Rising Star Burials Revised
Longtime readers will remember my obsession with the discoveries in the Rising Star Cave in South Africa, where thousands of bones of a hominin dubbed Homo naledi have been recovered over the past dozen or so years. By itself, that's not terribly shocking (there are lots of hominin fossils), but various features of the cave where they were found led researchers to suggest the bones were deliberately placed there. More recently, there have been a series of claims and pre-prints that have introduced some fascinating new dimensions to the work (fire, burials, tools, and engravings), but there has also been criticism of these findings that reflect some of my own misgivings about a few of the claims. Some of my concerns:
- The claims of fire evidence remains undocumented, and it's been nearly two and a half years now. I know in archaeology that it's somewhat common to make big announcements about discoveries with the research articles coming much later, but I'm really not a fan of publication by press release. In my view, extraordinary claims need to be backed by documentation of the research. I feel pretty strongly about that, and my opinions on this matter come from my history in creationism where extraordinary claims turn out to be largely baseless once the actual research is conducted. We all need to stop doing that.
- The alleged engravings have nothing that link them directly to Homo naledi. Not sure what else to say about that. Some researchers express skepticism that they're even engravings, but I'm not experienced enough to make a judgment on that subject. But I can see that other than spatial proximity to Homo naledi remains, there's nothing that really connects the markings to Homo naledi.
- The evidence of burial was just not clear to me. The scans show a jumble of bones, which the researchers assure us are either partially articulated or spatially arranged in a way that indicates anatomical connection. But it's not like there are bodies or even partial bodies that look like a more typical sapiens or Neandertal burial. The documentation was also not well labeled, so it was very difficult to verify their narrative of how the skeletal elements were spatially arranged. The evidence of disturbed sediments containing the bones adjacent to areas with few or no bones was quite limited in extent. I'm left very uncertain about the whole thing. If it is a burial, it looks like the bodies were either disturbed after burial or broken into pieces before burial.
- In one of the purported burials, researchers reported a "tool-shaped rock" near a skeletal hand, and they infer from high resolution scans that the rock had been knapped and used as a (presumably) cutting tool. But that rock remains encased in a field jacket, and it's only accessible by high resolution scans.
So that's where things stood when last I blogged about Rising Star Cave research seventeen months ago. Recently, Lee Berger, the lead researcher on Homo naledi, posted some updates on this research. The revised paper describing the engravings is not very different from the first version, and I won't comment much more on that here. They do address the idea that these could be weird erosional marks, but otherwise continue to insist that these are engravings that date to 241-335 ka, despite presenting no evidence of dating of the engravings.
The next paper was a new pre-print by van Rooyen and colleagues describing detailed scanning of the chute system that leads from the Dragon's Back antechamber into the Dinaledi subsystem where the Homo naledi bones are. I don't have much to say about that one. It reveals the complexity of the passage into the Dinaledi subsystem, but it does not confirm any other routes in.
Finally, eLife released a massive update to the original burial paper. Before I share my thoughts on this paper, it's helpful to remember that eLife has adopted a different publication model than a typical journal. The standard for publication is peer review that's overseen by an editor, where the reviewers recommend either accepting, revising, or rejecting a submitted paper. Most often, decent papers will go through revisions before appearing in print. eLife has adopted a public peer review style, where the decision to publish is not in the hands of the reviewer. Instead, editors make their own judgments on publication, then reviews are solicited and released along side the submitted manuscript. Authors can then revise their paper if they like, or not. You can imagine the peer review system is ripe for abuse, and I can see how the eLife strategy tries to address that (like preventing papers from being torpedoed by cranky reviewers or allowing reviewers to be just abusive in their reviews). This is the process I'm commenting on today. The initial burial paper was shorter and left me with a lot of questions. The published reviews were critical (although I would argue not in a way that would be helpful to improving the work). This new paper is a revised version of that original burial paper.
So how did they do? Well, it's a pretty big improvement in my judgment, and it's quite a lot longer and more detailed with new data. The main improvements in my estimation are:
- Better figures that better illustrate their findings. Multiple figures, like figure 6, attempt to illustrate how skeletal elements were arranged in the sediments.
- Additional excavation gives a much more complicated picture of the deposits in the Dinaledi subsystem. More about this below.
- A more careful consideration of possible deposition scenarios and how the evidence in the cave coheres or not with various scenarios. For example, they show evidence that rules out the idea that the bodies were dropped down the Chute and spread out from there. They really seemed to be carefully considering how different body accumulation ideas might cohere with the evidence in the Dinaledi subsystem.
- Just more information overall. I like information. It helps to evaluate what's going on, so more is better in my view.
Back to the actual deposits in the Dinaledi subsystem. To put this in context, the original publications nearly a decade ago left me with the impression that the floor of the Dinaledi Chamber was littered with bones and their excavation of one small area confirmed that there were more bones to be found under the surface. I'm not sure if this was their intention, but I assumed that wherever you dug in the Dinaledi Chamber, you'd find remains of Homo naledi. That turns out to be wrong. They describe both lateral and vertical extensions of their digging that shows that the Homo naledi bones were always spatially segregated. That means that the section they excavated 12 years ago happened to be a section where bones were found, but adjacent sections had comparatively few or no bones. In other words, the bones - all the bones - are restricted to particular locations. As I said, previously I thought there were bones throughout the Dinaledi Chamber, which made the identification of these spatially limited "burials" stand out as something unusual. Now it's apparent that spatially limited deposits of bone are the normal condition for all Homo naledi bones in the Dinaledi subsystem. I think that's important because it's more parsimonious to explain just one mode of bone deposition in spatially limited areas rather than two different accumulations, either uniformly in the Dinaledi subsystem or spatially limited.
But are they burials? Reconsidering all the bone deposits in this way leads to some very strange observations. First, there are definitely articulated and partially articulated skeletal elements all through the subsystem. Second, there are elements oriented in vertical positions that would be impossible to attain had bodies been left on the surface. Those elements must have been supported by sediment as the bodies decomposed. Third, the articulation is largely partial. There are elements that clearly represent an arm, a leg, or a skull, but there aren't full clear bodies laid out in supine or fetal position like we might expect from certain sapiens burials. Fourth, many of the skeletal elements appear to have post-mortem breakage including slight displacements, as if those bones were disturbed somehow. Considered together, these evidences seem to imply that bodies were deposited in graves, and then later those same graves were re-excavated, disturbing the original bones, and additional bodies were added. Over time, then, the deposits become more dense with bone but also more jumbled, all the while maintaining partial articulation of some skeletal parts.
They also mention once again in multiple places the presence of baboons, including a buried tooth and a few skeletal elements from a surface discovery. Part of their argument for intentional body deposition in the Dinaledi subsystem comes from the unusual monospecific accumulation: There's only Homo naledi bones down there, except for a tiny amount of other bones. That works well enough for 99.9% of the remains in the Dinaledi subsystem, but what about these baboons? I'm not sure what to think about that.
Do these observations add up to a solid case for deliberate burial? It seems plausible to me, but I'm curious to see how other anthropologists will take this. I thought the original critiques (which I discussed here) raised some interesting points, but others have been more curmudgeonly. A lot of critiques are not serious attempts to explain all the data. They are ad hoc strategies to explain away some of the observations because of skepticism that such a diminutive hominin could be doing such complex behaviors. They might claim that there's some other way into the chamber or that animals sometimes just get lost in caves, all of which just don't address the full body of data as we know it. The fact that these bone deposits are so unlike graves as we know them makes them easy targets.
In a blog post, naledi researcher John Hawks (and co-author on all three of these papers) suggests that this open approach to peer review is a desirable thing, that this encourages progress in the field and enhances public understanding of science. I tend to agree with him here. I like generally quick publication of results. I like making the public more aware of how research works, and I like making it more difficult for notorious "Reviewer #2" to effectively block publication of new findings. I'm not sure about this particular case though. The burial, engraving, and tool claims sure looked like they were rushed to publications in order to support a Netflix documentary and accompanying book (admittedly that's a pretty cynical read, but there it is). When (from my perspective) serious objections were raised, the research team was essentially silent for almost 18 months. It didn't help that in that period, the Twitter takeover occurred, and most research anthropologists stopped posting there. It kind of looked like the naledi team published some extraordinary claims, took sharp criticism, then withdrew from public view. I understand that revisions of the magnitude of this burial paper take time to research and write, and I definitely appreciate and acknowledge the outcome of that work. But if research is to be open, the public needs to see that revision process as well or at least some hint that it's happening, if for no other reason than to steal the thunder of the cynics who saw the burial claims as a crass publicity stunt. Openness is good. Be more open.
That's it for now. We're supposed to be getting more Homo naledi papers in the near future, including the fire paper we've been waiting 2+ years for. I hope that will give us at least some reason to connect the fire remains to Homo naledi. Time will tell!
Feedback? Email me at toddcharleswood [at] gmail [dot] com. If you enjoyed this article, please consider a contribution to Core Academy of Science. Thank you.
Have you read my book? You should check that out too!
