There is no new evidence that Gunung Padang is a 24000 year old pyramid

In fall 2022, an episode of Graham Hancock’s Netflix series Ancient Apocalypse heavily featured the impressive archaeological site of Gunung Padang in Indonesia. Gunung Padang is a hilltop site with several terraces and visually striking arrangements of the amazing naturally occurring columnar igneous rock that geology has blessed us with in many places on Earth. 

Figure 2 from Natawidjaja et al. showing Gunung Padang

Based on the work of Indonesian researcher Danny Hillman Natawidjaja, Hancock claimed that the site was in fact not a natural volcanic hill, but a pyramid built by an ancient lost civilization, perhaps as much as twenty five thousand years ago. Other archaeologists who work in the region, including Indonesian archaeologists, consider the terraces at Gunung Padang to be about 2000 years old. I wrote an initial critique of the claims about Gunung Padang at the time.

Now, as reported in many news and SciComm outlets, Natawidjaja et al. (2023) have published a follow up article in the journal Archaeological Prospection, claiming to demonstrate that Gunung Padang is indeed the oldest pyramid in the world, as much as twenty thousand years older than the pyramids of Egypt.

I see a massive disconnect between the claims made, especially in the conclusion, and the evidence presented in the body of this latest article. The title of the article, and the resulting media headlines do not match the evidence.

It’s concerning that major media outlets, including Canada’s National Post, have uncritically repeated these claims without seeking input from a range of researchers in archaeology and geology, who could point out some of the serious shortcomings of this new article.

Sculpted pyramid core, or natural andesite outcrop?

The authors claim that “The pyramid’s core consists of meticulously sculpted massive andesite lava (Unit 4), enveloped by layers of rock constructions (Unit 3, Unit 2 and Unit 1).” This core, according to them, is the oldest component of the pyramid. In other words, they claim that around twenty four thousand years ago, people sculpted a naturally occurring volcanic hill into a pyramid.

The first issue with this claim is that no evidence is presented of any sculpting of the stone core of the hill. The authors don’t present tool marks, and they don’t claim to have recovered any archaeological tools from the site. If I am told of “meticulous sculpting”, I am expecting it to be obvious even visually, but I would expect it to be measurable in some way.

Google Maps shows that Gunung Padang (circled in red) is surrounded by similar geological features.

The second problem is that this kind of andesite outcrop is common in the region. Gunung Padang is surrounded by similar formations. A quick glance at the region in Google maps confirms that Gunung Padang is not an isolated phenomenon. Natawidjaja et al. make no effort to tell us why Gunung Padang is different from any of the surrounding andesite outcrops. 

Presumably, the sculpting into a pyramid would have made it look different and remarkable. If so, we are not told in the latest article. Until there is a clear demonstration that Gunung Padang is not a natural volcanic hill in a region covered with volcanic hills, it is reasonable to think that it might be.

Buried, yes. But by humans, or by natural sedimentation?

Natawidjaja et al. repeatedly claim that different phases of the pyramid’s construction were deliberately buried (e.g. “Around 7900– 6100 BCE, Unit 3 was deliberately buried with substantial soil fills”). They present these claims as evidence that people, over the past 24k years, were gradually building up the hill into a pyramid.

However, the article opens with an acknowledgement of the natural forces that “bury” both archaeological and non-archaeological features in the region: “Indonesia’s tropical climate, characterized by intense weathering and sedimentation processes, combined with dense vegetation, has led to the burial and concealment of ancient cultural remains.” And again: “These dynamic natural processes have caused the disappearance of numerous ancient heritage sites in forests, underwater and buried underground.”

The authors are clearly very aware that there are many different processes that can lead to the apparent “burial” of both natural and human-made features over time, but they make no effort in the article to distinguish between weathering and sedimentation on the one hand, and human intervention on the other. If these non-human processes are so active in the region, and if they can result in burying sites, why must humans be involved in the “burying” parts of the Gunun Padang “pyramid?” Perhaps the authors have an answer to this question. If so, it doesn’t seem to be in the article. At least, I didn’t find it.

Is Gunung Padang 24k years old?

Natwidjaja et al. repeatedly claim that Gunung Padang is a pyramid whose first phase of construction goes back 24k years. They do present radiocarbon dates taken on organic fill material from the lower layers of the hill through coring. However, they present no evidence that this organic material is there as a result of human activity. 

Soils have an organic component that can be dated, such as remains of leaves, twigs, insects, rodent bones, etc. A radiocarbon date does not an archaeological site make. In order for a site to be archaeological, and therefore to be the result of human activity, the material dated must be related to human activity. The authors here present no evidence that the soil they are dating is related to human activity.

In summary…

Gunung Padang is an impressive archaeological site. It doesn’t need to be a 24k year old pyramid to be impressive. In order to reconcile the claims made in the article with the evidence presented, I would need to see at the very least 1) evidence that the andesite core of the hill is sculpted or worked, 2) evidence that the hill is structurally different from the surrounding ones, and 3) evidence that the organic material that was carbon dated is related to human activity, and not to environmental accumulation.

Postscript: a word about the peer review process

This article appeared in a peer reviewed journal put out by one of the major academic publishers. What I wrote above is essentially what I would have written to the editor of the journal, had I been asked to review this paper before publication.

Not being a specialist of the region, or even particularly knowledgeable about its archaeological record, I focus on the relationship between the claims and the evidence presented to support them. I am sure a proper specialist could find a lot of other stuff to say about how the claims and the evidence relate to what we know about the region and its past.

Articles like this one (and like the original Cerutti Mastodon paper, or the now retracted Hopewell cosmic impact paper) highlights for me the weakness of our current system of narrow pre-publication peer review by a few scholars, versus the strength of broad, post-publication review by the entire community.

The current system has trained the media to run uncritically with claims that appear in journals, to the extent that they often don’t even consult with anyone not involved in a project before repeating its claims. It’s in a journal, so it must be reliable, right?

The speed at which these unsupported (as opposed to false or impossible) claims spread through the media is stunning. As archaeologists, we have to be very active, and very responsive. Beyond the short-term effort of helping the public understand the limits of these claims, we have to participate in the long-term effort to contextualize them when they are used in support of pseudoarchaeological or other pseudoscientific narratives.