Great insight! Would the continuity of material culture from 130,000 to 40,000 years ago disfavor a hypothesis that the art was made by an earlier Out of Africa, or pre-OOA, homo sapien population?
Good question. I don't discount the possibility of an early dispersal from Africa, which might help explain some of record in China, Southeast Asia, and Sumatra of fossils that resemble modern humans. On the other hand, maybe some of those sites are late Denisovans, and others may be misdated. DNA evidence from a site with strong geochronology like Tam Pà Ling might be the only way we will know.
How far into island Southeast Asia could an early out-of-Africa population have gone? Is it possible that an early out-of-Africa population might be the first inhabitants of Australia? Were they on Sulawesi? Again, nothing precludes this possibility, but the data put some constraints on it. The DNA data from living people suggests that DNA from an early out-of-Africa group did not survive in any measurable fraction up to the present day. So if such a group existed, there must have been a near-total replacement by later people, replacement to a higher degree than Denisovans experienced.
There are several sites that I think provide some information about this. The Leang Bulu Bettue site with its consistent record from maybe as early as 200,000 up to 40,000 years ago is one of two deep chronologies. I think your observation is the correct one: An early out-of-Africa group should have shown up sometime after 120,000 years or so, and there's nothing in the record to suggest it. However, an important caveat is the sparse nature of all artifacts from these early parts of Leang Bulu Bettue. I'm not positive you'd notice a small difference with the small number of artifacts.
The other deep record is Liang Bua on Flores. That one is pretty unequivocally a case where the ancestral island population was present as late as 65,000 years ago, and was not an early out-of-Africa group. I'm waiting for more updates from Liang Bua on the post-floresiensis chronology. Trader's Cave on Borneo is starting to look like a deep record may emerge, and the one tooth may suggest this also was not an early out-of-Africa group at around 50,000 years ago. And then there is the Ngandong series of fossils, between 120,000 and 106,000 years old, also not an early out-of-Africa group.
So I don't rule out the possibility that an early out-of-Africa dispersal was part of the mix in this region but none have been found, and there seems to be at least some evidence that longer-term inhabitants were present and making archaeology.
The reason I bring up Denisovans is mainly inspired by genetics of present-day people. I take as a good chance that the Denisovans themselves had mastered seafaring to a degree that they could reach Sahul. I also think there's a good chance that the morphology of the Ngandong crania may be Denisovan rather than Homo erectus. But that, too, is not proved.
Much remains up in the air. Maybe the Madjedbebe, Australia, early date is incorrect, so the idea of an early habitation is not earlier than around 50,000 years ago. Maybe the Denisovan genes in today's people trace to different mainland Southeast Asia groups of Denisovans. And maybe the Sulawesi hominins were another case like floresiensis that was highly endemic. It may take a while to sort this all out.
Still I tend to think that early HS were in Sahul region well before 40 kY or even 45 kY - probably by 70-72 kY, and Sahul settlement was of two stages, consistent with "long chronology". There are too many anomalies with Austr. aboriginals and PNG genetics. First, they show a very substantial difference in coalescence time with Africans compared to all other Eurasians, even after filtering out Denisovan admixture, and attempts to explain it otherwise are not convincing. They are also huge outliers in PCA whole-genome space (especially if you look at higher components - 4,5,6, not just usual 1,2). The negation of the "long chronology" is largely based on the similar level of N. admixture in Austr/PNG to all other Eurasians (e.g. Allen & O'Connell, 2025). But if the older "middle-paleolithic" substrate is <10% of the modern Sahul region genomes, it would not give much difference in N. admixture, given that the latter is around just 2%. The latter, upper-paleolithic wave around 45 kY, likely coming from Siberia through China, of course, replaced all uniparental lines of early Sahul people, but some of the early genetic signal still remained in autosomes.
Another anomaly is very robust crania of some of the aboriginal fossils and even some modern aborigenes. I think there is more to this story than just a Denisovan introgression.
There is a consistent data for early presence of modern anatomy in the region (Tam Pa Ling, Lida Ajer, maybe Callao toe phalange which is very consistent with HS, just small), plus Chinese fossils of 120-80 kY (Zhirendong, Fuyan, etc). The early art in Sulavesi is consistent with this picture. Early humans might have a wide Eurasian expansion during Eemian interglacial and MIS 5, and a small level of this genetic signal might remained in EA, SEA and especially Sahul region.
I appreciate your perspective and also the rationale for thinking an early out-of-Africa dispersal may explain some of these data from East and Southeast Asia. I would add in favor of this hypothesis that the mtDNA chronology can be made partially consistent with the "long chronology" by choosing a slow mutation rate, if not across the whole tree then at least for deep branches. I don't rule out the possibility that the early Sahul habitation may have been
There are two directions to look at the authorship of these markings, one from the perspective of evidence of modern human presence in the region, and the other from the perspective of Denisovans. From the modern perspective, the site that influences me the most is Tam Pà Ling. I'm less persuaded by the Chinese sites and Lida Ajer for reasons that others have published; Niah now appears to show a break in the record. Interestingly the argument for arrival of an intrusive population in the region is reasonable from the anatomical/skeletal evidence but is very weak from the archaeological evidence. There are no artifacts at all from Tam Pà Ling and actually none of the sites with modern-looking skeletal remains of early date have any substantial artifact record. That's weird.
From the Denisovan side, the behavioral record has been unclear up to now. But the identification of Harbin skull makes it clearer that the later Middle Pleistocene record of China was Denisovan. Pigment use is less-well attested in East Asia than in Europe. But it is known at several sites, and at Lingjing is combined with engraved lines on bone. Handprints in travertine at Quesang also provide a parallel for the Neanderthal finger fluting sites, all showing that marking was a common behavior.
The difference in the archaeological datasets from Europe and East Asia across the period from 100,000 to 50,000 years is striking, and probably mostly due to biases of exploration and excavation. It's changing, and it's fascinating that the early progress in that change is coming from places like Sulawesi and Laos!
The Denisovan hypothesis should not be ruled out, of course, and there are supporters of it.
You know, of course, that some Russian archaeologists who worked in Denisova cave tend to attribute the earliest art objects from there to Denisovans. I am somewhat skeptical of this for the following reason - if Denisovans could make a leap to UP-type modernity, why had they disappeared so quickly and left so small a genetic imprint (except in Sahul)? We should have seen more of them and their influence. So I lean toward HS who are proven to make the UP transition. For similar reason I lean towards early HS as creators of SEA art. Denisovans were very successful in their adaptations to difficult and diverse Asian environment for hundreds of kY, but I think only HS were able to make transition to true modernity.
Great insight! Would the continuity of material culture from 130,000 to 40,000 years ago disfavor a hypothesis that the art was made by an earlier Out of Africa, or pre-OOA, homo sapien population?
Good question. I don't discount the possibility of an early dispersal from Africa, which might help explain some of record in China, Southeast Asia, and Sumatra of fossils that resemble modern humans. On the other hand, maybe some of those sites are late Denisovans, and others may be misdated. DNA evidence from a site with strong geochronology like Tam Pà Ling might be the only way we will know.
How far into island Southeast Asia could an early out-of-Africa population have gone? Is it possible that an early out-of-Africa population might be the first inhabitants of Australia? Were they on Sulawesi? Again, nothing precludes this possibility, but the data put some constraints on it. The DNA data from living people suggests that DNA from an early out-of-Africa group did not survive in any measurable fraction up to the present day. So if such a group existed, there must have been a near-total replacement by later people, replacement to a higher degree than Denisovans experienced.
There are several sites that I think provide some information about this. The Leang Bulu Bettue site with its consistent record from maybe as early as 200,000 up to 40,000 years ago is one of two deep chronologies. I think your observation is the correct one: An early out-of-Africa group should have shown up sometime after 120,000 years or so, and there's nothing in the record to suggest it. However, an important caveat is the sparse nature of all artifacts from these early parts of Leang Bulu Bettue. I'm not positive you'd notice a small difference with the small number of artifacts.
The other deep record is Liang Bua on Flores. That one is pretty unequivocally a case where the ancestral island population was present as late as 65,000 years ago, and was not an early out-of-Africa group. I'm waiting for more updates from Liang Bua on the post-floresiensis chronology. Trader's Cave on Borneo is starting to look like a deep record may emerge, and the one tooth may suggest this also was not an early out-of-Africa group at around 50,000 years ago. And then there is the Ngandong series of fossils, between 120,000 and 106,000 years old, also not an early out-of-Africa group.
So I don't rule out the possibility that an early out-of-Africa dispersal was part of the mix in this region but none have been found, and there seems to be at least some evidence that longer-term inhabitants were present and making archaeology.
The reason I bring up Denisovans is mainly inspired by genetics of present-day people. I take as a good chance that the Denisovans themselves had mastered seafaring to a degree that they could reach Sahul. I also think there's a good chance that the morphology of the Ngandong crania may be Denisovan rather than Homo erectus. But that, too, is not proved.
Much remains up in the air. Maybe the Madjedbebe, Australia, early date is incorrect, so the idea of an early habitation is not earlier than around 50,000 years ago. Maybe the Denisovan genes in today's people trace to different mainland Southeast Asia groups of Denisovans. And maybe the Sulawesi hominins were another case like floresiensis that was highly endemic. It may take a while to sort this all out.
Very interesting read!
Still I tend to think that early HS were in Sahul region well before 40 kY or even 45 kY - probably by 70-72 kY, and Sahul settlement was of two stages, consistent with "long chronology". There are too many anomalies with Austr. aboriginals and PNG genetics. First, they show a very substantial difference in coalescence time with Africans compared to all other Eurasians, even after filtering out Denisovan admixture, and attempts to explain it otherwise are not convincing. They are also huge outliers in PCA whole-genome space (especially if you look at higher components - 4,5,6, not just usual 1,2). The negation of the "long chronology" is largely based on the similar level of N. admixture in Austr/PNG to all other Eurasians (e.g. Allen & O'Connell, 2025). But if the older "middle-paleolithic" substrate is <10% of the modern Sahul region genomes, it would not give much difference in N. admixture, given that the latter is around just 2%. The latter, upper-paleolithic wave around 45 kY, likely coming from Siberia through China, of course, replaced all uniparental lines of early Sahul people, but some of the early genetic signal still remained in autosomes.
Another anomaly is very robust crania of some of the aboriginal fossils and even some modern aborigenes. I think there is more to this story than just a Denisovan introgression.
There is a consistent data for early presence of modern anatomy in the region (Tam Pa Ling, Lida Ajer, maybe Callao toe phalange which is very consistent with HS, just small), plus Chinese fossils of 120-80 kY (Zhirendong, Fuyan, etc). The early art in Sulavesi is consistent with this picture. Early humans might have a wide Eurasian expansion during Eemian interglacial and MIS 5, and a small level of this genetic signal might remained in EA, SEA and especially Sahul region.
I appreciate your perspective and also the rationale for thinking an early out-of-Africa dispersal may explain some of these data from East and Southeast Asia. I would add in favor of this hypothesis that the mtDNA chronology can be made partially consistent with the "long chronology" by choosing a slow mutation rate, if not across the whole tree then at least for deep branches. I don't rule out the possibility that the early Sahul habitation may have been
There are two directions to look at the authorship of these markings, one from the perspective of evidence of modern human presence in the region, and the other from the perspective of Denisovans. From the modern perspective, the site that influences me the most is Tam Pà Ling. I'm less persuaded by the Chinese sites and Lida Ajer for reasons that others have published; Niah now appears to show a break in the record. Interestingly the argument for arrival of an intrusive population in the region is reasonable from the anatomical/skeletal evidence but is very weak from the archaeological evidence. There are no artifacts at all from Tam Pà Ling and actually none of the sites with modern-looking skeletal remains of early date have any substantial artifact record. That's weird.
From the Denisovan side, the behavioral record has been unclear up to now. But the identification of Harbin skull makes it clearer that the later Middle Pleistocene record of China was Denisovan. Pigment use is less-well attested in East Asia than in Europe. But it is known at several sites, and at Lingjing is combined with engraved lines on bone. Handprints in travertine at Quesang also provide a parallel for the Neanderthal finger fluting sites, all showing that marking was a common behavior.
The difference in the archaeological datasets from Europe and East Asia across the period from 100,000 to 50,000 years is striking, and probably mostly due to biases of exploration and excavation. It's changing, and it's fascinating that the early progress in that change is coming from places like Sulawesi and Laos!
The Denisovan hypothesis should not be ruled out, of course, and there are supporters of it.
You know, of course, that some Russian archaeologists who worked in Denisova cave tend to attribute the earliest art objects from there to Denisovans. I am somewhat skeptical of this for the following reason - if Denisovans could make a leap to UP-type modernity, why had they disappeared so quickly and left so small a genetic imprint (except in Sahul)? We should have seen more of them and their influence. So I lean toward HS who are proven to make the UP transition. For similar reason I lean towards early HS as creators of SEA art. Denisovans were very successful in their adaptations to difficult and diverse Asian environment for hundreds of kY, but I think only HS were able to make transition to true modernity.
Fascinating! Thanks for the reply