Hominins: Was I wrong????
 |
| Me and Lu at the Cleveland Museum (She still isn't human.) |
Back in July, Answers Research Journal published a significant challenge to statistical baraminology, certainly the most knowledgeable I had ever seen. When I first looked closely at the papers, I saw enough simple errors to give me pause, but I confess that I was still a little nervous. What if I had gotten the hominin story wrong for an entire decade? If so, it would have been an honest mistake but still very embarrassing. And I knew there would be a lot of people who would be gleeful to see me end up publicly shamed like that.
As I described previously, my response focused entirely on creatures other than hominins, and I only seriously looked at hominins once that paper was in print. Today, my new analysis of hominin fossils has been published in ARJ. It's co-authored with a student collaborator who did a lot of the "grunt work" and helped me write the paper. Our intention was only to re-analyze craniodental characters, which means attributes of the skull and teeth. We wanted to know whether my previous statements on hominins were really justified or if I had been misled by using only one clustering method (distance correlation, which it turns out was not nearly as bad as that July critique implied). We are well aware that we should be looking at whole bodies rather than just heads, but that work is currently ongoing and not ready for publication. In the mean time, this paper is a pretty important one, given the shadow of doubt cast on that previous work.
We looked at two sets of characteristics: 1. The original characters from eleven years ago that led me to place Australopithecus sediba in the human holobaramin. 2. The most recent set of characteristics that includes new character states for Ar. ramidus, Au. anamensis, and Homo floresiensis. We used the most recent methods recommended in my paper from this summer.
So what did we find? Was I wrong about Homo naledi and Australopithecus sediba being human? As far as we can tell from these results, the answer is no. Or at least, the new results give me no reason to revise my previous conclusion. Pretty much everything we did in this study agreed nearly perfectly with what I had done in all my previous studies. The only outstanding question that is not resolved is the position of Homo floresiensis, which remains a kind of annoying problem.
Now it's important to place this work in the larger context of creationist discussions of human origins. None of this work proves me right or proves that Au. sediba is human or that Homo naledi is human. I simply failed to falsify my original conclusions that Au sediba and Homo naledi were human. That may sound pedantic, but it's an important distinction. In science, "not being shown to be wrong" is not the same as "being right." There are still outstanding questions, primarily about the rest of the skeleton. As I said, that work on the skeleton is presently underway, and we're hoping to have something publishable in 2022 or 2023.
With that caution in mind, it's still a pretty happy result, at least for me. Given the data that we have and the methods we used (conventional and otherwise), my description of humanity including Au. sediba and Homo naledi is still the best inference and entirely justifiable. This to me closes the book on this most significant challenge to statistical baraminology yet published. In my previous work, I found that concern was warranted, but the empirical results were very favorable to the original conclusions. Here, I found no new reason to suspect I had been mistaken about hominins.
Check back tomorrow for commentary on Lee Berger's new announcement, and in the mean time, check out our new paper:
Sinclair and Wood. 2021. Revising Hominin Baraminology with Medoid Partitioning and Fuzzy Analysis. ARJ 14:451-462.
Have you read my book? You should check that out too!
