Bantu Expansion Shows That Habitat Alters the Route and Pace of Human Dispersals

Bantu expansion shows that habitat alters the route and pace of human dispersals

Rebecca Grollemunda,1, Simon Branforda, Koen Bostoenb, Andrew Meadea, Chris Vendittia, and Mark Pagela,c,1

aEvolutionary Biology Group, School of Biological Sciences, University of Reading, Reading RG6 6BX, England; bKongoKing Research Group, Department of Languages and Cultures, Ghent University, 9000 Ghent, Belgium; and cThe Santa Fe Institute, Santa Fe, NM 87501

Edited by Peter S. Bellwood, Australian National University, Canberra, ACT, Australia, and accepted by the Editorial Board August 10, 2015 (receivedfor review February 25, 2015) Unlike most other biological species, humans can use cultural inno- at ∼2,500 B.P. affected amongst others the western part of the vations to occupy a range of environments, raising the intriguing Congo Basin, creating patches of more or less open forests and question of whether human migrations move relatively indepen- wooded or grassland savannahs (14, 15). These areas eventually dently of habitat or show preferences for familiar ones. The Bantu merged into a corridor known as the “Sangha River Interval” that expansion that swept out of West Central Africa beginning ∼5,000 y repeatedly facilitated the north–southspreadofcertaintypical ago is one of the most influential cultural events of its kind, even- savannah plant and animal species (17, 20–22). tually spreading over a vast geographical area a new way of life in The Sangha River Interval may also have been a crucial pas- which farming played an increasingly important role. We use a new sageway for the initial north–south migration of Bantu speech dated phylogeny of ∼400 Bantu languages to show that migrating communities across the Equator. The archaeological evidence is Bantu-speaking populations did not expand from their ancestral not yet detailed enough on its own to test this idea (17). How- homeland in a “random walk” but, rather, followed emerging sa- ever, the geographical expansion of the Bantu linguistic family, vannah corridors, with rainforest habitats repeatedly imposing coupled with phylogenetic trees that make use of archaeological temporal barriers to movement. When populations did move evidence, provides an opportunity to reconstruct how and when from savannah into rainforest, rates of migration were slowed, de- this cultural expansion moved through the varying habitats of laying the occupation of the rainforest by on average 300 y, com- West Central Africa. pared with similar migratory movements exclusively within savannah Here we use a new time-calibrated phylogenetic tree describing or within rainforest by established rainforest populations. Despite the patterns of descent of ∼400 Bantu languages to study the route unmatched abilities to produce innovations culturally, unfamiliar hab- and pace of Bantu speakers as they migrated from their ancestral itats significantly alter the route and pace of human dispersals. homelands. Our data include a dense sampling of languages that descend from the early phases of the Bantu expansion, along with human dispersal | phylogeography | phylogenetics | languages | Bantu five now-extinct northern Bantu languages and several Bantu languages spoken in the northeastern Democratic Republic of the ost biological species are confined to areas of the world for Congo (DRC). In combination with information on present-day Mwhich their genes have adapted them, but humans, relying geographical positions of the Bantu languages, the phylogenetic on cultural innovations passed down for generations, have been able to inhabit nearly every environment on Earth (1). Even so, Significance from our earliest migrations as a species, there is reason to believe that modern humans, despite all of their cultural evolutionary Humans are uniquely capable of using cultural innovations to potential, might have preferred to follow habitats that did not occupy a range of environments, raising the intriguing question require them to master new environments. The so-called “beach- of whether historical human migrations have followed familiar comber” or “coastal routes” hypothesis proposes that the first habitats or moved relatively independently of them. Beginning migrations out of Africa might have followed a coastal route via ∼5,000 y ago, savannah-dwelling populations of Bantu-speaking India to the Far East and eventually to Australia (2). Much more peoples swept out of West Central Africa, eventually occupying recently, there was a suggestion that during the occupation of the ∼ a vast geographical area. We show that this expansion avoided Pacific by Austronesian people 3,500 y ago (3), there were several unfamiliar rainforest habitats by following savannah corridors periods during which the migration paused while people acquired that emerged from the Congo rainforest, probably from climate the sailing technology to attempt further voyages (4). This tech- change. When Bantu speakers did move into the rainforest, nology, in the form of boat designs, might also have been under migration rates were delayed by on average 300 y compared strong natural selection (5), showing that cultural innovations are with similar movements on the savannah. Despite unmatched – not just a matter of whimsy. East west migrations might in general abilities to produce innovations culturally, unfamiliar habitats be more common than north–south movements because the former significantly alter the route and pace of human dispersals. are less likely to encounter variation in climate and habitat (6). Bantu migrations swept out of West Central Africa beginning Author contributions: R.G., S.B., K.B., A.M., C.V., and M.P. designed research; R.G., S.B., K.B., ∼5,000 y ago (B.P.) and eventually moved all the way down to the A.M., C.V., and M.P. performed research; R.G., S.B., K.B., A.M., C.V., and M.P. contributed southern tip of the African continent. It was one of the most in- new reagents/analytic tools; R.G., S.B., K.B., A.M., C.V., and M.P. analyzed data; and R.G., S.B., K.B., A.M., C.V., and M.P. wrote the paper. fluential cultural events of its kind, spreading over a vast geo- The authors declare no conflict of interest. graphical area a new, more sedentary way of life that was — This article is a PNAS Direct Submission. P.S.B. is a guest editor invited by the Editorial fundamentally different from that of indigenous forest foragers Board. ancestral Bantu speakers had mixed-subsistence economies, in Freely available online through the PNAS open access option. which farming gradually gained in importance (7–9). Data deposition: The Bantu language data and the multistate encoding of the language Two major events in the recent paleoenvironmental history of data are available at www.evolution.reading.ac.uk/DataSets.html. Central Africa might have influenced the route of the Bantu ex- 1To whom correspondence may be addressed. Email: [email protected] or r.b.grollemund@ pansion (10–18). The first was a contraction at ∼4,000 B.P. of the reading.ac.uk. Congo rainforest at its periphery, for instance along the coasts of This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. South Cameroon, Gabon, and Congo (11, 16, 19). A second event 1073/pnas.1503793112/-/DCSupplemental.

13296–13301 | PNAS | October 27, 2015 | vol. 112 | no. 43 www.pnas.org/cgi/doi/10.1073/pnas.1503793112 Downloaded by guest on September 24, 2021 tree allows us to infer ancestral migration routes and then test The ∼4,800 B.P. date for node 1 can be compared with the among proposed scenarios for how Bantu speakers moved through results from two recent genetic studies on the assumption that the the savannah and rainforest habitats of Central, Eastern, and in-group Bantu node coincides with the beginning of the Bantu Southern Africa. expansion. Gignoux et al. (24) report a population expansion of “sub-Saharan” people at ∼4,600 y ago, and Li et al. (25) find ev- Results idence for a Bantu population expansion at ∼5,600 y ago. Dated Phylogenetic Tree of the Bantu. We derived a Bayesian posterior sample of n = 100 phylogenetic trees from linguistic Historical Migration Route. We used information on the latitudinal data on 424 Bantu and related languages (Materials and Methods and longitudinal positions of the languages to reconstruct the and SI Materials and Methods). The consensus phylogeny (Fig. probable ancestral geographical locations of each of the internal 1andFig. S1) depicts a progressive “backbone” or pectinate nodes of the trees in the posterior sample (Materials and Methods radiation from a common ancestor with the out-group Grass- and SI Materials and Methods). We then used these reconstruct- fields languages. This radiating tree occurs in 100% of the trees ions to record the routes of dispersal of Bantu speakers from their in the posterior sample (SI Materials and Methods). The tree’s homeland, and we linked the reconstructed geographical position broad outlines are similar to the tree that Currie et al. (23) at each node to its inferred time, as recorded on the tree, and to report, but where those authors find paraphyletic groups for information from palynological and paleoenvironmental studies the central-western and west-western Bantu, we reconstruct (13–15, 26) on the likely habitats at different times in the past. monophyletic groups. The reconstructions (Fig. 2 A and B) locate the ancestral On the basis of four calibration ranges supported by archae- homeland of the common ancestors to the Bantu and outgroup ological studies (Materials and Methods and SI Materials and Grassfields speakers (node 0, Fig. 1) in the savannah habitat of Methods), the root of the tree estimates a common ancestor with Northwestern Cameroon. The pectinate nature of the tree means the outgroup Grassfields speakers at ∼6,900 B.P. (node 0, Fig. 1; that the Bantu language groups that descended from the Bantu age = 6,929.7 ± 418.6 B.P.), a date considerably older than the common ancestor (node 1, Fig. 1) would themselves become the 5,000-B.P. younger limit suggested by our calibration range. The ancestors to the major radiation of the Bantu that eventually oc- tree then dates the remaining Bantu in-group (node b) to ∼4,800 cupied large parts of Central, Eastern, and Southern Africa. A B.P. (4,846.5 ± 138.1), a time that is near to the older end of The principal dispersal route (Fig. 2 ) first moves in a south- – dates suggested by archaeology (node b prior range = 4,000 B.P. easterly direction (approximately nodes 1 8), before traversing in to 5,000 B.P.; SI Materials and Methods). a predominantly easterly direction along the southern boundary of the Congo rainforest [this is in contrast to Currie et al. (23), whose reconstructed route moves in alternating south and east steps, crossing the Congo rainforest]. We find no evidence for the sug- a Jarawan NORTH-WESTERN gestion (27, 28) that the main migration followed a coastal route A31-40-50-60 Cameroon c 0 A10-20-30 (Fig. 2). A few early groups did explore coastal routes (Fig. 1), but

A40-60-70 these groups moved in from the east after having branched off the b 1

B20-A80-90 main backbone migration, rather than being ancestral to it. EVOLUTION NORTH-WESTERN B20 At least three principal southern migrations branched off from 2 B10-30 Gabon C50-D10 the backbone as it moved east along the southern boundary of 3 D20-30-C40 the rainforest (Fig. 2A), the last of which were the ancestors to CENTRAL-WESTERN 4 C10-20-30-40 modern-day South African Bantu speakers. This migration route C60-70-80 is consistent with proposals (29–35) that the ancestors of the B80 5 modern-day Eastern Bantu groups diverged from the Western B50-60-70 WEST-WESTERN Bantu ∼2,000 y ago in the Congo region. H10-30-42-B40 – 6 However, our results reject the suggestion (36 38) that the L10-20-H21a-H30-40 SOUTH-WESTERN Eastern Bantu speakers in the Great Lakes region of East Africa 7 L50-60-K10-30-R10-20-30 8 L20-30-40 trace their ancestry back to Bantu-speaking peoples who had D10-20-30-43-55 9 M40-50-60 migrated from the northern Congo. Instead, we find that the JD50-60

d JE10-20-30-40 Eastern Bantu are the descendants of people who moved north into the Great Lakes region from the main backbone (brown F10-20-30 EASTERN M10-20-30 lines, Fig. 2A). This result emerges despite the fact that our tree

G50-60-N10-P10-20 includes five now-extinct Bantu languages, along with several

P30-K21-S10-20-30-40-50-N20-30-40 contemporary Bantu languages, all spoken in the northeastern G10-20-30 DRC and that have been proposed (39) to have shared a more G40-E70 E74a-G20-G39 recent common ancestor with Eastern Bantu. Our findings are

E46-50-60 also consistent with genetic studies (40) that have found a pos- itive correlation between genetic and linguistic distances, which 6000 5000 4000 3000 2000 1000 0 suggests that a northern migration route was less probable. Fig. 1. Consensus time tree of n = 424 Bantu languages, derived from n = 100 Our principal interest is in whether the early Bantu migration trees drawn from the Bayesian posterior distribution. Triangles are pro- (nodes 1–8 in Fig. 2A) took advantage of changes to the climate portional to the number of languages in the group, and the labels are the and habitat in the western Congo basin that created north–south codes used by Guthrie (65). Phylogenetic methods and full tree are reported in “corridors” through the core of the Central African rainforest SI Materials and Methods. The four calibrations used are identified by red (dashed curve Fig. 2A). Before ∼4,000 B.P. (11, 16, 19), nearly letters (a, 5,000 B.P. or older; b, 4,000–5,000 B.P.; c, 3,000–3,500 B.P.; and d, the entire light- and darker-shaded regions of Fig. 2A were cov- 2,500 B.P.; SI Materials and Methods). (Inset) Map of Africa with colored dots ered by rainforest (SI Materials and Methods and Fig. S2A). Then, to represent the current location of the languages. Note: The age of the root on the consensus tree differs from the average root in the posterior sample palynological and geological data (11, 16, 19, 41) indicated that, by (text). This is because the ages of nodes on the consensus tree were recon- at least 4,000 y ago, climate changes had created encroaching structed by fitting the phylogenetic model to the fixed consensus tree topol- savannah habitats in the periphery of the rainforest (white and ogy. All statistics reported in the text are based on the posterior sample, not light green shading, Fig. 2A)—for instance, along the coasts of the consensus tree. Gabon and Congo. It is only toward 2,500 B.P. that climate change

Grollemund et al. PNAS | October 27, 2015 | vol. 112 | no. 43 | 13297 Downloaded by guest on September 24, 2021 AB 0 ? 12 ?

34 5 6 7 9 8

Fig. 2. Ancestral migration route reconstruction. (A) Ancestral migration route reconstructed on consensus tree by using geographical locations of contemporary languages and connecting ancestral locations by straight lines (true route will differ). Numbered positions correspond to nodes on the consensus tree (Fig. 1). Curved dashed line indicates suggested migration route through savannah corridors (B). Lighter green shading corresponds to the delimitation of the rainforest at 5,000 B.P.; the darker green corresponds to the delimitation of the rainforest at 2,500 B.P. (text and SI Materials and Methods). (B)Map showing the ancestral locations of the backbone nodes (Fig. 1) for the 100 trees in the Bayesian posterior sample; curved arrow is suggested route for the early migration based on a small number of reconstructed points that fall in rainforest. (C) Same as B but showing the ancestral locations of random migration routes for nodes 0–8 (text and SI Materials and Methods).

also led to the development of savannah vegetation in central negligible level (∼47% of routes; Materials and Methods and SI parts of the Congo rainforest, yielding corridors such as the San- Materials and Methods). gha River Interval in the western part of the Congo Basin (SI An intriguing alternative to the proposal that the Bantu fol- Materials and Methods and Fig. S2B), which connected northern lowed emerging savannah habitats is that they created their mi- and southern savannahs (14, 15, 17). gration route by deforesting the Sangha River Interval region (42). To test the savannah-corridor hypothesis (that the backbone However, we think this scenario is unlikely to have played a major Bantu migration followed savannah rather than rainforest habi- role in determining the Bantu’s route. The thinning of the rain- tats), we reconstructed the ancestral geographical positions of forest occurred simultaneously over much of the region from nodes 0–8 (Fig. 1) for each of the trees in our posterior sample. Cameroon to the Congo (42), and it grew from southern as well as Then, using dates from the trees along with the paleoclimatic northern areas (Fig. 2A). This thinning has been linked to climatic data (SI Materials and Methods and Fig. S2 A–C), we asked changes, but not to human deforestation (43, 44), suggesting that whether at the time the Bantu speakers are inferred to have been if Bantu populations contributed to thinning, it was to a process at those positions, the habitat had changed from rainforest to that was already underway. either savannah or other nonrainforest habitat. The last of these There is also no evidence to suggest that the predominantly nodes (node 8) roughly corresponds to the point at which the north–south movement of the Bantu through the savannah cor- southeasterly Bantu migration reaches the southern boundary of ridor followed or was aided by rivers. Archaeological evidence the rainforest, before turning east. from the Inner Congo Basin (45, 46) suggests movement of Bantu We find that in all 100 trees in the posterior sample, the communities along rivers mostly in a west to east direction and backbone moves in a southeasterly direction toward the southern involving groups that are not part of the backbone or main boundary of the rainforest (Fig. 2B). A small number of ancestral migration lineage. positions are reconstructed in a “bulb” of rainforest habitat in the northwest, but the majority are not, suggesting that the main mi- Migration Rates Within and Between Savannah and Rainforest Habitats. gration moved around it (curved arrow). Thus, in n = 96 (96%) of The tree, along with the dates and palynological and paleoenvir- the trees, the reconstructed positions of at least 7 of the 9 ancestral onmental information, can be used to identify “habitat transitions,” nodes miss the rainforest entirely (routes plotted in Fig. 2B): all defined as instances in which the geographical position and date 9 nodes miss the forest in n = 73 of the trees, and at least 8 miss reconstructed at the beginning of a branch on the tree implies a the forest in n = 87 trees, giving an average of 8.53 ± 0.96 of 9 of different habitat from the one implied by the geographical position the backbone nodes falling in nonrainforest habitat. and date at the end of the branch. It is unlikely that the reconstructed migration route and fit to Across trees, we found an average of 52.7 ± 4.4 independent the habitat could have arisen by chance: When we simulate mi- habitat transitions, with 35.8 ± 3.4 corresponding to transitions grations as random walks from the ancestral homeland, and by from savannah into forest and 16.9 ± 2.5 from forest back to using conservative criteria that favor the random-walks hypothesis savannah: We say “back” to savannah because most rainforest- (simulation details in SI Materials and Methods), we find that, at dwelling Bantu speech communities have an ancestral history of most, 6.3–9.7% of the random-migration routes follow the sa- residing in savannah. The consensus tree records 48 transitions vannah corridor as closely as the real data (corresponding to 7, 8, between habitats, 31 corresponding to transitions from savannah or 9 nodes outside the forest; Fig. 2C). Only when we restrict the to rainforest, and 17 from rainforest back to savannah (Fig. 3). simulations to move exclusively in a southeasterly direction do our The remaining branches record movement within the same habitat, simulated routes coincide with the savannah corridor beyond a either forest or savannah.

13298 | www.pnas.org/cgi/doi/10.1073/pnas.1503793112 Grollemund et al. Downloaded by guest on September 24, 2021 in a southeasterly direction, taking advantage of a savannah corridor that began to appear by ∼4,000 y ago. This route avoided rainforest habitats and spawned numerous migratory branches that led to the occupation of nearly all of southern Africa, along with several independent movements north into the Great Lakes region of East Africa. When savannah-dwelling Bantu-speaking groups did move into the rainforest, their rate of migration was significantly slowed. On its own, this result might not be surprising—the rainforest is covered with dense vegetation that might have made subsistence (and especially farming) more difficult, and rainforest habitats might harbor more predators and organisms causing infectious disease. What is surprising, however, and relevant to the question of human cultural innovation, is the extent to which the rainforest slowed human movement. Vansina (47) has written that “[Bantu] Farmers took some 2000 y to settle the rainforests of equatorial Africa, and then, about another half millennium to absorb new technologies and to become finely attuned to all of the potential of their habitats.” Our phylogenetic reconstructions, showing that Savannah -> Rainforest transitions into the rainforest were delayed by ∼300 y compared Rainforest -> Savannah with movements of a similar distance within savannah habitats, are Fig. 3. Consensus time tree with panels that enlarge the clades that have in good agreement with Vansina’s observations and correspond to savannah to rainforest (n = 31 independent transitions) and rainforest to a 50% reduction in the pace of human expansion. savannah (n = 17 independent transitions). Numbers of each kind of tran- Could transitions into the rainforest really delay movements sition vary in the posterior sample (text). Both kinds of transition are widely by hundreds of years? Our results curiously seem to fit with distributed among the clades near to the rainforest, and S->F transitions are modern studies that suggest that human innovation has less to do > always ancestral to F- S transitions. Some lineages have experienced three with thinking hard until the right solution comes to mind (the transitions in their history. lightbulb switching on in our minds), than with the slow accu- mulation of knowledge and technology principally resulting from “ ” On average, Bantu speaking groups that moved into the rain- trial and error. Thus, Basalla (48) and Arthur (49) both em- ∼ phasize the cumulative nature of human innovations, downplaying forest (F) were significantly delayed, taking on average 300 y “ ” ’ longer than comparable transitions within savannah (S) habitats the role of genius innovators. For instance, Henry Ford sfa- (Fig. 4). This significant delay is observed separately in at least 90% mous assembly line production drew on earlier experiments with streamlining assembly lines, and Watt’s steam engine was less of of the trees in the posterior sample and is not an artifact of S->F an “out of thin air” invention than a development of Newcombe’s EVOLUTION transitions covering a greater distance: Our analyses control for the “ > earlier engine. Thomas Edison is often credited with inventing distance moved, implying that S- F transitions proceed at an ab- ” > the light bulb, but records show that his patent was for a better solutely slower pace. We think it is unlikely that the slower S- F filament to a lightbulb, and his notebooks reveal that he tried transitions could arise from a higher extinction rate of groups that thousands of filament materials before alighting by chance on his attempted this transition: Even if there were higher extinction rates, favored material. The typically low population densities of sub- because the analyses control for the distance moved, the finding of sistence peoples such as early Bantu speakers would only have a slower rate of movement of successful transitions stands. exaggerated the difficulties of accumulating new technologies (50). By comparison, transitions from the rainforest back to sa- Our approach shows that evidence bearing on subtle questions vannah take no longer on average than movements within either of human history can be investigated by using phylogenies de- rainforest or savannah (not significant in any tree; Fig. 4). This rived from languages, combined with relevant information on finding might suggest that the savannah is an easier habitat to contemporary cultures and appropriate statistical modeling. In- occupy or, more interestingly, that the rainforest-dwelling Bantu deed, there is reason to believe that language phylogenies might cultures in our tree tend to descend from ancestrally savannah- even be preferable to gene-based trees in this regard (51). dwelling cultures and retained some cultural knowledge of how Languages typically evolve at a higher rate than genes, meaning to exploit the savannah environment. that they can resolve shorter time scales, but languages might have an even more fundamental role. Languages track the in- Discussion heritance of culture, and it is this inheritance that is normally Together, our results show that the Bantu expansion was char- pertinent to questions of human cultural evolution. Genes, by acterized by a measureable preference for following familiar sa- comparison, can readily move among cultures, without neces- vannah habitats as it moved from present-day northwest Cameroon sarily taking their cultures with them.

hannavaS 300 400 500 600 700 800 hannavaS 300 400 500 600 700 800 tseroF 300 400 500 600 700 800 tseroF

Fig. 4. Posterior distribution of times taken for four different habitat transitions, controlling for distance moved. Savannah to forest transitions are significantly slowed (Tukey honest significant difference test; P < 0.05) compared with transitions within savannah in 90 of 100 trees in the posterior sample. Rainforest to savannah transitions take no longer on average than movements within either rainforest or savannah (not significant in any tree). Mean in years ± SD: S->S = 368.6 ± 13.9; S->F = 662.8 ± 78.7; F->S = 446.0 ± 64.7; F->F = 420.6 ± 24.2. All significance tests were performed on log-transformed data to normalize variances.

Grollemund et al. PNAS | October 27, 2015 | vol. 112 | no. 43 | 13299 Downloaded by guest on September 24, 2021 Materials and Methods Ancestral Reconstructions. We inferred ancestral latitude and longitude for Linguistic Data. We collected lexical data from published sources and from each node of our tree using a Brownian motion model applied to the con- fieldwork for 409 Bantu and 15 Bantoid languages. Our wordlist is a modified temporary data that allowed for rates of geographical movement to vary version of the Atlas Linguistique du GABon (52). This list comprises 159 words throughout the tree, following methodology we have reported elsewhere from which we have sampled 100 words that are the best documented for (63) and as implemented in our BayesTraits software. the languages we studied (Materials and Methods and SI Materials and Methods). We then classified the words into cognate sets and built a binary- Simulated Migrations of Savannah Corridor Route. We generated random coded dataset (each column identifies a unique cognate class), yielding 3,859 dispersal routes from the Bantu homeland for the nine nodes (nodes 0–8of cognate classes for the 424 languages. Fig. 1) corresponding to the southeasterly movement through the savannah corridor and out into the savannah south of the Congo rainforest. We held Phylogenetic-Statistical Methodology. We inferred a time-dated phylogeny constant the consensus phylogenetic tree and the timings at its nodes, so as from the lexical dataset using a variable-rates molecular clock model that allows not to introduce a large and unknown additional source of possible geo- the rate of evolution to vary among branches of the tree. The variable-rate graphical movements. Simulated routes were allowed to go to places that clock is modeled by applying a scalar multiplier to each branch of the tree that Bantu have actually inhabited historically or at present. Moves into the sea alters the rates by some fixed amount (53). We assume these scalars are drawn from a log-normal prior distribution with μ = 1 and unknown σ2 that we es- or other bodies of water were prohibited, and a newly simulated position timate from the data. Node ages were estimated by using a Yule process (54). was not allowed to occupy a space already occupied (defined as within Trees were inferred using Markov chain Monte Carlo methods (55) that 10 km of any previously simulated point, unless the distance to be traveled implemented Tuffley and Steel’s covarion model (56), which allows the rate was less than this). of evolution to jump between an “on” and an “off” state throughout the These constraints narrowed the space of possible migration routes, making tree (model testing and selection is detailed in SI Materials and Methods and it more likely the simulated routes would coincide with the savannah cor- Table S1). The covarion model is well suited to binary-coded cognate data, ridor. We then simulated two dispersal scenarios. In the first, the distances owing to the fact that each cognate class ideally arises just once on the tree. The moved along the backbone on the tree followed those actually observed variable rates and covarion models were implemented in our BayesPhylogenies along the same branches but in a random order; in the second, these distances software (57). The ladderised or pectinate phylogeny of Fig. 1 is robust to were drawn from a random distribution but normalized to have the same subsampling of the n = 100 words (SI Materials and Methods). total distance moved as observed in the real data. The first yielded 9.7% of Chains were run for 3 × 108 iterations, with a sampling period of 10,000 routes with seven or more nodes falling in the savannah corridor, and the iterations. We used the Tiv and the Grassfields languages as out-groups to root the tree. second returned 6.3% using the same criterion. Only when we constrain the simulations to move exclusively in a southeasterly direction do our simulated —∼ Calibration Ranges. We used archaeological data to propose date ranges, and in routes coincide with the savannah corridor beyond a negligible level 47% one case a fixed date, for four nodes of our tree (labeled a–dinFig.1).Thefour of route falls in savannah corridor. calibrations are as follows: (a) 5,000 B.P. or older for Bantoid, non-Bantu (58); (b) 4,000–5,000 B.P. for Narrow Bantu (13, 14, 16, 44, 59, 60); (c) 3,000–3,500 B.P. ACKNOWLEDGMENTS. We thank Gérard Philippson for the data on Eastern for the Mbam-Bubi ancestor (61); and (d) 2,500 B.P. for Eastern Bantu (62). We Bantu languages, Jean-Marie Hombert for the Grassfields languages, the used a uniform prior in our Bayesian tree inference for all calibration ranges. KongoKing research group (leader KB) for the data on the Kikongo languages (H), and Jean-Pierre Donzo and Guy Kouarata for data on C languages spoken in Congo and DRC. This work was supported by European Research Council Geographical Data. We recorded the latitude and longitude of the approxi- Advanced Investigator Award 268744 (Mother Tongue; to M.P.). K.B. was sup- mate centroid of each of our languages (Dataset S1), using data provided by ported by European Research Council Starting Grant No. 284126 (KongoKing) Bastin et al. (38) and fieldwork studies. and by the Special Research Fund of Ghent University.

1. Pagel M (2012) Wired for Culture: Origins of the Human Social Mind (Penguin, New 16. Ngomanda A, et al. (2009) Western equatorial African forest-savanna mosaics: A York), p 432. legacy of late Holocene climatic change. Clim Past 5(4):647–659. 2. Mellars P (2006) Going east: New genetic and archaeological perspectives on the 17. Bostoen K, et al. (2015) Middle to Late Holocene Palaeoclimatic change and the early modern human colonization of Eurasia. Science 313(5788):796–800. Bantu expansion in the rain forests of West Central-Africa. Curr Anthropol 56(3): 3. Hung H-c, et al. (2011) The first settlement of Remote Oceania: The Philippines to the 354–384. Marianas. Antiquity 85(329):909–926. 18. Klieman KA (2003) The Pygmies were our compass: Bantu and Batwa in the History of 4. Gray RD, Drummond AJ, Greenhill SJ (2009) Language phylogenies reveal expansion West Central Africa, Early Times to c. 1900 C.E. (Heinemann, Portsmouth, NH), p 253. pulses and pauses in Pacific settlement. Science 323(5913):479–483. 19. Vincens A, Schwartz D, Bertaux J, Elenga H, de Namur C (1998) Late Holocene climatic 5. Rogers DS, Ehrlich PR (2008) Natural selection and cultural rates of change. Proc Natl changes in western equatorial Africa inferred from pollen from Lake Sinnda, south- Acad Sci USA 105(9):3416–3420. ern Congo. Quat Res 50(1):34–45. 6. Diamond J (1997) Guns, Germs, and Steel: The Fates of Human Societies (Norton, New 20. White F (1979) The Guineo-Congolian Region and its relationships to other phy- York), p 480. tochoria. Bulletin Jardin Botanique National Belgique/Bulletin Nationale Plantentuin 7. Diamond J, Bellwood P (2003) Farmers and their languages: The first expansions. Belgie 49(1/2):11–55. Science 300(5619):597–603. 21. Gond V, et al. (2013) Vegetation structure and greenness in Central Africa from Modis 8. Russell T, Silva F, Steele J (2014) Modelling the spread of farming in the Bantu-speaking multi-temporal data. Philos Trans R Soc Lond B Biol Sci 368(1625):20120309. regions of Africa: An archaeology-based phylogeography. PLoS One 9(1):e87854. 22. Doumenge C, Gonmadje C, Doucet J-L, Maley J (2014) Intervalle de la Sangha: Mythe 9. Neumann K, et al. (2012) First farmers in the Central African rainforest: A view from ou réalité biogéographique? (Résumés, Stellenbosch, Afrique du Sud). southern Cameroon. Quat Int 249(6):53–62. 23. Currie TE, Meade A, Guillon M, Mace R (2013) Cultural phylogeography of the Bantu 10. Schwartz D (1992) Assèchement climatique vers 3000 BP et expansion bantu en Languages of sub-Saharan Africa. Proc R Soc 280(1762):20130695. Afrique Centrale atlantique: Quelques réflexions. Bull Soc Geol Fr 163(3):353–361. 24. Gignoux CR, Henn BM, Mountain JL (2011) Rapid, global demographic expansions 11. Vincens A, et al. (1994) Changement majeur de la végétation du lac Sinnda (vallée du after the origins of agriculture. Proc Natl Acad Sci USA 108(15):6044–6049. Niari, Sud-Congo) consécutif à l’assèchement climatique holocène supérieur: Apport 25. Li S, Schlebusch C, Jakobsson M (2014) Genetic variation reveals large-scale pop- de la palynologie. Comptes Rendus à l’Academie des Sciences de Paris 318:1521–1526. ulation expansion and migration during the expansion of Bantu-speaking peoples. 12. Vincens A, Buchet G, Servant M (2010) Vegetation response to the” African Humid Proc R Soc 281(1793):20141448. Period” termination in Central Cameroon (7° N)–new pollen insight from Lake Mbalang. 26. Adams J, Faure H, eds (1997) Review and Atlas of Palaeovegetation: Preliminary Land Clim Past 6(3):281–294. Ecosystem Maps of the World Since the Last Glacial Maximum (Oak Ridge National 13. Maley J, Brenac P (1998) Vegetation dynamics, palaeoenvironments and climatic Laboratory, Oak Ridge, TN). changes in the forests of western Cameroon during the last 28,000 years BP. Rev 27. Phillipson DW (1977) The Later Prehistory of Eastern and Southern Africa (Africana, Palaeobot Palynol 99(2):157–187. New York), p 323. 14. Maley J (2001) La destruction catastrophique des forêts d’Afrique centrale survenue il 28. Blench R (2012) Two vanished African maritime traditions and a parallel from South ya environ 2500 ans exerce encore une influence majeure sur la répartition actuelle America. Afr Archaeol Rev 29(2/3):273–292. des formations végétales. Syst Geogr Plants 71:777–796. 29. Heine B, Hoff H, Vossen R (1977) Neuere Ergebnisse zur Territorialgeschichte der 15. Maley J, Willis K (2010) Did a savanna corridor open up across the Central African Bantu. Zur Sprachgeschichte und Ethnohistorie in Afrika, eds Möhlig WJG, Rottland F, forests 2500 years ago? CoForChange. Available at www.coforchange.eu/products/ Heine B (Dietrich Reimer, Berlin), pp 57–72. notes/did_a_savanna_corridor_open_up_across_the_central_african_forests_2500_years_ago. 30. Henrici A (1973) Numerical classification of Bantu languages. African Language Studies Accessed 18 August 2015. 14:82–104.

13300 | www.pnas.org/cgi/doi/10.1073/pnas.1503793112 Grollemund et al. Downloaded by guest on September 24, 2021 31. Heine B (1973) Zur genetischen Gliederung der Bantu-Sprachen. Afr Uebersee 56(3): 63. Venditti C, Pagel M (2010) Speciation as an active force in promoting genetic evo- 164–185. lution. Trends Ecol Evol 25(1):14–20. 32. Ehret C (1999) Subclassifying Bantu: The evidence of stem morpheme innovation. 64. Guthrie M (1948) The Classification of the Bantu Languages (Oxford Univ Press for the Bantu Historical Linguistics. Theoretical and Empirical Perspectives, eds Hombert JM, International African Institute, London), p 91. Hyman LM (CSLI, Stanford, CA), pp 43–147. 65. Guthrie M (1967–1971) Comparative Bantu: An Introduction to the Comparative Lin- 33. Ehret C (2001) Bantu expansions: Re-envisioning a central problem of early African guistics and Prehistory of the Bantu Languages (Gregg International, London), 4 vols. history. Int J Afr Hist Stud 34(1):5–41. 66. Grollemund R (2012) Nouvelles approches en classification: Application aux langues 34. Holden CJ (2002) Bantu language trees reflect the spread of farming across sub- bantu du Nord-Ouest. Thesis (Université Lumière Lyon 2, Lyon). Saharan Africa: A maximum-parsimony analysis. Proc Biol Sci 269(1493):793–799. 67. Grollemund R, Branford S, Hombert J-M, Watters JR (2015) Towards a new phylo- 35. Holden CJ, Gray RD (2006) Rapid radiation, borrowing and dialect continua in the genetic classification of Bantoid languages. Towards Proto-Niger-Congo: Comparison Bantu languages. Phylogenetic Methods and the Prehistory of Languages, eds and Reconstruction, ed Watters JR (Language Science Press, Berlin). Forster P, Renfrew C (MacDonald Institute for Archaeological Research, Cambridge, 68. Yang Z (1996) Among-site rate variation and its impact on phylogenetic analyses. U.K.), pp 19–31. Trends Ecol Evol 11(9):367–372. 36. Möhlig WJG (1981) Stratification in the history of the Bantu languages. Sprache und 69. Pagel M, Meade A (2006) Estimating rates of lexical replacement on phylogenetic Geschichte in Afrika 3:251–317. trees of languages. Phylogenetic Methods and the Prehistory of Languages, eds 37. Bastin Y, Coupez A, de Halleux B (1983) Classification lexicostatistique des langues Forster P, Renfrew C (McDonald Institute Monographs, Cambridge, U.K.), pp 173–182. – bantoues (214 relevés). Bull Seances Acad R Sci Outre Mer 27(2):173 199. 70. Cornelissen E (2003) On microlithic quartz industries at the end of the Pleistocene in 38. Bastin Y, Coupez A, Mann M (1999) Continuity and Divergence in the Bantu Central Africa: The evidence from Shum Laka (NW Cameroon). Afr Archaeol Rev ’ Languages: Perspectives from a Lexicostatistic Study (Musée Royal de l Afrique Cen- 20(1):1–24. trale, Tervuren, Belgium), p 225. 71. Lavachery P (2001) The Holocene archaeological sequence of Shum Laka rock shelter ‘ ’ 39. Vansina J (1995) New linguistic evidence and the Bantu expansion . J Afr Hist 36(2): (Grassfields, Cameroon). Afr Archaeol Rev 18:213–247. – 173 195. 72. Asombang R, De Maret P (1992) Re-investigating Shum Laka (Cameroun). Nsi 10/11: 40. de Filippo C, Bostoen K, Stoneking M, Pakendorf B (2012) Bringing together linguistic 13–16. and genetic evidence to test the Bantu expansion. Proc R Soc B Biol Sci 279(1741): 73. de Maret P, et al. (1993) Preliminary results of the 1991-1992 field season at Shum – 3256 3263. Lake, Northwestern province, Cameroon. Nyame Akuma 39:13–15. 41. Desjardins T, et al. (2013) d13C variation of soil organic matter as an indicator of 74. de Maret P, Asombang R, Cornelissen E, Lavachery P, Moeyersons J (1995) Continuing – vegetation change during the Holocene in central Cameroon. C R Geosci 345:266 271. research at Shum Laka rock shelter, Cameroon (1993-1994 field season). Nyame Akuma 42. Bayon G, et al. (2012) Intensifying weathering and land use in Iron Age Central Africa. 43:2–3. Science 335(6073):1219–1222. 75. de Maret P, Clist B, Van Neer W (1987) Résultats des premières fouilles dans les abris- 43. Maley J, Giresse P, Doumenge C, Favier C (2012) Comment on “Intensifying weath- sous-roche de Shum Laka et Abeke au Nord-Ouest du Cameroun. Anthropologie ering and land use in Iron Age Central Africa”. Science 337(6098):1040, author reply 91(2):559–584. 1040. 76. Lavachery P (1998) De la pierre au métal: Archéologie des dépôts holocènes de l’abri 44. Neumann K, et al. (2012) Comment on “Intensifying weathering and land use in Iron sous roche de Shum Laka (Cameroun). PhD thesis (Université Libre de Bruxelles, Age Central Africa”. Science 337(6098):1040, author reply 1040. Bruxelles). 45. Eggert MKH (1992) The Central African rain forest: Historical speculation and ar- 77. Lavachery P (2003) A la lisière de la forêt: 10.000 ans d’interactions entre l’homme et chaeological facts. World Archaeol 24(1):1–24. l’environnement dans les Grassfields (Cameroun). Peuplements Anciens et Actuels de des 46. Wotzka H-P (1995) Studien zur Archäologie des Zentralafrikanischen Regenwaldes: Forêts Tropicales, Actes du Séminaire-Atelier Orléans 15-16 Octobre 1998, Collection Die Keramik des Inneren Zaïre-Beckens und Ihre Stellung im Kontext der Bantu- Colloques et Séminaires, eds Froment A, Guffroy J (IRD Editions, Paris), pp 89–102. Expansion (Heinrich-Barth-Institut, Köln, Germany). 78. Lavachery P, Cornelissen E (2000) Natural and cultural spatial patterning in the Late 47. Vansina J (1990) Paths in the Rainforests (Univ of Wisconsin Press, Madison, WI), p 428. Holocene deposits of Shum Laka Rock Shelter, Cameroon. J Field Archaeol 27(2): 48. Basalla G (1988) The Evolution of Technology (Cambridge Univ Press, Cambridge, U.K.). 153–158. 49. Arthur WB (2009) The Nature of Technology: What It Is and How It Evolves (Penguin, 79. Ribot I, Orban R, de Maret P (2001) The Prehistoric Burials of Shum Laka Rockshelter London). (NorthWest Cameroon) (Royal Museum for Central Africa, Tervuren, Belgium). 50. Powell A, Shennan S, Thomas MG (2009) Late Pleistocene demography and the ap- EVOLUTION – – pearance of modern human behavior. Science 324(5932):1298–1301. 80. de Maret P (1994-1995) Pits, pots and the Far West streams. Azania 29 30:318 323. 51. Pagel M (2009) Human language as a culturally transmitted replicator. Nat Rev Genet 81. Hassan FA (1996) Abrupt Holocene climatic events in Africa. Aspects of African – 10(6):405–415. Archaeology, eds Pwiti G, Soper R (Univ of Zimbabwe, Harare, Zimbabwe), pp 83 89. 52. Hombert J-M, Van der Veen L, Medjo Mve P (2011) ALGAB, Atlas Linguistique du 82. de Maret P (1992) Sédentarisation, agriculture et métallurgie du Sud-Cameroun: ’ GABon (Laboratoire Dynamique du Langage, Lyon). Available at www.ddl.ish-lyon. Synthèse des recherches depuis 1978. L archéologie au Cameroun. Actes du premier cnrs.fr/equipes/index.asp?Langue=FR&Equipe=8&Page=Action&ActionNum=48. Ac- Colloque international de Yaoundé, 6-9 janvier 1986, ed Essomba JM (Karthala, Paris), – cessed November 10, 2014. pp 247 262. 53. Drummond AJ, Ho SY, Phillips MJ, Rambaut A (2006) Relaxed phylogenetics and 83. de Maret P (1982) New survey of archaeological research and dates for West-Central – dating with confidence. PLoS Biol 4(5):e88. and North-Central Africa. J Afr Hist 23:1 15. 54. Yule GU (1925) A mathematical theory of evolution, based on the conclusions of 84. de Maret P, Clist B, Mbida C (1983) Belgian archaeologial mission in Cameroon, 1983 – – Dr. JC Willis, FRS. Philos Trans R Soc Lond, B Contain Pap Biol Character 213:21–87. field season. Nyame Akuma 23(5 6):5 6. 55. Larget B, Simon DL (1999) Markov chain Monte Carlo algorithms for the Bayesian 85. Clist B (1987) A critical reappraisal of the chronological framework of the early Urewe analysis of phylogenetic trees. Mol Biol Evol 16:750–759. Iron Age industry. Muntu 6:35–62. 56. Tuffley C, Steel M (1998) Modeling the covarion hypothesis of nucleotide substitution. 86. Phillipson DW (1993) African archaeology (Cambridge Univ Press, Cambridge, U.K.), p 389. Math Biosci 147(1):63–91. 87. Organ CL, Shedlock AM, Meade A, Pagel M, Edwards SV (2007) Origin of avian ge- 57. Pagel M, Meade A, Barker D (2004) Bayesian estimation of ancestral character states nome size and structure in non-avian dinosaurs. Nature 446(7132):180–184. on phylogenies. Syst Biol 53(5):673–684. 88. Lemey P, Rambaut A, Welch JJ, Suchard MA (2010) Phylogeography takes a relaxed 58. Lavachery P (2001) The Holocene archaeological sequence of Shum Laka Rock Shelter random walk in continuous space and time. Mol Biol Evol 27(8):1877–1885. (Grassfields, Western Cameroon). Afr Archaeol Rev 18(4):213–247. 89. Bouckaert R, et al. (2012) Mapping the origins and expansion of the Indo-European 59. Oslisly R (2001) The history of human settlement in the middle Ogooué valley (Gabon). language family. Science 337(6097):957–960. African Rain Forest Ecology and Conservation, eds Weber W, White LJT, Vedder A, 90. Anhuf D, Frankenberg P (2000) Die mittelholozäne Feuchtphase 5000 BP–Eine Veg- Naughton-Treves L (Yale Univ Press, New Haven), pp 101–118. etationsrekonstruktion für Afrika. Regensburger Geographische Arbeiten 33:99–126. 60. Ngomanda A, et al. (2005) Vegetation changes during the past 1300 years in western 91. Maley J (2002) A catastrophic destruction of African forests about 2,500 years ago still equatorial Africa: A high-resolution pollen record from Lake Kamalété, Lopé Reserve, exerts a major influence on present vegetation formations. IDS Bull 33(1):13–30. Central Gabon. Holocene 15(7):1021–1031. 92. White F (1983) The Guineo-Congolian Regional Centre of Endemism. The Vegetation of 61. Lavachery P (2003) A la lisière de la forêt. Peuplements Anciens et Actuels des Forêts African: A descriptive Memoir to Accompany the UNESCO/AETFAT/UNSO Vegetation Tropicales, eds Froment A, Guffroy J (IRD Editions, Paris), pp 89–102. Map of Africa (UNESCO, Paris), p 356. 62. Ashley CZ (2010) Towards a socialised archaeology of ceramics in Great Lakes Africa. 93. White F (1986) La Vegetation de L’Afrique (The Vegetation of Africa) (ORSTOM – UNESCO, Afr Archaeol Rev 27(2):135–163. Paris), p 384.

Grollemund et al. PNAS | October 27, 2015 | vol. 112 | no. 43 | 13301 Downloaded by guest on September 24, 2021