Rock art from at least 67,800 years ago in Sulawesi

A noteworthy finding of the past decade has been the discovery and dating of Pleistocene-aged rock art in Sulawesi, the largest island in Wallacea—the oceanic island zone separating the continental regions of Southeast Asia (Sunda) and Australia–New Guinea (Sahul)—and on the edge of the Sunda continental shelf (present-day Borneo) to its immediate west^1,2,3,4,5,8. In Sulawesi, U-series dating of rock art motifs (n = 23), including hand stencils (n = 13) and figurative paintings of animals (n = 7) and anthropomorphs (n = 3), has yielded minimum ages ranging from 51.2 kyr to 17 kyr (refs. ^1,2,3,4,5). All of the dated rock art sites are located in the approximately 450 km² lowland ‘tower’ karst region of Maros-Pangkep in the southwestern peninsula of Sulawesi. On the opposing southern peninsula, known as Southeast Sulawesi, an area that includes several satellite islands, the presence of rock art inside limestone karst caves was first reported in 1977 (ref. ⁹). The art in this region, however, has remained underinvestigated compared with Maros-Pangkep art (itself poorly known until recently). Beginning in 2019, we have undertaken a programme of recording and dating of parietal imagery in Southeast Sulawesi, resulting in the documentation of 44 sites (including 14 new localities), and the dating of 11 individual rock art motifs at 8 sites (Fig. 1). The dated motifs comprised seven hand stencils, and four other paintings (two of human figures and two non-figurative, geometric motifs). We used a new method of U-series dating (LA-U-Series analysis) reported elsewhere⁵ to obtain age estimates for calcium carbonate deposits that had formed in association with these rock art motifs. In all cases, we obtained minimum ages for the parietal images by dating calcite materials that had formed directly on top of the rock art through natural calcium carbonate precipitation. In some instances, we also obtained maximum ages by dating calcium carbonate layers immediately below the pigment layers.

a, The proposed modern human migration routes from Sunda to Sahul; the northern route through Wallacea is delineated by the red arrows, and the southern route is delineated by the blue arrow²¹. The red dots represent the areas with Pleistocene rock art in eastern Borneo and southwestern Sulawesi. b, Map of Sulawesi showing the location of dated rock art sites in Southeast Sulawesi: (1) Liang Metanduno, (2) Liang Pominsa, (3) Gua Kaghofighofine, (4) Madongka 3 rock shelter, (5) Waburi 1 rock shelter, (6) Lia Bunta rock shelter, (7) Gua Anawai, (8) Gua Mbokita and (9) Gua Berlian. m + MSL, metres above mean sea level. Map data adapted from General bathymetric maps (https://www.gebco.net/data_and_products/gridded_bathymetry_data/) and DEMNAS Badan Informasi Geospasial (https://tanahair.indonesia.go.id/demnas/#/demnas).

Full size image

All but one of the sampled hand stencils (n = 6) from Southeast Sulawesi date to the Pleistocene, whereas the remaining paintings mainly consist of motifs with more recent (that is, Holocene) minimum ages (Supplementary Table 1). The oldest minimum-age constraint we obtained is for a hand stencil (sample LMET2) from Liang Metanduno in Muna Island (Fig. 2). This motif is partly covered by ancient coralloid speleothems. This stencil is in a poor state of preservation, comprising only a 14 × 10 cm patch of faded pigment bearing a portion of the fingers and the adjoining palm area. The tip of one finger appears to have been artificially narrowed, either through the additional application of pigment or by moving the hand during pigment application⁹, a distinct type of hand stencil art thus far identified only in Sulawesi¹⁰. Dating results show that this stencil was produced at least 67.8  thousand years ago (ka) (71.6 ± 3.8 ka) (Fig. 3). An additional coralloid speleothem (denoted as LMET1) was sampled over another hand stencil located on the same panel 11 cm to the left of the previous one (Fig. 4). This motif is also partly covered over by ancient coralloid speleothems, with only a portion of the original stencil remaining in the form of a 14 × 9 cm area of sprayed pigment bearing negative impressions of three fingers. The pigment appears darker in colour than the other visible hand stencils on the panel. Sample LMET1 comprises two distinct pigment layers embedded within the calcium carbonate. The stratigraphically oldest pigment layer has a minimum date of 60.9 ka (70.5 ± 9.5 ka), whereas the second pigment layer has a minimum date of 21.5 ka (22.8 ± 1.3 ka) and a maximum date of 32.8 ka (30 ± 2.8 ka). Close inspection shows that the darker pigment overlies the lighter pigment on the visible portion of the image. The lighter pigment may correspond to an earlier hand stencil or represent residual sprayed pigment associated with the hand stencil sampled as LMET2. Alternatively, the lighter layer may be the remains of a different motif, now obscured or no longer visible, that predates the overlying pigment. Regardless, it is evident from LMET1 that two episodes of Pleistocene rock art production on this panel are separated by a period of at least 35 kyr. Finally, as the oldest narrow-fingered stencil previously dated yielded a minimum date of 17 ka (ref. ¹), we may conclude that this regionally unique variant of stencil art is much older than previously thought¹⁰.

a,b, Samples LMET1 and LMET2 collected from over distinct hand stencils: photograph of the rock art panel with the two hand stencils (a) and digital tracing (b).

Full size image

a, Transect view of sample LMET2 after removal from over a distinct hand stencil, highlighting the paint layer and the three integration zones (region of integration (ROI)) and associated date calculation. b, Schematic showing the main features. c, LA-MC-ICPMS imaging of LMET2 isotopic activity ratios for ²³²Th/²³⁸U. d, As in c, for ²³⁰Th/²³⁸U. e, As in c, for ²³⁴U/²³⁸U. f, As in c, for concentrations for thorium-232. g, As in c, for concentrations of uranium-238.

Full size image

a, Transect view of sample LMET1 after removal from over a distinct hand stencil, highlighting the paint layer and the five integration zones (ROI) and associated date calculation. b, Schematic showing the major features. c, LA-MC-ICPMS imaging of LMET2 isotopic activity ratios for ²³²Th/²³⁸U. d, As in c, for ²³⁰Th/²³⁸U. e, As in c, for ²³⁴U/²³⁸U. f, As in c, for concentrations for thorium-232. g, As in c, for concentrations of uranium-238.

Full size image

Liang Metanduno also contains paintings that appear to be much younger than the two dated hand stencils, based on the extent of weathering and other observations. These much fresher-looking artworks, executed using a brownish pigment, have been applied over existing speleothems and the ‘new’ cave wall surface exposed by the exfoliation of the older surface bearing the hand stencils, as well as over the dated hand stencils themselves. Among the images, a figurative representation of a bird, probably a chicken (Gallus sp.), measuring 13.5 cm × 12 cm, is located between the two dated hand stencils (Supplementary Information). The depicted subject matter (for example, introduced domestic fauna) attributes these figures to or after the arrival of Austronesian-speaking ‘Neolithic’ societies in the region^11,12 (about 4–3.5 ka). At a nearby cave, Gua Pominsa, a similar brown-pigment image showing a human figure possibly holding material culture objects in both hands yielded a minimum date of 3.9 ka (5 ka ± 1.1 ka) (GPOM1) (Extended Data Fig. 1). This figure is possibly the oldest evidence of Austronesian culture identified thus far in the region.

Dating of several other hand stencils from southeastern Sulawesi produced minimum dates falling within the Late Pleistocene epoch. In Gua Mbokita, a limestone cave in the Menui Kepulauan district, two samples provided minimum dates of 44.7 ka (50.8 ± 6.1 ka) (MBK1) and 25.9 ka (31.1 ± 5.2 ka) (MBK2) (Extended Data Fig. 2). Two separate hand stencils from Gua Anawai were also dated to a minimum of 19.1 ka (19.9 ± 0.8 ka) (ANW1) and 20.1 ka (22.1 ± 2.1 ka) (ANW2), respectively (Extended Data Figs. 3 and 4). A maximum date of 20.4 ka was also obtained for ANW1, suggesting that these hand stencils were produced between 20.1 and 20.4 ka and that people were creating them until at least the Last Glacial Maximum.

Further dating results for other motifs, including artworks that are likely to be much younger than these hand stencils (although we only have minimum ages for the former), are reported in Extended Data Figs. 5–9 and Supplementary Table 1.

Our dating investigation in Southeast Sulawesi shows that extremely old rock art is not just concentrated in the Maros-Pangkep district in the southwestern peninsula but occurs in other parts of this large Wallacean island (about 174,000 km²). These findings support the growing view that Sulawesi was host to a vibrant and longstanding artistic culture during the Late Pleistocene epoch^10,13. Two artworks we dated in Southeast Sulawesi also have far-reaching implications for our understanding of early human history. With minimum-age constraints of 67.8 ka (71.6 ± 3.8 ka) and 60.9 ka (70.5 ± 9.5 ka), the hand stencils from Liang Metanduno provide the oldest demonstrated minimum-age constraints reported to date worldwide for parietal art that can be attributed to our species. Previously reported evidence for very early rock art in Europe, a region first inhabited by Homo sapiens around 54 ka (ref. ¹⁴) (although this remains debated¹⁵), includes a sprayed disc with a published minimum date of 40.8 ka (41.4 ± 0.6 ka)¹⁶ and a hand stencil from northern Spain (Maltravieso) with a published minimum date of 66.7 ka (70.1 + 3.8–3.4 ka)⁶. The latter motif has been attributed to Neanderthals, but its dating has been controversial^{17,18,19,20,21,22,23,24} (Methods, section ‘Interpretation of U-series ages’).

There is no obvious method available to us to determine which human taxon produced the Liang Metanduno stencil (LMET2), and, to further complicate matters, Sulawesi was host to archaic hominins before colonization by modern humans^25,26—although it remains unclear whether there was a chronological overlap with our species on the island²⁷. Nevertheless, we attribute the earliest cave art of Muna to H. sapiens based on the added technical and stylistic complexity of the intentionally modified fingers on the hand stencil and the close fit with the known arrival time of our species in the region.

About the latter, the colonization of Sahul by modern humans is now thought to have occurred by at least 68.7–59.3 ka (65.0 ka ± 3.7, 5.7 ka), based on excavated findings at Madjedbebe in northern Australia⁷. The initial peopling of this continent is widely considered to have involved the first planned, long-distance sea crossing undertaken by our species. The earliest intentional ocean voyages over shorter distances took place when H. sapiens spread beyond the extreme eastern edge of the Eurasian continent (Sunda) and crossed the ‘Wallace Line’, the main biogeographical boundary marking the easternmost limit of the floral and faunal communities of Asia. Modelling data, based on palaeogeographic, environmental and demographic reconstructions, favours two main routes from Sunda to Sahul^{28,29,30,31,32}. The northern route began in present-day Borneo (Sunda) and passed through Sulawesi and from there to Sahul through Misool or the western tip of Papua²⁸. The southern route runs eastwards from Java through the Lesser Sunda Islands to northwestern Australia (through Timor)²⁸ (Fig. 1). Current archaeological evidence demonstrates an early presence of modern humans at the nominal beginning and end points of these routes^7,33,34 (Laos by 86–68 ka, Sumatra by 73–63 ka and Australia by 69–59 ka). Although modelling suggests an early dispersal to Sahul through the northern route, possibly with a later (that is, about 45 ka) secondary dispersal along the southern route^30,31,35, there has been no archaeological evidence so far to bolster this scenario. A gap of about 15 kyr in modern human sites between Sumatra and Australia and the broader region between them has made it impossible to support any particular migratory pathway to Sahul.

The newly discovered Liang Metanduno hand stencils are the oldest archaeological evidence revealed so far for the presence of our species in Wallacea. Notably, the dated rock art presented here is the easternmost in Wallacea and is located along the northern route to Sahul, as is all of the dated Pleistocene cave art reported thus far from Island Southeast Asia^1,2,3,4,5,8. This strongly supports the view that the initial peopling of the Sahul landmass involved maritime journeys through northern Wallacea. It is also evident from our findings that the first modern humans to reach Sahul about 65 ka brought with them a sophisticated artistic culture, with implications for the likely age of early painted rock art sequences in northern Australia³⁶. Given the extreme antiquity of the Liang Metanduno stencil, it seems reasonable to anticipate that evidence for rock art production of a similar age will be present at other locations along the northern route to Sahul.

All dating work was undertaken at the BIOMICS laboratory in the Geoarchaeology and Archaeometry Research Group (GARG) of Southern Cross University (Lismore, New South Wales, Australia). U-series measurements were obtained using an ESI NW193 ArF excimer laser ablation unit coupled to a MC-ICPMS ThermoFishers Neptune XT. Each sample was measured by a succession of parallel rasters across the exposed polished cross-section (geochemical imaging), allowing us to reconstruct an isotopic map of the precipitated calcite, or by successive parallel rasters along the concretion formation of the sample. Imaging rasters had different lengths to adapt to the irregular shape of the sample using the following parameters for mapping: a square spot size of 44 μm × 44 μm using the infinite aperture of the laser system matched by a translation speed of 21 μm s⁻¹ and integration time of 2.097 s on the MC-ICPMS Neptune XT. This combination of parameters allowed us to obtain within <0.1% a pixel on the map equivalent to a 44 μm × 44 μm data point (the exact translation speed to obtain an exact data-pixel of 44 μm × 44 μm would be 20.982 μm s⁻¹).

On non-mapped samples, individual rasters had systematically the same length within the same sample around about 700 μm following parameters for mapping: a square spot size of 110 μm × 110 μm using the infinite aperture of the laser system with a translation speed of 5 μm s⁻¹ and integration time of 4.194 s on the MC-ICPMS Neptune XT.

Other parameters for the data acquisition were as follows: 900 ml min⁻¹ UHP He and 6 ml min⁻¹ UHP nitrogen for the gas flow from the chamber to the ICPMSs, rep rate of 100 Hz for the laser frequency and an average of 1.92 J cm⁻² sample fluence. ²³⁴U and ²³⁰Th were measured simultaneously, with uranium in the centre Faraday cup coupled with a secondary electron multiplier (SEM) and thorium on the L3 Faraday cup coupled to an ion counter (IC). All other Faraday cups were set to using high-gain 10¹¹ Ω amplifiers. The cup configuration was as follows: L3/IC(230), L2(232), L1(233), C/SEM(234), H1(235), H2(236) and H3(238). Baseline and drifts were corrected using NIST 610 and NIST 612 glass standards, whereas two corals (the MIS7 Faviid and MIS5 Porites corals from the Southern Cook Islands)³⁷ were used to correct ²³⁴U/²³⁸U and ²³⁰Th/²³⁸U ratios and assess the accuracy of measurements.

Image and data processing

Isotopic mapping obtained by LA-MC-ICP-MS data was extracted using the Iolite 4 software package³⁸. The data were accumulated in a single file on the MC-ICPMS Neptune XT as follows: 5 min background, NIST610 (3×), NIST612 (3×), STD1 (3×), STD2 (3×), sample rasters (n×), STD2 (3×), STD1 (3×), NIST612 (3×), NIST610 (3×) and 5 min background. For sample imaging sequences longer than 2 h, a set of standards (for example, STD1 (3×) and STD2 (3×)) was incorporated in the middle of the measurement. Data reduction used NISTs to assess drift and the 5-min background on each side of the measurements for baseline. One standard (MK10) was used for correction of the isotopic ratios, whereas the other one (MK16) was used as known values to check data accuracy (including for matrix effect). The images were produced using a spectrum gradient colour distribution, with either linear or logarithmic scale (specified for each sample on the isotopic maps). Regions of integration (ROIs) were carefully selected on the ²³²Th/²³⁸U and ²³⁰Th/²³⁸U isotopic ratio maps and Uppm maps to be as close to the paint layer as possible, while avoiding diagenetic zones. Data errors were extracted and reported at 2 standard error. ROIs located immediately above the pigment layers were selected for calculating the minimum ages relating to the underlying paintings. U-series data were integrated for individual ROIs, resulting in U-series ages and associated errors. Sufficient data points were also selected to minimize the errors. The integration area of each ROI is reported in μm² (Supplementary Table 1).

Raster measurement

The data were accumulated in a single file on the MC-ICPMS Neptune XT as follows: 5 min background, NIST610 (3×), NIST612 (3×), STD1 (3×), STD2 (3×), sample rasters (n×), STD2 (3×), STD1 (3×), NIST612 (3×), NIST610 (3×) and 5 min background. For sample imaging sequences or several raster measurements in the row longer than 2 h, a set of standards (for example, STD1 (3×) and STD2 (3×)) was incorporated in-between samples. Data reduction used NISTs to assess drift and the 5-min background on each side of the measurements for baseline. One standard (MK10) was used for the correction of the isotopic ratios, and the other (MK16) was used as known values to check data accuracy (including for matrix effect). Rasters are placed starting as close as the paint layer and spaced uniformly along the axis perpendicular to the first raster. Yet, to avoid an obvious diagenetic zone, the position of the raster in certain circumstances may have been adjusted (Extended Data Figs. 1, 3, 7). U-series data were integrated for each individual raster, resulting in U-series ages and associated errors (Supplementary Table 1).

Tuning procedures were carried out using a NIST 610 glass standard with the following parameters: a 50 µm spot size, translation speed set to 5 µm s⁻¹, and a fluence of >1 J cm⁻² at the sample surface. The Multicollector Zoom Optic is optimized to ensure optimal peak shape for ²³⁸U in Faraday cup H3, with signal adjustments made to achieve a minimum of 1 V for ²³⁸U and maintain a 0.85 ratio factor between ²³²Th and ²³⁸U voltages (Faraday cups L2 and H3, respectively). In the case of the samples analysed in this study, the average measures for ²³⁸U exceeded 3 V. Following tuning with NIST 610, fine-tuning was carried out using A MIS7 (MK10) Faviid coral from the southern Cook Islands³⁷, with particular attention to comparing R48 values between the laser and the solution. For geochemical imaging, a spot size of 44 µm with a laser rastering speed of 21 µm s⁻¹ with a 2.097 s integration time was optimal for most circumstances, with the number of blocks varying depending on the measurement sequence. Yet, on average, measurements were conducted with a total block number of about 10, with 200 cycles per block. Individual raster measurements were carried out a spot size of 100 µm with a laser rastering speed of 5 µm s⁻¹ and a 4.194 s integration time.

The MK10 coral standard is also used to correct the ²³⁴U/²³⁸U and ²³⁰Th/²³⁸U ratios. Another coral sample, a MSI5 (MK16) Porite coral also from the southern Cook Islands³⁷, is used to independently check the accuracy of the measurements and dating results. Both coral samples were reduced to powder to homogenize their values before being run by solution MC-ICPMS. The remaining homogeneous powder samples were consolidated using a uranium- and thorium-free resin and used as standards for LA-MC-ICPMS. Standards were measured in the following order: 3× NIST610, 3× NIST612, 3× CoralSTD1 and 3× CoralSTD2 at the beginning of each sample and at the end in a mirroring order. The aforementioned list of standards was measured in between each sample with a maximum of 2 h in-between standard sets. We used the same coral-based laser-ablation U–Th calibration approach as in our previous study, in which we demonstrated that laser ablation and solution U-series ages of calcite crusts were indistinguishable within analytical uncertainty, indicating that any matrix effects between coral (aragonite) and calcite are negligible under comparable analytical conditions⁵.

It is not unusual for secondary calcium carbonate to contain detrital materials such as wind-blown or waterborne sediments, which can contaminate the sample and lead to U-series ages that appear older than they actually are. This occurs because of pre-existing ²³⁰Th in the detrital components. As it is impossible to physically separate detrital/initial ²³⁰Th from radiogenic ²³⁰Th for measurement, corrections are made using an assumed ²³⁰Th/²³²Th activity ratio. Typically, the average continental crust value of ²³²Th/²³⁸U = 3.8 is used for these corrections, with an uncertainty of 100%. The measured ²³⁰Th/²³²Th activity ratio reflects the degree of detrital contamination, with the higher values (>20) indicating a smaller effect on the calculated age, whereas the lower values (<20) suggest a major correction is needed. In samples with a measured ²³⁰Th/²³²Th activity ratio >20, the detrital ²³⁰Th accounts for only a small percentage of the total ²³⁰Th in the sample.

Sometimes, the assumed ²³⁰Th/²³²Th activity ratio of 0.8 (±100%) for the detrital component may not cover all situations³⁹. If the actual ²³⁰Th/²³²Th activity ratio in the detrital component substantially deviates from this assumed range, the detrital correction scheme may introduce considerable bias, especially in samples with a ²³⁰Th/²³²Th activity ratio <20. In these situations, the ²³⁰Th/²³²Th activity ratio in the detrital component can be obtained through direct measurement of sediments associated with speleothems⁶ or computed using isochron methods or stratigraphic constraints⁴⁰. In our case, our samples were relatively pure: the ²³⁰Th/²³²Th activity ratio of individual aliquots was extremely high. Corrections for detrital components were therefore calculated assuming the bulk-Earth values.

A conceivable problem with the U-series dating method is that calcium carbonate accretions can behave as an open system for uranium, in which the element can be leached out of the accretions or remobilized⁴¹. In these instances, the calculated ages will be too old because the dating method relies on the accurate measurement of uranium compared with its decay product ²³⁰Th. In this study, this problem was tackled by avoiding porous samples and by measuring ROIs and multiple rasters from every sample. The ages of these subsamples were in chronological order or of similar ages within error, confirming the integrity of the dated coralloids. If uranium had leached out of the samples, a reverse age profile would have been evident (the ages would have gotten older towards the surface). Age calculations were performed using IsoplotR v.6.8 (ref. ⁴²). Ages are reported at 2 standard error.

Interpretation of U-series ages

Ages derive from LA-U-series analysis of carbonate overgrowths and therefore provide minimum-age constraints, that is, conservative lower bounds on painting age. Differences between minima do not by themselves demonstrate that one painting predates another. Accordingly, comparisons throughout the text refer to differences in minimum-age constraints unless stated otherwise.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

The data supporting the findings of this study are provided in the Supplementary Information. Raw and additional source data are available from publicly available at Zenodo⁴³ (https://zenodo.org/uploads/17561904).

1. Aubert, M. et al. Pleistocene cave art from Sulawesi, Indonesia. Nature 514, 223–227 (2014).

   Article  ADS  CAS  PubMed  Google Scholar 

2. Aubert, M. et al. Palaeolithic cave art in Borneo. Nature 564, 254–257 (2018).

   Article  ADS  CAS  PubMed  Google Scholar 

3. Aubert, M. et al. Earliest hunting scene in prehistoric art. Nature 576, 442–445 (2019).

   Article  ADS  CAS  PubMed  Google Scholar 

4. Brumm, A. et al. Oldest cave art found in Sulawesi. Sci. Adv. 7, eabd4648 (2021).

   Article  ADS  PubMed  PubMed Central  Google Scholar 

5. Oktaviana, A. A. et al. Narrative cave art in Indonesia by 51,200 years ago. Nature 631, 814–818 (2024).

   Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

6. Hoffmann, D. L. et al. U-Th dating of carbonate crusts reveals Neandertal origin of Iberian cave art. Science 359, 912–915 (2018).

   Article  ADS  CAS  PubMed  Google Scholar 

7. Clarkson, C. et al. Human occupation of northern Australia by 65,000 years ago. Nature 547, 306–310 (2017).

   Article  ADS  CAS  PubMed  Google Scholar 

8. Aubert, M., Brumm, A. & Taçon, P. S. C. The timing and nature of human colonization of Southeast Asia in the late Pleistocene. Curr. Anthropol. 58, S553–S566 (2017).

   Article  Google Scholar 

9. Oktaviana, A. A. in The Archaeology of Sulawesi: Current Research on the Pleistocene to the Historic Period (eds O’Connor et al.) Vol. 48, 61–77 (ANU Press, 2018).

10. Oktaviana, A. A. et al. Hand stencils with and without narrowed fingers at two new rock art sites in Sulawesi, Indonesia. Rock Art Res. 33, 32–48 (2016).

    Google Scholar 

11. Bellwood, P. First Islanders: Prehistory and Human Migration in Island Southeast Asia (John Wiley, 2017).

12. Cochrane, E. E., Rieth, T. M. & Filimoehala, D. The first quantitative assessment of radiocarbon chronologies for initial pottery in Island Southeast Asia supports multi-directional Neolithic dispersal. PLoS ONE 16, e0251407 (2021).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

13. Brumm, A. et al. Early human symbolic behavior in the late Pleistocene of Wallacea. Proc. Natl Acad. Sci. USA 114, 4105–4110 (2017).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

14. Slimak, L. et al. Modern human incursion into Neanderthal territories 54,000 years ago at Mandrin, France. Sci. Adv. 8, eabj9496 (2022).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

15. Zilhão, J., d’Errico, F., Banks, W. E. & Teyssandier, N. A data-driven paradigm shift for the Middle-to-Upper Palaeolithic transition and the Neandertal debate. Quat. Environ. Hum. 2, 100037 (2024).

    Google Scholar 

16. Pike, A. W. G. et al. U-series dating of Paleolithic art in 11 caves in Spain. Science 336, 1409–1413 (2012).

    Article  ADS  CAS  PubMed  Google Scholar 

17. Aubert, M., Brumm, A. & Huntley, J. Early dates for ‘Neanderthal cave art’ may be wrong. J. Hum. Evol. 125, 215–217 (2018).

    Article  PubMed  Google Scholar 

18. Pearce, D. G. & Bonneau, A. Trouble on the dating scene. Nat. Ecol. Evol. 2, 925–926 (2018).

    Article  PubMed  Google Scholar 

19. White, R. et al. Still no archaeological evidence that Neanderthals created Iberian cave art. J. Hum. Evol. 144, 102640 (2020).

    Article  Google Scholar 

20. Hoffmann, D. L. et al. Response to comment on “U-Th dating of carbonate crusts reveals Neandertal origin of Iberian cave art”. Science 362, eaau1736 (2018).

    Article  PubMed  Google Scholar 

21. Hoffmann, D. L. et al. Dates for Neanderthal art and symbolic behaviour are reliable. Nat. Ecol. Evol. 2, 1044–1045 (2018).

    Article  PubMed  Google Scholar 

22. Hoffmann, D. L. et al. Response to Aubert et al.’s reply ‘Early dates for “Neanderthal cave art”may be wrong’ [J. Hum. Evol. 125 (2018), 215–217]. J. Hum. Evol. 135, 102644 (2019).

    Article  Google Scholar 

23. Hoffmann, D. L. et al. Response to White et al.’s reply: ‘Still no archaeological evidence that Neanderthals created Iberian cave art’ [J. Hum. Evol. (2020) 102640]. J. Hum. Evol. 144, 102810 (2020).

    Article  PubMed  Google Scholar 

24. Standish, C. D. et al. The age of hand stencils in Maltravieso cave (Extremadura, Spain) established by U-Th dating, and its implications for the early development of art. J. Archaeol. Sci. Rep. 61, 104891 (2025).

    Google Scholar 

25. van den Bergh, G. D. et al. Earliest hominin occupation of Sulawesi, Indonesia. Nature 529, 208–211 (2016).

    Article  ADS  PubMed  Google Scholar 

26. Hakim, B. et al. Hominins on Sulawesi during the Early Pleistocene. Nature 646, 378–383 (2025).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

27. Brumm, A. et al. A reassessment of the early archaeological record at Leang Burung 2, a Late Pleistocene rock-shelter site on the Indonesian island of Sulawesi. PLoS ONE 13, e0193025 (2018).

    Article  PubMed  PubMed Central  Google Scholar 

28. Birdsell, J. in Sunda and Sahul: Prehistoric Studies in Southeast Asia, Melanesia and Australia (eds Allen, J. et al.) 113–167 (Academic Press, 1977).

29. Kealy, S., Louys, J. & O’Connor, S. Reconstructing palaeogeography and inter-island visibility in the Wallacean Archipelago during the likely period of Sahul colonization, 65–45 000 years ago. Archaeol. Prospect. 24, 259–272 (2017).

    Article  Google Scholar 

30. Kealy, S., Louys, J. & O’Connor, S. Least-cost pathway models indicate northern human dispersal from Sunda to Sahul. J. Hum. Evol. 125, 59–70 (2018).

    Article  PubMed  Google Scholar 

31. Norman, K. et al. An early colonisation pathway into northwest Australia 70–60,000 years ago. Quat. Sci. Rev. 180, 229–239 (2018).

    Article  ADS  Google Scholar 

32. Bird, M. I. et al. Early human settlement of Sahul was not an accident. Sci. Rep. 9, 8220 (2019).

    Article  ADS  PubMed  PubMed Central  Google Scholar 

33. Westaway, K. E. et al. An early modern human presence in Sumatra 73,000–63,000 years ago. Nature 548, 322–325 (2017).

    Article  ADS  CAS  PubMed  Google Scholar 

34. Freidline, S. E. et al. Early presence of Homo sapiens in Southeast Asia by 86–68 kyr at Tam Pà Ling, Northern Laos. Nat. Commun. 14, 3193 (2023).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

35. Bradshaw, C. J. A. et al. Minimum founding populations for the first peopling of Sahul. Nat. Ecol. Evol. 3, 1057–1063 (2019).

    Article  PubMed  Google Scholar 

36. Brumm, A., Oktaviana, A. A. & Aubert, M. in Deep-Time Images in the Age of Globalization (eds Abadia, O. M. et al.) 31–44 (Springer, 2024).

37. Woodroffe, C. D., Short, S. A., Stoddart, D. R., Spencer, T. & Harmon, R. S. Stratigraphy and chronology of late Pleistocene reefs in the southern Cook Islands, South Pacific. Quat. Res. 35, 246–263 (1991).

    Article  Google Scholar 

38. Paton, C., Hellstrom, J., Paul, B., Woodhead, J. & Hergt, J. Iolite: freeware for the visualisation and processing of mass spectrometric data. J. Anal. At. Spectrom. 26, 2508 (2011).

    Article  CAS  Google Scholar 

39. Zhao, J., Yu, K. & Feng, Y. High-precision ²³⁸U–²³⁴U–²³⁰Th disequilibrium dating of the recent past: a review. Quat. Geochronol. 4, 423–433 (2009).

    Article  Google Scholar 

40. Hellstrom, J. U–Th dating of speleothems with high initial 230Th using stratigraphical constraint. Quat. Geochronol. 1, 289–295 (2006).

    Article  Google Scholar 

41. Plagnes, V. et al. Cross dating (Th/U-¹⁴ C) of calcite covering prehistoric paintings in Borneo. Quat. Res. 60, 172–179 (2003).

    Article  CAS  Google Scholar 

42. Vermeesch, P. IsoplotR: a free and open toolbox for geochronology. Geosci. Front. 9, 1479–1493 (2018).

    Article  CAS  Google Scholar 

43. Oktaviana, A. A. et al. Data for ‘Rock art from at least 67,800 years ago in Sulawesi’. Zenodo https://zenodo.org/uploads/17561904 (2025).

Download references

Author notes

1. These authors contributed equally: Adhi Agus Oktaviana, Renaud Joannes-Boyau, Adam Brumm, Maxime Aubert

2. Deceased: Pindi Setiawan

Authors and Affiliations

1. School of Humanities, Languages and Social Science, Griffith University; Griffith Centre for Social and Cultural Research (GCSCR), Griffith University, Nathan, Queensland, Australia

   Adhi Agus Oktaviana, Shinatria Adhityatama, Andrea Jalandoni & Maxime Aubert

2. Pusat Riset Arkeometri, Organisasi Riset Arkeologi, Bahasa, dan Sastra, Badan Riset dan Inovasi Nasional, Jakarta, Indonesia

   Adhi Agus Oktaviana, Dewangga Eka Mahardian, Fadhlan S. Intan & Sofwan Noerwidi

3. Geoarchaeology and Archaeometry Research Group, Southern Cross University, East Lismore, New South Wales, Australia

   Renaud Joannes-Boyau & Maxime Aubert

4. ARC Training Centre for Archaeology in the Resources Sector, Southern Cross University, Lismore, New South Wales, Australia

   Renaud Joannes-Boyau

5. Pusat Riset Arkeologi Prasejarah dan Sejarah, Organisasi Riset Arkeologi, Bahasa, dan Sastra, Badan Riset dan Inovasi Nasional, Jakarta, Indonesia

   Budianto Hakim & Irfan Mahmud

6. Pusat Kolaborasi Riset Arkeologi Sulawesi, Makassar, Indonesia

   Budianto Hakim, Basran Burhan, Ratno Sardi, Iwan Sumantri, Rustan Lebe, Andi Jusdi, Sofwan Noerwidi, Marlon N. R. Ririmasse, Irfan Mahmud & Akin Duli

7. Australian Research Centre for Human Evolution (ARCHE), School of Environment and Science, Griffith University, Nathan, Queensland, Australia

   Basran Burhan & Adam Brumm

8. Balai Pelestarian Kebudayaan Wilayah XIX, Makassar, Indonesia

   Ratno Sardi,  Iswadi, Imran Ilyas, Abdullah Abbas, Andi Jusdi & Laode Darma

9. Pusat Riset Arkeologi Lingkungan, Maritim, dan Budaya Berkelanjutan, Organisasi Riset Arkeologi, Bahasa, dan Sastra, Badan Riset dan Inovasi Nasional, Jakarta, Indonesia

   Shinatria Adhityatama & Marlon N. R. Ririmasse

10. Korps Pencinta Alam, Universitas Hasanuddin, Makassar, Indonesia

    Hamrullah & Iwan Sumantri

11. Departemen Arkeologi, Fakultas Ilmu Budaya, Universitas Hasanuddin, Makassar, Indonesia

    Iwan Sumantri, Akin Duli, Laode M. Aksa, M. Nur,  Suryatman & Andi Muhammad Saiful

12. Balai Pelestarian Kebudayaan Wilayah XVIII, Palu, Indonesia

    M. Tang, Nasrullah Aziz & Sri Wigati

13. Badan Layanan Umum, Museum dan Cagar Budaya, Direktorat Jenderal Kebudayaan, Jakarta, Indonesia

    Rustan Lebe

14. Museum Sulawesi Tengah, Palu, Indonesia

    Iksam

15. Balai Pelestarian Wilayah XVII, Manado, Indonesia

    Faiz

16. Departemen Arkeologi, Universitas Indonesia, Depok, Indonesia

    M. Sabri

17. Yayasan Bumi Toala Indonesia, Makassar, Indonesia

    Fardi Ali Syahdar

18. Museum Mulawarman, Tenggarong, Indonesia

    Eriani

19. Jurusan Arkeologi, Universitas Haluoleo, Kendari, Indonesia

    N. A. Hidayatullah,  Nurmin & S. H. Sindara

20. Dinas Kebudayaan Buton Selatan, Batauga, Indonesia

    Laode Zulman

21. KK Desain Komunikasi Visual, Fakultas Seni Rupa dan Desain, Institute Teknologi Bandung, Bandung, Indonesia

    Pindi Setiawan

22. School of Environment and Science, Griffith University, Nathan, Queensland, Australia

    Adam Brumm

Contributions

A.A.O., M.A., R.J.-B. and A.B. conceived and led the research with senior collaborators B.H., M.N.R.R., S.N., I.M., A.D., M.N., M.T., F.S.I., L.M.A., Iksam, I.S., P.S. and R.L. Rock art dating samples at Metanduno Cave and Pominsa Cave were collected by M.A., A.A.O., B.H., L.M.A. and L.D. Rock art dating samples from Anawai Cave were collected by A.A.O., N.A.H., A.M.S., R.L., I.I., A. Jusdi and A.A., and rock art dating samples at Mbokita Cave and Berlian Cave by A.A.O., N.A.H., Faiz, N.A., Iksam, F.A.S. and S.W. Rock art dating samples at Kaghofighofine Cave, Madongka 3, Lia Bunta and Waburi 1 were collected by A.A.O., N.A.H., Eriani, B.B., L.Z., Nurmin, S.H.S. and Suryatman. Rock art was recorded and analysed in the field by A.A.O., Hamrullah, A. Jusdi, B.B., D.E.M. and R.S., and A.A.O. produced the digital tracings of parietal motifs. A.A.O., D.E.M., A. Jalandoni, R.S. and I.I. produced the 3D model of Leang Metanduno. A.A.O. and M.A. identified and collected the coralloid speleothem samples at the rock art sites and prepared the samples for LA-U-series analysis. All in-field sampling involving rock art was done under the direct supervision of R.L., Faiz and A.A.O.; R.J.-B. and M.A. conducted the U-series dating, interpreted the data, and discussed and approved correction factors and other methodological details pertinent to the dating results. A.A.O., M.A., R.J.-B. and A.B. wrote the paper, with key contributions from the other authors. The figures were produced and/or designed by A.B., R.J.-B., S.A., A.A.O. and M.A. All authors reviewed and edited the paper.

Corresponding author

Correspondence to Maxime Aubert.

Competing interests

The authors declare no competing interests.

Peer review information

Nature thanks the anonymous reviewer(s) for their contribution to the peer review of this work. Peer reviewer reports are available.

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions