26

It was the first day of the workshop, and I was briefing the group. We had pulled together a great mix of early career and more senior scientists, all of whom had come together in Johannesburg to study the Rising Star assemblage of fossils. Our deputy vice chancellor of research had just welcomed the group to Wits, and now I was introducing the goals of the workshop.

“We are going to describe these fossils without a date,” I told them.

A laboratory full of scientists stared back at me. A few probably thought I was joking, but most had looks of disbelief. Ever since the application of more sophisticated dating methods at Olduvai Gorge, almost every discovery of a fossil hominin in Africa had been accompanied by some kind of date. For many scientists, geological age had become a kind of insurance policy. After all, the lack of a clear idea of the geological age of the Taung Child had been part of why Raymond Dart had faced early criticism. In the popular mind, and in the minds of many paleo​anthropo​logists, the geological age of a fossil is the key to interpreting its place in human evolution. I was telling a roomful of scientists that we were going to blindfold ourselves to that key to understanding.

In part, this decision came from necessity. Our geological team, led by Paul Dirks, had been involved with interpreting the site from the start. It was clear to everyone that it would be difficult to work out the age of the fossils in the 101 Chamber.

At Malapa, we had enjoyed a run of geological good luck. The fossils there had been sandwiched neatly between flowstone layers, and we were able to use that material to date the skeletons precisely. The nonhuman animal bones (which included saber-toothed cats, hyenas, antelopes, and a horse) gave evidence that confirmed the age of the site as determined by the flowstones.

No such luck at Rising Star. To begin with, there were only the bird bones and rodent fragments, and none of these seemed to be preserved in the same way as the hominin fossils. Unlike at Malapa, these bones were not embedded in a hard breccia with flowstone draped over the top or bottom—they were in a mass of soft, loose sediment. Bits of flowstone had eroded against the wall, but it would be hard to determine how these formations related to the fossils. Still, we tried to determine the flowstone ages using the same dating method that had worked at Malapa, but here it failed. The thin, eroded flowstones had been contaminated to some degree by sediments in the Chamber, and we had no way of knowing how long the fossils had been in the cave before the flowstone formed. Only the bottom of the Chamber floor could give us a way to use geological testing to figure out the maximum age of the fossils.

All in all, it was a problem. One day, I had spoken over Skype with Paul and his young colleague, Eric Roberts. “You are going to have to sacrifice some material to get a date,” Paul told me.

“I can’t do that until it’s described,” I replied, “and even then, you know there is no direct dating method that’s likely to give us a date on this material—it’s probably too old.”

The logical method to try was electron spin resonance (ESR) dating, a method that uses the atomic structure of crystals formed as teeth fossilize to gauge how long a tooth has been buried. But there were two problems. First, the geologist has to know the radiation levels of the environment where the teeth were buried, and these measurements can be misleading. Worse, this analysis requires that a sample of enamel be removed from the tooth. Usually, that’s not a problem, because we can sample teeth from nonhominin animals found at the site. Dating an antelope tooth is just as good as dating a hominin tooth as long as they clearly come from the same geological situation. But the 101 Chamber had no teeth from fauna—only hominins. To date them, we would have to drill into some of these precious teeth—not an option. None of us knew what these hominins were, and destroying any of these fossils before describing them would be the wrong scientific choice.

THE MORE I THOUGHT about it, the more I realized that what seemed like a barrier was actually an opportunity. There was a real advantage to analyzing these fossils without dating them. We were about to study this set of extraordinary fossils. Our assessment of them would depend on their anatomy alone. They might be very similar to some other collections of fossils and very different from others. Some of those similarities would be shared with very distant common ancestors—so-called “primitive” traits—while others would be unique and distinctive features—“derived” traits. The pattern of traits would tell us their relationships—not the age of the fossils. If our collection of fossils from the 101 Chamber shared many derived traits with one particular species of hominin, we might assign them to that species. If we could not show such a pattern with any existing collection, we might name a new species. These relationships had nothing whatsoever to do with the fossils’ age. So the fact that we would not know their age was not a drawback—it was an advantage!

Our team had begun to learn this lesson with sediba. The fossil skeletons that we had found at Malapa were possibly the best dated fossils in all of Africa. We had imagined that the highly precise dating of the site was a scientific triumph. In some ways it was, adding much more information to our knowledge of the time line of sites in the Cradle. Still, that date had also caused us some unexpected trouble.

When we described sediba for the first time, we emphasized that its anatomy was a mosaic—a combination of traits, some like Homo, others like Australopithecus or other primitive hominins. We thought that this species did not belong to Homo, but we could not deny that some comparisons placed it as a close relative. The combination of features made sediba look like what we might expect in an ancestor of Homo. It connected the primitive body plan of australopiths with more humanlike features of our genus. The mosaic of features suggested that sediba might be an ancestor of humans.

This seemed fairly innocuous to our team, but some other scientists asserted that it was impossible for sediba to be the ancestor of Homo. The fossils were not old enough. They pointed to a fossil found in Ethiopia that seemed to come from a rock layer 2.33 million years old. This was a mere fragment of a jaw, but its finders supported the idea that this was the earliest specimen of Homo ever found. The Malapa skeletons were only 1.977 million years old. It was obvious, some scientists claimed: sediba was simply too young to have been the ancestor of Homo. This was exactly the same logic by which Arthur Keith argued against the importance of the Taung Child back in 1925. If the fossil was not old enough, it could have no place in the ancestry of humans.

This logic was based on the assumption, though, that the Malapa fossils represented the earliest possible evidence of their species. Paleontology doesn’t work that way. We found some individuals of sediba at Malapa, but other members of sediba almost certainly lived much earlier—how much earlier, no one could say. But in paleontology, the individuals we find as fossils are literally fewer than one in a million of the individuals that ever existed. Did sediba exist early enough to give rise to a 2.33-million-year-old fossil? The fossil record could not answer that question either way.

To the sediba team this argument about dates was beside the point. The fossil skeletons from Malapa were the best evidence from any early humanlike australopith species. If we were going to learn how our ancestors took the path toward humanity, we needed to use evidence from this branch of the family tree. Whether sediba itself gave rise to Homo, or whether instead sediba and early Homo both emerged from some other species that was their shared ancestor, the fossils provided the best clues we had to know how our evolution had happened. Dates didn’t necessarily help us to understand the relationships of fossils. In the case of sediba, the dates were getting in the way.

So I was eager to see the team study these new fossils without any preconceptions about their age, and I convinced almost everyone of the advantages of this approach. Some workshop scientists who hadn’t been involved with the sediba project did raise their eyebrows at first. But as they encountered the fossils, they found that not knowing their age gave them the freedom to explore ideas they might never have considered otherwise. If the anatomy of these fossils did not tell us how they were connected to our family tree, no single number representing their age would ever help. On the other hand, if their anatomy gave us a fairly clear picture of relationships, the age of the fossils, when we determined it, might serve to test some of the most cherished assumptions held by scientists in the field.

That would be worth seeing.