The Cave of Homo naledi, or A Textbook Example of How to Do Science
“How often have I said to you that when you have eliminated the impossible, whatever remains, however improbable, must be the truth?” —Sherlock Holmes
Recently, a magnificent new hominin fossil was announced: Homo naledi, known from a single chamber in an almost-inaccessible cave in South Africa. NCSE’s Stephanie Keep nicely explains the significance of this “weird as hell” fossil in posts here, here, and here.
Why is it so weird? In short, the problem is this: H. naledi had a very small, orange-sized brain, but there is strong evidence that it went to great efforts—extremely difficult movement of corpses perhaps requiring group cooperation and the use of fire as a light source—to place its dead in this almost-inaccessible chamber. We don’t see obvious signs of burial until the relatively big-brained Neanderthals appear on the scene. The idea of something with such a small brain conceiving of and executing the complex behaviors required to repeatedly dispose of their dead in a super-hard-to-get-to cave blows open a lot of our assumptions about cognition. I think the scientific term that best describes this is “cray-cray.”
One of the set of papers released as part of the announcement of this fossil involved the geology of the cave. Dirk et al.’s “Geological and taphonomic context for the new hominin species Homo naledi from the Dinaledi Chamber, South Africa” provides a brilliant description and explanation of the cave environment in which at least fifteen individual Homo naledi found themselves, however that happened. It also explains what happened to the bodies once they were in there (that’s the taphonomy part). But perhaps best of all, the paper provides a fabulous example of how scientists think about a difficult problem. Teachers could base an entire lesson plan on this one paper.
When I heard initial reports about this fossil find, I was shaving and I was so startled by what I heard that I cut myself, exclaiming, “Occam!” Occam’s razor is a good first step in thinking about scientific claims; the elevator-pitch explanation of Occam’s razor is that given two possible explanations, the simpler one is probably the right one. So when the initial news reports came out, I anticipated that very soon someone would find a flaw and blow open the unlikely, improbable, seemingly-impossible claim that Homo naledi had been put its dead in the cave deliberately.
After all, in 2010 when NASA announced the discovery of bacteria using arsenic in its DNA, it seemed to take only days before people identified severe methodological problems with this research. (Turns out the bacteria, like every other known living thing, use phosphorus, not arsenic.) Surely the same thing would happen here with H. naledi. It hasn’t yet.
One of the keys to the strangeness of the H. naledi find is that there seems to be only one entrance to the cave. Surely if there were another, easier entrance that would really change the significance of this find. So as I ruminated on this, I thought:
1. There’s got to be another entrance, perhaps on the cave ceiling, meaning that Homo naledi could have dropped into this isolated spot; maybe subsequent carbonate growth (flowstone) covered and hid this entrance
2. If there was another cave entrance, then there should be other critters in the cave deposits, not just H. naledi
3. If there was an easier cave entrance, then there should also be outside debris—rocks different from the cave rocks, twigs, and leaves from local plants, etc.
If any of those three things were true, I reasoned, then there could be another explanation for how H. naledi got into this spot. So then I read the Dirk et al. paper I was pleased to find that each of these considerations had been addressed—and refuted.
1. Other entrance
My initial assumption that the team had not thoroughly addressed the issue of a second entrance, perhaps on the ceiling, perhaps now covered with carbonate deposits, was discussed in the paper this way:
An exhaustive search by a professional caving team and researchers has failed to find any other plausible access points into the Dinaledi Chamber, and there is no evidence to suggest that an older, now sealed, entrance to the chamber ever existed. Furthermore, detailed surface mapping of the landscape overlying the Rising Star cave system illustrates that no large flowstone-filled fractures occur in the region above the Dinaledi Chamber.
Okay, so strike one for my skepticism. The researchers seem to have anticipated this problem and done due diligence—looking not only inside but outside and above—for signs of another way into this remote cave.
2. Other critters
If there was a second entrance, or if the cave was in some way more accessible in the past, then one would expect to find significant evidence of animals other than H. naledi. All manner of animal—rodents, bats, birds—take up residence in accessible caves. Here’s how this paper addressed this issue:
The lack of other contemporaneous fauna in the assemblage, and complete lack of surface modifications by vertebrates (carnivores, scavengers or rodents) further suggests that the Dinaledi Chamber remained undisturbed by other animals, which could not reach the chamber.
In all, six birds and several rodents were found in cave, but this is hardly the wealth of animals fossils one would expect if this cave were easily open to the outside. Moreover, the lack of disturbance suggests that the birds and rodents were rare; certainly if there was a food source lying on the cave bottom, we should expect heavy disturbance of scavengers. Instead we find many fossils that are well articulated.
The lack of disturbance touches on another possibility: maybe other H. naledi did not move these remains into the cave. Could large predators have dragged them in? Here again the researchers have anticipated the argument:
Nor have we found any trace of carnivore remains or the remains of other likely prey animals. Thus, the predator would have had to select a single prey species—H. naledi—carrying into the chamber all age and size categories (Berger et al., 2015) without leaving a trace of its own presence. We consider this very unlikely.
Strike two for my skepticism.
3. Outside debris
Sediment inside a cave can tell you a lot about what’s going on in the cave—is there regular traffic from outside? Does water flood the cave, perhaps transporting and trapping individuals inside the cave? Most cave deposits have measurable debris from the outside—rocks, branches, leaves, or litter from modern humans. Nothing of the sort was found.
Analysis of the sediment shows it to be fragments of the surrounding rock—and little else. No leaves, no branches, no anomalous pebbles carried in on the toe of some scavenger. It’s as if the cave has been sealed from the outside except for a few enterprising H. naledi. To have a cave so cut off is unusual:
The sedimentary deposits in the Rising Star cave system that host the remains of the new hominin species of Homo naledi (Berger et al., 2015) are anomalous when compared to all other deposits of hominin remains in the Cradle of Humankind in a number of significant ways.
The presence of clay in the sediment suggest periodic flooding, leading to the idea that maybe these fossils were somehow drowned in this cave system and swept into this remote location. But even here the researchers have anticipated this idea:
...the remains of H. naledi could have accumulated as a result of a catastrophic event during which a large group of animals was trapped in the cave. This could have happened either during a single event when a large number of hominin individuals were in the chamber, or in a death trap scenario over a period of time as individuals repeatedly entered the Dinaledi Chamber and died. Either scenario would have to explain why the animals chose to penetrate this deep into the cave, into the dark zone, moving away from all entrance points into the cave system. The sedimentological evidence presented suggests that accumulation of the fossils occurred over a period of time during deposition and reworking of Units 2, and 3, which refutes a single event hypothesis. Apart from this, and noting that the assemblage recovered to date represents only a small part of the total fossil content in the chamber, the sheer number of remains encountered in the Dinaledi Chamber, is hard to explain as the result of a single calamity.
Strike three. After reading this, I’m impressed and convinced.
The work on H. naledi is amazing science, and this section on geology and taphonomy by Dirk et al. is an excellent example of how scientists think. There are a whole bunch of possible alternate explanations, and they get whittled down one-by-one, until the only thing left is the improbable, counter-intuitive conclusion that H. naledi individuals were deliberately dropped into this cave. Dirk et al. have addressed virtually every other explanation and demonstrated why they don't fit with the evidence.
One thing they found particularly intrigued me. Among all of the fossils, they found old survey pegs left behind in this chamber, and evidence that some of the fossils on the surface had been moved. Apparently, the cavers that discovered the chamber were not the first ones to have stumbled upon it. Yet until quite recently, no one knew this cave existed; whoever left those survey pegs did not recognize the importance of this find and didn’t bother to note it on a map.
So the history of this cave, like the history of the fossils it contains, shows us another important aspect of scientific discovery: sometimes important findings are seen, but not understood. Science is far from finished; latent discoveries abound, waiting for new eyes. We do not find the revolutionary H. naledi except in this one cave—what else are we missing?