–232.
13. Montinaro, F. and Capelli, C. (2018) The evolutionary history of Southern Africa. Curr. Opin.
Genet. Dev., 53, 157–164.
14. Skoglund, P., Thompson, J. C., Prendergast, M. E., Mittnik, A., Sirak, K., Hajdinjak, M., Salie, T.,
Rohland, N., Mallick, S., Peltzer, A., et al. (2017) Reconstructing Prehistoric African Population
Structure. Cell, 171, 59-71.e21.
15. Wang, K., Goldstein, S., Bleasdale, M., Clist, B., Bostoen, K., Bakwa-Lufu, P., Buck, L. T.,
Crowther, A., Dème, A., McIntosh, R. J. et al. (2020) Ancient genomes reveal complex patterns of
population movement, interaction, and replacement in sub-Saharan Africa. Science Advances, 6,
eaaz0183.
16. Schlebusch, C. M., Sjödin, P., Breton, G., Günther, T., Naidoo, T., Hollfelder, N., Sjöstrand, A.
E., Xu, J., Gattepaille, L. M., Vicente, M., et al. (2020) Khoe-San Genomes Reveal Unique
Variation and Confirm the Deepest Population Divergence in Homo sapiens. Mol. Biol. Evol.,
Advance Online https://doi.org/10.1093/molbev/msaa140.
17. Barbieri, C., Vicente, M., Rocha, J., Mpoloka, S. W., Stoneking, M. and Pakendorf, B. (2013)
Ancient substructure in early mtDNA lineages of southern Africa. Am. J. Hum. Genet., 92, 285–
292.
18. Gunnink H., Sands B., Pakendorf B. and Bostoen, K. (2015) Prehistoric language contact in the
Kavango-Zambezi transfrontier area: Khoisan influence on southwestern Bantu languages. J. Afr.
Lang. Linguist., 36, 193–232.
Downloaded from https://academic.oup.com/hmg/advance-article/doi/10.1093/hmg/ddaa221/5930650 by guest on 24 November 2020

UNCORRECTED MANUSCRIPT
19. Fan, S., Kelly, D. E., Beltrame, M. H., Hansen, M. E. B., Mallick, S., Ranciaro, A., Hirbo, J.,
Thompson, S., Beggs, W., Nyambo, T., et al. (2019) African evolutionary history inferred from
whole genome sequence data of 44 indigenous African populations. Genome Biol., 20, 82.
20. Lorente-Galdos, B., Lao, O., Serra-Vidal, G., Santpere, G., Kuderna, L. F. K., Arauna, L. R.,
Fadhlaoui-Zid, K., Pimenoff, V. N., Soodyall, H., Zalloua, P., et al. (2019) Whole-genome
sequence analysis of a Pan African set of samples reveals archaic gene flow from an extinct basal
population of modern humans into sub-Saharan populations. Genome Biol., 20, 77.
21. Bergström, A., McCarthy, S. A., Hui, R., Almarri, M. A., Ayub, Q., Danecek, P., Chen, Y.,
Felkel, S., Hallast, P., Kamm, J., et al. (2020) Insights into human genetic variation and
population history from 929 diverse genomes. Science, 367, eaay5012.
22. Almarri, M. A., Bergström, A., Prado-Martinez, J., Yang, F., Fu, B., Dunham, A. S., Chen, Y.,
Hurles, M. E., Tyler-Smith, C., Xue, Y. (2020) Population Structure, Stratification, and
Introgression of Human Structural Variation. Cell, 182, 189-199.e15.
23. Schlebusch, C. M., Malmström, H., Günther, T., Sjödin, P., Coutinho, A., Edlund, H., Munters, A.
R., Vicente, M., Steyn, M., Soodyall, H., et al. (2017) Southern African ancient genomes estimate
modern human divergence to 350,000 to 260,000 years ago. Science, 358, 652–655.
24. Chan, E. K. F., Timmermann, A., Baldi, B. F., Moore, A. E., Lyons, R. J., Lee, S.-S., Kalsbeek,
A. M. F., Petersen, D. C., Rautenbach, H., Förtsch, H. E. A., et al. (2019) Human origins in a
southern African palaeo-wetland and first migrations. Nature, 575, 185–189.
25. Ackermann, R. R., Athreya, S., Black, W., Cabana, G. S., Hare, V., Pickering, R. and Schroeder,
L. (2019) Upholding ‘good Science’ in Human Origins Research: A Response to Chan Et Al.
AfricArXiv. November 6. doi:10.31730/osf.io/qtjfp.
26. Schlebusch, C. M., Loog, L., Groucutt, H. S., King, T., Rutherford, A., Barbieri, C., Barbujani,
G., Chikhi, L., Jakobsson, M., Eriksson, A., et al. (2019) Human origins in southern african
Palaeo-wetlands?: strong claims from weak evidence.
https://www.preprints.org/manuscript/201911.0193/v1.
Downloaded from https://academic.oup.com/hmg/advance-article/doi/10.1093/hmg/ddaa221/5930650 by guest on 24 November 2020

UNCORRECTED MANUSCRIPT
27. Kim, H. L., Ratan, A., Perry, G. H., Montenegro, A., Miller, W. and Schuster, S. C. (2014)
Khoisan hunter-gatherers have been the largest population throughout most of modern-human
demographic history. Nat. Commun., 5, 5692.
28. Hammer, M. F., Woerner, A. E., Mendez, F. L., Watkins, J. C. and Wall, J. D. (2011) Genetic
evidence for archaic admixture in Africa. P. Natl. Acad. Sci. USA, 108, 15123–15128.
29. Pickrell, J. K., Patterson, N., Barbieri, C., Berthold, F., Gerlach, L., Güldemann, T., Kure, B.,
Mpoloka, S. W., Nakagawa, H., Naumann, C. et al. (2012) The genetic prehistory of southern
Africa. Nat. Commun., 3, 1143.
30. Schlebusch, C. M., Skoglund, P., Sjödin, P., Gattepaille, L. M., Hernandez, D., Jay, F., Li, S., De
Jongh, M., Singleton, A., Blum, M. G. B., et al. (2012) Genomic variation in seven Khoe-San
groups reveals adaptation and complex African history. Science, 338, 374–379.
31. Montinaro, F., Busby, G. B. J., Gonzalez-Santos, M., Oosthuitzen, O., Oosthuitzen, E.,
Anagnostou, P., Destro-Bisol, G., Pascali, V. L. and Capelli, C. (2017) Complex Ancient Genetic
Structure and Cultural Transitions in Southern African Populations. Genetics, 205, 303–316.
32. Uren, C., Kim, M., Martin, A. R., Bobo, D., Gignoux, C. R., Helden, P. D. van, Möller, M., Hoal,
E. G. and Henn, B. M. (2016) Fine-Scale Human Population Structure in Southern Africa Reflects
Ecogeographic Boundaries. Genetics, 204, 303–314.
33. Vicente, M., Jakobsson, M., Ebbesen, P. and Schlebusch, C. M. (2019) Genetic Affinities among
Southern Africa Hunter-Gatherers and the Impact of Admixing Farmer and Herder Populations.
Mol. Biol. Evol., 36, 1849–1861.
34. Barbieri, C., Güldemann, T., Naumann, C., Gerlach, L., Berthold, F., Nakagawa, H., Mpoloka, S.
W., Stoneking, M. and Pakendorf, B. (2014) Unraveling the complex maternal history of Southern
African Khoisan populations. Am. J. Phys. Anthropol., 153, 435–448.
35. Bajić, V., Barbieri, C., Hübner, A., Güldemann, T., Naumann, C., Gerlach, L., Berthold, F.,
Nakagawa, H., Mpoloka, S. W., Roewer, L., et al. (2018) Genetic structure and sex-biased gene
flow in the history of southern African populations. Am. J. Phys. Anthropol., 167, 656–671.
Downloaded from https://academic.oup.com/hmg/advance-article/doi/10.1093/hmg/ddaa221/5930650 by guest on 24 November 2020

UNCORRECTED MANUSCRIPT
36. Pickrell, J. K., Patterson, N., Loh, P.-R., Lipson, M., Berger, B., Stoneking, M., Pakendorf, B. and
Reich, D. (2014) Ancient west Eurasian ancestry in southern and eastern Africa. P. Natl. Acad.
Sci. USA, 111, 2632–2637.
37. Lander, F. and Russell, T. (2018) The archaeological evidence for the appearance of pastoralism
and farming in southern Africa. PLOS ONE, 13, e0198941.
38. Breton, G., Schlebusch, C. M., Lombard, M., Sjödin, P., Soodyall, H. and Jakobsson, M. (2014)
Lactase persistence alleles reveal partial East african ancestry of southern african khoe
pastoralists. Curr. Biol., 24, 852–858.
39. Macholdt, E., Lede, V., Barbieri, C., Mpoloka, S. W., Chen, H., Slatkin, M., Pakendorf, B. and
Stoneking, M (2014) Tracing Pastoralist Migrations to Southern Africa with Lactase Persistence
Alleles. Curr. Biol., 24, 875–879.
40. Tishkoff, S. A., Reed, F. A., Ranciaro, A., Voight, B. F., Babbitt, C. C., Silverman, J. S., Powell,
K., Mortensen, H. M., Hirbo, J. B., Osman, M., et al. (2007) Convergent adaptation of human
lactase persistence in Africa and Europe. Nat. Genet., 39, 31–40.
41. Pinto, J. C., Oliveira, S., Teixeira, S., Martins, D., Fehn, A.-M., Aço, T., Gayà‐Vidal, M. and
Rocha, J. (2016) Food and pathogen adaptations in the Angolan Namib desert: Tracing the spread
of lactase persistence and human African trypanosomiasis resistance into southwestern Africa.
Am. J. Phys. Anthropol., 161, 436–447.
42. Oliveira, S., Hübner, A., Fehn, A.-M., Aço, T., Lages, F., Pakendorf, B., Stoneking, M. and
Rocha, J. (2019) The role of matrilineality in shaping patterns of Y chromosome and mtDNA
sequence variation in southwestern Angola. Eur. J. Hum. Genet., 27, 475–483.
43. Semo, A., Gayà-Vidal, M., Fortes-Lima, C., Alard, B., Oliveira, S., Almeida, J., Prista, A.,
Damasceno, A., Fehn, A.-M., Schlebusch, C., et al. (2020) Along the Indian Ocean Coast:
Genomic Variation in Mozambique Provides New Insights into the Bantu Expansion. Mol. Biol.
Evol., 37, 406–416.
44. Choudhury, A., Ramsay, M., Hazelhurst, S., Aron, S., Bardien, S., Botha, G., Chimusa, E. R.,
Christoffels, A., Gamieldien, J., Sefid-Dashti, M. J., et al. (2017) Whole-genome sequencing for
an enhanced understanding of genetic variation among South Africans. Nat. Commun., 8, 2062.
Downloaded from https://academic.oup.com/hmg/advance-article/doi/10.1093/hmg/ddaa221/5930650 by guest on 24 November 2020

UNCORRECTED MANUSCRIPT
45. Pakendorf, B., Gunnink, H., Sands, B. and Bostoen, K. (2017) Prehistoric Bantu-Khoisan
language contact. Language Dynamics and Change, 7, 1–46.
46. Garrison, N. A., Hudson, M., Ballantyne, L. L., Garba, I., Martinez, A., Taualii, M., Arbour, L.,
Caron, N. R. and Rainie, S. C. (2019) Genomic Research Through an Indigenous Lens:
Understanding the Expectations. Annu. Rev. Genom. Hum. G., 20, 495–517.
47. Hudson, M., Garrison, N. A., Sterling, R., Caron, N. R., Fox, K., Yracheta, J., Anderson, J.,
Wilcox, P., Arbour, L., Brown, A., et al. (2020) Rights, interests and expectations: Indigenous
perspectives on unrestricted access to genomic data. Nat. Rev. Genet., 21, 377–384.
48. Chennells, R. and Steenkamp, A. (2018) International genomics research involving the San
people. In Schroeder, D., Cook, J., Hirsch, F., et al. (eds.), Ethics dumping. Case studies from
North-South research collaborations, Springer Briefs in Research and Innovation Governance,
Springer, Cham, pp. 15–22.
49. Stokstad, E. (2019) Genetics lab accused of misusing African DNA. Science. 2019, pp. 555–556.
50. Thami, P. K. and Chimusa, E. R. (2019) Population Structure and Implications on the Genetic
Architecture of HIV-1 Phenotypes Within Southern Africa. Front. Genet., 10, 905.
51. Swart, Y., van Eeden, G., Sparks, A., Uren, C. and Möller, M. (2020) Prospective avenues for
human population genomics and disease mapping in southern Africa. Mol. Genet. Genomics, 295,
1079–1089.
52. Schlebusch, C. M., de Jongh, M. and Soodyall, H. (2011) Different contributions of ancient
mitochondrial and Y-chromosomal lineages in ‘Karretjie people’ of the Great Karoo in South
Africa. J. Hum. Genet., 56, 623–630.
53. Mallick, S., Li, H., Lipson, M., Mathieson, I., Gymrek, M., Racimo, F., Zhao, M., Chennagiri, N.,
Nordenfelt, S., Tandon, A., et al. (2016) The Simons Genome Diversity Project: 300 genomes
from 142 diverse populations. Nature, 538, 201–206.
54. Meyer, M., Kircher, M., Gansauge, M.-T., Li, H., Racimo, F., Mallick, S., Schraiber, J. G., Jay,
F., Prüfer, K., de Filippo, C., et al. (2012) A high-coverage genome sequence from an archaic
Denisovan individual. Science, 338, 222–226.
Downloaded from https://academic.oup.com/hmg/advance-article/doi/10.1093/hmg/ddaa221/5930650 by guest on 24 November 2020

UNCORRECTED MANUSCRIPT
55. Schuster, S. C., Miller, W., Ratan, A., Tomsho, L. P., Giardine, B., Kasson, L. R., Harris, R. S.,
Petersen, D. C., Zhao, F., Qi, J., et al. (2010) Complete Khoisan and Bantu genomes from
southern Africa. Nature, 463, 943–947.
Downloaded from https://academic.oup.com/hmg/advance-article/doi/10.1093/hmg/ddaa221/5930650 by guest on 24 November 2020

UNCORRECTED MANUSCRIPT
Figure Legends
Figure 1: Map showing the approximate location of Khoisan-speaking groups, based on
ethnolinguistic data from (1) and (2), and information on the Karretjie from (52). Colours
indicate the language family affiliation: blue = Kx’a, red = Tuu, green = Khoe-Kwadi.
Languages that are extinct are indicated by crosses. Some of the ǂKhomani still remember
Nǀuu. ǂ’Amkoe is the actual name of the language spoken by the ǂHoan, but since the initial
publication presenting genetic data from this group referred to them by the old language
name, this has been maintained in genetic publications.
Downloaded from https://academic.oup.com/hmg/advance-article/doi/10.1093/hmg/ddaa221/5930650 by guest on 24 November 2020

UNCORRECTED MANUSCRIPT
Figure 2: Plot of the first two dimensions of a multidimensional scaling analysis of Khoisan-
speaking groups, illustrating the northern, central, and southern genetic groupings of Khoisan-
speaking groups. The plot is based on genome-wide SNP array data with non-Khoisan-related
ancestry masked. The vertical axis is the first dimension and the horizontal axis is the second
dimension. The contours depict 90% utilization distribution densities, i.e. the smallest area of
the plot in which there is a 90% probability of locating the individuals from the same group,
and are color-coded according to language family: blue, Kx’a; red, Tuu; green, Khoe.
Modified from (31), which should be consulted for further details.
Downloaded from https://academic.oup.com/hmg/advance-article/doi/10.1093/hmg/ddaa221/5930650 by guest on 24 November 2020

UNCORRECTED MANUSCRIPT
Figure 3: Variable amounts of Khoisan-related, Bantu-related, and East African-related
ancestries in Khoisan-speaking groups, based on (36).
Downloaded from https://academic.oup.com/hmg/advance-article/doi/10.1093/hmg/ddaa221/5930650 by guest on 24 November 2020

UNCORRECTED MANUSCRIPT
Table 1. Salient features of recent studies of whole genome sequences that included Khoisan-
speaking groups.
Study
Sample sizes
and groups
Deepest divergence time
between Khoisan-speaking
and other groups
Divergence time among
Khoisan-speaking groups
Fan et al. 2019
(19)
4 Juǀ’hoan1
2 ǂKhomani1
~200 kya
~30 kya
Lorente-Galdos
et al. 2019 (20)
2 Juǀ’hoan2
1 Taa3
1 ǂKhomani
~190 kya
n/a
Bergström et al.
2020 (21)
6 Juǀ’hoan4
~162 kya
n/a
Schlebusch et al.
2020 (16)
5 Juǀ’hoan
5 ǂKhomani
5 Nama
5 Karretjie
5 Gǀui/Gǁana5
~200-300 kya
~160-190 kya
1sequences from (53); Juǀ’hoan samples from HGDP
21 sequence from (53) and 1 sequence from (54); samples from HGDP
3sequence from (55), where this individual (KB1) is labelled a Tuu-speaker
4all samples from HGDP and include the 4 Juǀ’hoan from (53)
5mixed group of Gǀui and Gǁana individuals, see (30) for details
Downloaded from https://academic.oup.com/hmg/advance-article/doi/10.1093/hmg/ddaa221/5930650 by guest on 24 November 2020