July 25, 2013

Madagascar inhabited long before Austronesian arrival

These findings revolutionize the understanding of the prehistory of the island, until now believed to have remained uninhabited until the arrival of peoples of Austronesian stock some 1500 years ago (with some scattered evidence of earlier inhabitation but nothing too conclusive). New archaeological data pushes back the first colonization period to some 4000 years ago, a time when Malayo-Polynesian culture was still restricted to, roughly, the Philippine archipelago.

Robert E. Deward et al., Stone tools and foraging in northern Madagascar challenge Holocene extinction models. PNAS 2013. Pay per view (6-month embargo) → LINK [doi:10.1073/pnas.1306100110]


Past research on Madagascar indicates that village communities were established about AD 500 by people of both Indonesian and East African heritage. Evidence of earlier visits is scattered and contentious. Recent archaeological excavations in northern Madagascar provide evidence of occupational sites with microlithic stone technologies related to foraging for forest and coastal resources. A forager occupation of one site dates to earlier than 2000 B.C., doubling the length of Madagascar’s known occupational history, and thus the time during which people exploited Madagascar’s environments. We detail stratigraphy, chronology, and artifacts from two rock shelters. Ambohiposa near Iharana (Vohémar) on the northeast coast, yielded a stratified assemblage with small flakes, microblades, and retouched crescentic and trapezoidal tools, probably projectile elements, made on cherts and obsidian, some brought more that 200 km. 14C dates are contemporary with the earliest villages. No food remains are preserved. Lakaton’i Anja near Antsiranana in the north yielded several stratified assemblages. The latest assemblage is well dated to A.D. 1050–1350, by 14C and optically stimulated luminescence dating and pottery imported from the Near East and China. Below is a series of stratified assemblages similar to Ambohiposa. 14C and optically stimulated luminescence dates indicate occupation from at least 2000 B.C. Faunal remains indicate a foraging pattern. Our evidence shows that foragers with a microlithic technology were active in Madagascar long before the arrival of farmers and herders and before many Late Holocene faunal extinctions. The differing effects of historically distinct economies must be identified and understood to reconstruct Holocene histories of human environmental impact.

Notice that this colonization is also older than the Bantu expansion and therefore these settlers must have been pre-Bantu peoples of East African roots.

Source: Al-Hakawati
The sites are located in the North tip of the island, what is consistent with arrival through Comoros, the most natural route between East Africa and Madagascar, which  requires the sailing of some 190 miles (~350 Km) of open sea.

Otherwise the narrowest extent of the Mozambique Channel, between Angoche and Tambohorano, is of some 460 Km. The small island or Juan de Nova (uninhabited except for a military garrison) lies to the south of this other potential sailing route.

The toolkit found in the key site of Lakaton'i Anja includes many microliths, as well as some larger tools, made of chert and obsidian. This last must have been brought from far away, as there are no sources of the volcanic glass in Northern Madagascar.

An important point is that these new dates show that human inhabitation did not kill the Malagasy megafauna right away but that instead humans and giant animals shared the environment without immediate catastrophic consequences, which would only happen in the last two millennia.


  1. Totally unexpected. A fascinating find.

  2. Totally expected. Many anti-African people undermind the Bantu-Expansion

    1. This newly discovered African arrival to Madagascar would be clearly pre-Bantu as well.

  3. To expand on why this is unexpected:
    1. There is no prior archaeological evidence of pre-Austronesian human habitation of Madagascar.
    2. As Maju notes, the first two thousand years of human habitation apparently did not result in megafauna extinction, despite the fact that this is a substantially post-Upper Paleolithic and Last Glacial Maximum population and known modern human arrivals that late historically in many other places have resulted in megafauna extinctions.
    3. African admixture in modern Madagascar shows affinity to East African Bantus, rather than to pre-Bantu populations of the eastern coast of Africa (e.g. the San, the non-Bantu components in East Africa, the non-Bantu autosomal genetic component of Mozambique).

    1. What I could find online was that there were, before these findings, some indications of older inhabitation but were inconclusive and at most pushed the date of arrival to c. 300 or 400 BCE, which was surely pre-Austronesian but not 100% certain.

      Personally I think that megafauna extinctions, while real and in many cases surely partly caused by human action (especially in islands), are a bit hyped.

      " African admixture in modern Madagascar shows affinity to East African Bantus, rather than to pre-Bantu populations of the eastern coast of Africa (e.g. the San, the non-Bantu components in East Africa, the non-Bantu autosomal genetic component of Mozambique)".

      Are you sure? I don't know of any such detailed analysis, maybe you can point me to the relevant study or a synthesis of it. Even for Mozambique I find myself often begging for better analysis and Madagascar has been rather neglected in African genetic studies (with exceptions very limited in scope) AFAIK.

  4. This comment has been removed by a blog administrator.

    1. Megafauna extinctions in Australia and New Guinea, in the Americas, in Siberia, and in Europe all predated the Neolithic revolution.

    2. DDeden is a spammer, please don't feed the troll.

  5. The Mozambique study is Sikora, et al., "A genomic analysis identifies a novel component in the genetic structure of sub-Saharan African populations" European Journal of Human Genetics (2011) 19, 84–88; doi:10.1038/ejhg.2010.141.

    A key passage in that paper notes: "The southeastern Bantu from Mozambique are remarkably differentiated from the western Niger-Congo speaking populations, such as the Mandenka and the Yoruba, and also differentiated from geographically closer Eastern Bantu samples, such as Luhya.

    These results suggest that the Bantu expansion of languages, which started ~5000 years ago at the present day border region of Nigeria and Cameroon . . . was not a demographic homogeneous migration with population replacement in the southernmost part of the continent, but acquired more divergence, likely because of the integration of pre-Bantu people. . . . the singularity of the southeastern population of Mozambique (poorly related to present Khoisan) could be attributed to a complete assimilation of ancient genetically differentiated populations (presently unknown) by Bantu speakers in southeastern Africa, without leaving any pre-Bantu population in the area to compare with."

    Some of the pertinent Madagascar population genetics papers are: Regueiro, et al., "Austronesian genetic signature in East African Madagascar and Polynesia" Journal of Human Genetics (2008) 53, 106–120; doi:10.1007/s10038-007-0224-4


    Poetsch, "Determination of population origin: a comparison of autosomal SNPs, Y-chromosomal and mtDNA haplogroups using a Malagasy population as example", European Journal of Human Genetics (24 April 2013) doi:10.1038/ejhg.2013.51


    Hurles, et al. "The dual origin of the malagasy in island southeast Asia and East Africa: evidence from maternal and paternal lineages.", Am J Hum Genet. 2005;76;894-901. PMID: 15793703


    Tofanelli, et al., "On the origins and admixture of Malagasy: new evidence from high-resolution analyses of paternal and maternal lineages." Mol. Biol. Evol. 26, 2109–2124 (2009) (doi:10.1093/molbev/msp120)
    ("The pattern of diffusion of uniparental lineages was compatible with at least two events: a primary admixture of proto-Malay people with Bantu speakers bearing a western-like pool of haplotypes, followed by a secondary flow of Southeastern Bantu speakers unpaired for gender (mainly male driven) and geography (mainly coastal); all African loan words in the Malagasy languages are Bantu in origin; the remainder of its words are Asian with 90% from a particular language spoken in Borneo."; African mtDNA haplogroups werw from L* 5% of tlt, L2b1b 2% of tlt, L3b1 28% of tlt, L3e1a 4% of tlt (remainder Asian in origin)l African Y-DNA haplogroups were E1b1a, B2*, E2b (E3a 36% of tlt, lesser amounts of other hgs).

    Off topic (focuses on Austonesian mtDNA) but worth a mention for completeness is: Cox, et al., "A small cohort of Island Southeast Asian women founded Madagascar", Proc. R. Soc. B. (21 March 2012) doi: 10.1098/rspb.2012.0012 (the mtDNA gene pool of Madagascar is traceable to about 30 women).

    1. Razib Khan also personally did an independent study of large numbers of autosomal DNA loci from two Malagasy individuals which he compared to various publicly available reference samples:

      This result found an absence of or only trace percentages of the African autosomal components that are common in the Sandawe (about 1-2%), San and Mbuti Pygmies (about 1-2%), Biaka Pygmies (about 1-2%), or Maasi and Ethiopian Jews (3-4%). There were no SW Asian components (e.g. Yemen Jews) or European components. The balance was a mix of clearly Asian components.

      The proportions of different African autosomal components in Razib's samples were roughly proportional to the Luhya (except that the Maasi component in the Luhya was a bit less common, proportionately, in the two Malagasy individuals), which is an East African Bantu population. In other words, the African component in the Malagasy autosomal DNA profile is more purely Nigeria-Cameroon Bantu than the intermediate Luhya (suggesting that East African Bantus have been seen moderate introgression from genetically Maasi-like populations after joining the Austronesian peoples en route to Madagascar ca. 1000-2500 years ago).

      If the source were Mozambique (and especially if it were a pre-Bantu source) you would expect a substantial component of a unique Paleoafrican component that was distinct from the Bantu component that makes up about 95% of Yoruba and Mandenka individuals. Since SE African Bantu have a very different autosomal profile than the Yoruba and Mandenka and Luhya per Sikora (2011).

    2. Interesting, thanks. It is still only two individuals of two of the many ethnicities of Madagascar. The Merinas (whose historical country is in the central plateau, around Antananrivo) are well-known for being the most Asian ones of all Malagasy, while the Betsileo are their neighbors by the South.

      It'd be much more interesting to study Western, Northern and Southern coastal peoples, like the Sakalava (West), Antanarakana (North), Mahafaly or Antandroy (South). It'd be also interesting to compare with Mozambicans directly, because Mozambicans seem to have a major distinct component to other Bantus, which may be a remnant from before the Bantu expansion or something like that (affinities in some Pygmy groups, especially the Twa of Rwanda and Burundi). See:



      (The latter I never got access to, so if any reader does, feel free to send me a copy, because it looks most interesting - thanks in advance).

      "If the source were Mozambique (and especially if it were a pre-Bantu source) you would expect a substantial component of a unique Paleoafrican component that was distinct from the Bantu component that makes up about 95% of Yoruba and Mandenka individuals".

      The problem is that if you don't have a good comparison you get nothing, especially with the huge African diversity. It's like expecting Hadza component in the Maasai... it does not happen, at least not significantly, yet both are East African aboriginal populations (but from further North than we would require). If at least there'd be a Sandawe sample (or Twa, or Mozambicans), but nope. And the two Malagasy individuals weight too little to bring forward whatever peculiarities of Madagascar in such a large global sample.

      Still, in my understanding the African component (for whatever is worth that comparison) is much closer to the Luhya complexity than to the West African simplicity.

    3. Two studies on Malagasy haploid affinities:

      http://www.sciencedirect.com/science/article/pii/S0002929707607368 (Hurles 2005)

      http://mbe.oxfordjournals.org/content/26/9/2109.full (Tofanelli 2009)

      The first one only uses a single generic Malagasy sample of mostly Merina and Betsileo individuals, finding affinities with SE Asia and Africa but it's not too precise.

      The second one is maybe more interesting because it explores the differences between the Merina (highlanders) and three populations from the SE (collectively known as Côtiers = "coast people" in French). They both show similar mtDNA affinity with SE Asia (~60%) and Africa (~40%) BUT the Côtiers have much more Y-DNA ancestry from Africa (almost 75%) than the Merina (50%). There are other differences as well:

      SE Asian mtDNA: Côtiers have more B4a1a1 (Polynesian motif) while the Merina have more E1a and M(xM7).

      African mtDNA: Côtiers have a very diverse African (L(xM,N)) ancestry, while the Merina are dominated by L3b1 in this aspect.

      African Y-DNA: all the difference of Côtiers vs Merina is owed to "E3a" (named as E1b1a in other sections of the same paper). Fig. 3 explores the affinities of Malagasy E1b1a, finding that there are five clusters: three of East African affinity and two of W/E African affinity with the occasional West affinity haplotype (so these could well be described as "Bantu"). The East African affinities are exclusive of the Côtiers and strongest among the Antandroy (South vs SE, although they are neighbors). This would be consistent with an expectation of more of that East African specificity in the (so far unresearched) West coast of the island.

      Other Y-DNA: most (Eur-)Asian paternal ancestry differences are negligible, although maybe the differences for haplogroup J could be worth a second look (?).

      So, in essence, all (studied) Malagasy have an African male ancestry bias but it is much stronger (and much more diverse) among the Côtiers. Some of these (notably the Antandroy) seem to have much stronger East African specific (pre-Bantu) paternal ancestry, leaving very open the possibility that this also happen in other coastal areas of the West and North.

    4. A very intriguing people are the Mikea, a semi-forager people of SW Madagascar. Among them and also among the Vezo (who inhabit most of the Western coast, being specialist fishers) a new mtDNA M lineage was discovered years ago: http://leherensuge.blogspot.com/2009/12/novel-mtdna-m-sublineage-found-in.html

      Both could be loosely included among the Sakalava, which is a catch-all term for various peoples united under a single kingdom once upon a time.

      The Mikea consider themselves descendants of historical foragers known as Vazimba or Beosi, whose language(s), now extinct, was documented in 1920, showing resemblance to no other. It has been argued that the Malagasy dialect of the Mikea has such pre-Austronesian substratum: http://www.rogerblench.info/Language/Isolates/Vazimba%20vocabulary.pdf

    5. Oops, I was replying to your second comment. I only now realized that it was preceded by another much more extensive comment with many links, including several of the ones I mentioned as reply. My apologies.

      Still, I think that what I said does stand.

      I was not aware that the Sikora paper on Mozambique was freely available. As I expected Mozambicans are clearly not typical Bantu/West African by ancestry but something else. Even more interesting is that the Luhya, when compared to Mozambicans as well, show much lesser affinity to West Africa and appear as an East African population instead. At K=5 they also develop their own distinct component.

      As I said above the Mozambican component is very akin to Mountain Pygmies (Twa), who seem different from both West and East Jungle Pygmies (and therefore the Sikora comparison is not very useful).

  6. Im really glad you are reporting on this important find. This effectively doubles the timeline of human occupation of the island. I think that it is entirely possible that the owners and users of this paleolithic technology were a different people from the waves of (or single landing of) Austronesian settlers beginning around 500 AD. It will be really interesting to see the history of this people emerge. There is a strong oral history of the Vazimba throughout the island and too little documented of it, with the focus being on the Mikea in the south. The Vazimba were said to be a small forest dwelling hunter-gather people that lived here before the Malagasy. You don't say... Id be interested in any articles people could link to on the history of Vazimba as told through story. http://marshinmadagascar.blogspot.com/2013/07/hard-evidence-of-2000bce-settlement-of.html

    1. Indeed. It is a really fascinating story and, as you say in your blog, challenges many ideas.

      You emphasize the issue of environmental destruction or lack of it. I have got my face rubbed more than once with that but the reality is that some of the areas where humans have lived for longer (Africa, Tropical Asia) the megafauna or most of it survived, while in other areas (Europe, America), it survived until the big climate changes of the end of the Ice Age (and some species at least did much further in time). So it's not just human presence but a more complex scenario.

      Islands are almost invariably more fragile ecosystems that the big continents, just because of their isolation. And Madagascar is archetypal to this kind of challenge. But it's not just humans who almost automatically destroy everything: it's at the very least techno-cultural development, and within it certain kinds are worse than others.

      But there is more to it: the myth of Africans not being into navigation collapses with this finding. Hunter-gatherers or not, these people obviously had mariner skills that surpass those of the first settlers of Australasia and are comparable to those of Epipaleolithic (and some times older) Europeans, who began settling some of the Mediterranean Islands (also rather recent findings).

      "There is a strong oral history of the Vazimba throughout the island and too little documented of it, with the focus being on the Mikea in the south. The Vazimba were said to be a small forest dwelling hunter-gather people that lived here before the Malagasy".

      As I've said in the comments above, genetic research of the island has been relatively limited and the West almost totally ignored so far. A proper sampling strategy, not just in the island but also in neighboring countries: Mozambique, Tanzania and Comoros particularly, should be most interesting towards better understanding the origins of Malagasy in their diversity. If the Vazimba were hunter-gatherers, it is quite probable that they only left some remnants but there must be such a detectable substrate to be found with due research.


Please, be reasonably respectful when making comments. I do not tolerate in particular sexism, racism nor homophobia. Personal attacks, manipulation and trolling are also very much unwelcome here.The author reserves the right to delete any abusive comment.

Preliminary comment moderation is... ON (your comment may take some time, maybe days or weeks to appear).