Australopithecus sediba: human by accident?

Australopithecus sediba, everyone’s second favourite Australopith after Lucy, was back in the news last week as more research on this fascinating specie was published. First discovered in South Africa in 2008, the species made headlines when preliminary investigations revealed that they were a mixture of earlier ape-like (Australopith) and modern traits (Homo). This raised the intriguing possibility that we’d discovered the “ape man” (or ape woman, if you prefer) from which the more modern-looking species evolved.  

MH1; one of the Australopithecus sediba skeletons found

MH1; one of the Australopithecus sediba skeletons found

Such hopes were short lived however, as it was discovered Au. sediba was ~1.9 million years old (Pickering et al., 2011). Since the first members of Homo lived 2.3 million years ago (Aiello and Wells, 2002),  Au. sediba obviously couldn’t have been their ancestor (unless they had a Delorean. That always leads to weird family relations).  Some maintain that the earliest members of Homo don’t actually belong to our genus, and that it actually evolved much later (Wood and Collard, 1999). If this is the case then Au. sediba could still take center stage in the story of our evolution.

At the very least Au. sediba appears to be closely related to the ancestor of Homo, even if they aren’t that ancestor themselves. As such they can provide valuable insights into the origins of our genus. Now the initial investigations have been completed and published, revealing how Australopithecus sediba moved, what they looked like and where they fit in our evolutionary tree (Berger, 2013).

These results confirm our earlier suspicions that Au. sediba had a mixture of ape-like Australopith and more modern Homo traits (when wanting to sound fancy, scientists refer to such a mixture as a “mosaic”). For example, they had a cone shaped rib-cage like Australopiths and modern apes (compared with our barrel shaped rib cage). However, the bottom of this rib-cone is much less slender than it is in an ape, being more similar to the human rib cage (Shmid et al., 2013). 

The hand of sediba, with short stubby human-like fingers

Other such mosiac traits include the arm. This research shows that shoulder blade was orientated upwards (like an ape), so  sediba would look like they were shrugging their shoulders. Such an orientation makes it easy to reach over their head, so their really long arms (again, like an ape) could grab onto branches (Churchill et al., 2013). However, whilst apes and earlier Australopiths have long curved fingers to “hook” onto branches, Au. sediba‘s fingers are short and stubby (Kivell et al., 2011). This is a human adaptation to better handle tools. Currently we only know of one species of Australopith that used stone tools; so this discovery suggests it may have been a lot more widespread than we thought.

Almost every aspect of Au. sediba‘s anatomy has a similar mix of modern and old traits; from the jaw through to their backbone. Put them all together and you wind up with something that looks like the image below. Definitely an ape-like Australopith, but with some key modern traits (Berger, 2013).

A modern human (left); sediba (middle) and chimp (right)

A modern human (left); sediba (middle) and chimp (right)

However, what I find the most interesting are those traits unique to sediba, not present in either other members of Australopithecus or Homo. Most notably this includes sediba‘s way of walking. Although they were bipdeal, like other Australopiths and humans (with many similar adaptations), an analysis of their lower limbs reveals they walked with their feet  twisted slightly outwards. No other member of our family walks like this (DeSilva et al, 2013). 

This unique way of walking has led some of the researchers involved in these new papers to suggest that Australopithecus sediba may not be as closely related to modern humans as early research suggested. They claim that sediba may have split from the human lineage millions of years earlier, before even Lucy lived (Schmid et al., 2013)!

A possible reconstruction of the human family tree, taking these new findings into account

A possible reconstruction of the human family tree, taking these new findings into account

If this is the case then it would mean that all their human-like adaptations were the result of convergant evolution. That there were multiple lineages independently evolving towards the modern human body plan. Perhaps this speaks to just how well adapted our form is for life in Africa. It also raises the fascinating possibility that had things played out a little differently there may have been another genus of tall, upright, smart hominins. Remember that Neanderthals, for all their differences, are still classified as the same genus as us. Imagine if a completely different genus of intelligent hominins had developed!

Although that’s a fascinating possibility, not all the researchers agree with it. Irish et al (2013) point out that some of sediba’s traits – such as the shape of their teeth – are so similar to Homo that it’s very unlikely both groups evolved them coincidentally. Their detailed analysis of sediba’s location in the human family tree place them much closer to humans; indicating both us and them developed from southern Australopiths, like Australopithecus africanus.  This shows that the emergence of Homo was more complex than we thought, with multiple lineages emerging from our ancestors.

In short, this research shows that Australopithecuses sediba was the mosaic we thought it was. However, it’s unique features force us to change where we place it in the human family tree and by doing so we revolutionise our understanding of the entire  course of human evolution; either by showing another lineage of hominins existed that was developing human-like traits coincidentally, or revealing the genesis of our genus was a lot more complicated than we thought. This is the final nail in the coffin of the idea that Au. sediba was the ancestor of modern humans, but frankly I am not disappointed. The ideas which replace it are just revolutionary.


Aiello, L. C., & Wells, J. C. (2002). Energetics and the evolution of the genus Homo. Annual Review of Anthropology, 323-338.

Berger, L. R. (2013). The Mosaic Nature of Australopithecus sediba. Science,340(6129), 163-165.

Churchill, S. E., Holliday, T. W., Carlson, K. J., Jashashvili, T., Macias, M. E., Mathews, S., … & Berger, L. R. (2013). The upper limb of Australopithecus sediba. Science340(6129).

Irish, J. D., Guatelli-Steinberg, D., Legge, S. S., de Ruiter, D. J., & Berger, L. R. (2013). Dental Morphology and the Phylogenetic “Place” of Australopithecus sediba. Science, 340(6129).

Kivell, T. L., Kibii, J. M., Churchill, S. E., Schmid, P., & Berger, L. R. (2011). Australopithecus sediba hand demonstrates mosaic evolution of locomotor and manipulative abilities. Science333(6048), 1411-1417.

Pickering et al, (2011). Australopithecus sediba at 1.977 Ma and implications for the origins of the genus Homo. Science 333, 1421

Schmid, P., Churchill, S. E., Nalla, S., Weissen, E., Carlson, K. J., de Ruiter, D. J., & Berger, L. R. (2013). Mosaic Morphology in the Thorax of Australopithecus sediba. Science340(6129).

Wood, B., & Collard, M. (1999). The human genus. Science284(5411), 65-71.

(As an aside, I predict young earth creationists will respond to these findings by 1. playing up the ape-like nature of the ape/human mosaic traits and 2. using the fact that it’s relationship to humans is being re-evaluated to question whether it – and other Australopiths – is related to humans at all. They may also use the fact that first impressions were wrong to question the validity of palaeoanthropology all  together) 

19 thoughts on “Australopithecus sediba: human by accident?

  1. Just came across your blog and had a question that’s been bothering me for some time. You write,
    “Such hopes were short lived however, as it was discovered Au. sediba was ~1.9 million years old (Pickering et al., 2011). Since the first members of Homo lived 2.3 million years ago (Aiello and Wells, 2002), Au. sediba obviously couldn’t have been their ancestor….”

    I read this sort of thing all the time in discussions of evolution, and it just seems wrong. OK, we have fossils of sediba that date to 1.9 Mya. But that only tells us they were around at that time. It doesn’t tell us when they arose or when they died out. IIRC, it’s not unusual for a species to be around for a million years or so, right? So sediba could easily have arisen 2.5 Mya or earlier, no? And therefore could have been ancestral to Homo.

    Any chance you can clear this up for me?

    • The problem of “there may be earlier/later examples” applies to just about every discovery in palaeoanthroplogy. There may be an earlier sediba, an earlier first stone tool, earlier first art etc. However, most scientists refrain from spending much time speculating about these possibilities since, although it’s likely we will discover earlier/later examples, any claims we make about them would be baseless until we’ve actually found them.

      However – even if we did discover an earlier sediba that lived at the right time – the fact they walked differently to us suggests they weren’t our ancestor after all.

  2. Thanks for the reply, Adam, and yes, I get that there are now other reasons to rule out sediba.

    It’s the logic of ruling out one species as an ancestor based on very incompletely known dates that bothers me. OK, we shouldn’t speculate where we don’t have evidence. But to use the dates to rule out ancestry is itself a form of speculation. It requires speculating that sediba WASN’T around 500,000 years earlier. That’s just as unjustified as speculating that it WAS around.

    Sorry to harp on this, as I said, it’s sort of a bugbear for me. I’ve even seen actual published papers that say “X couldn’t have been ancestral to Y because they existed AT THE SAME TIME.” That’s like saying my mom can’t be my mom because we are both still alive.

    • The thing to remember is that they don’t use this data to rule out various ideas for certain. Scientists will happily adapt their ideas with little fuss as new evidence is revealed.

  3. I’m another blow-in via Pharyngula, so apologies that I’m sure this is such a basic question it’s quite silly … but reading this I’m confused about what the ‘rule’ is that divides Homo and non-Homo. I know there were multiple species that fit the class, though H.Sapiens and H.Erectus are the only two I recall from my education-via-BBC-documentary, but how do you all decide that something is a Homo not an Australopithecus or whatever??

    • No worries, it’s a good question I may use for next weeks “question of the week”. It’s always worth reminding me not everyone has read everything I’ve ever written (and I don’t take that as a personal slight).

      At any rate, the traditional distinguishing features between us and Austro was stone tool use and large brains (our earliest members had around 900cc brain capacity, comapred to 350/400 for Austro). Then we found Homo habilis, which used stone tools but only had a brain ~700cc. So the “big brained” part of the definition of Homo was dropped. Then we recently found some clearly Australopithecus specimens using tools! So now that part of the definition does’t work either.

      So now they’re separated by shades of grey rather than the black and white of things like “tool use”. How flat was their face? What size was their teeth? How long were there legs etc? This gradation is what you’d expect to see if they evolved, so although it’s confusing for us it’s a nice bloody nose for any creationists watching.

      At any rate, all of this means it’s hard to give a specific definition of what makes a creature Homo. The upshot of this is that some early members of our genus (like Homo habilis) are being reconsidered, with some arguing they’re actually Austro!

      Sorry I can’t give a properly definitive answers, I hope you decide to stick around nonetheless.

  4. Adam
    I was thinking of sending this comment to AiG – unless you think it inaccurate:
    “ (item 1)

    “Other evolutionary researchers have suggested that Lucy (Australopithecus afarensis)—a competitor in this race to the root of the human evolutionary tree—did not have a human sort of bipedalism, but a different “australopithecine bipedalism” altogether. (In other words, the tree-swinging arboreal ape wasn’t really bipedal. See A Look at Lucy’s Legacy and Lucy, the Knuckle-walking “abomination”? to learn more.)” UNTRUE. ‘Lucy’ was bipedal.

    “On the other hand, the special issue of Science devoted to sediba shows that it was well-designed for climbing and swinging. It was an ape. No amount of wishful thinking can make it the “transitional form” the researchers claim that it is.” UNTRUE.
    “The less well-preserved elements of the lower rib cage suggest a degree of human-like mediolateral narrowing to the lower thorax, indicating a rather unsuspected mosaic anatomy in the chest that is not like that observed in Homo erectus or H. sapiens.”
    “The new analysis shows this species – Australopithecus sediba – had a human-like pelvis, hands and teeth, and a chimpanzee-like foot”.”

    • As I predicted at the end of this article, creationists would likely respond to it by playing up the ape-like characteristics an ape/human intermediary naturally has. AiG have fulfilled this prophecy superbly.

      For example, they talk about how well suited her arm was to climbing, missing out the fact their fingers were comparatively short and more human like; which isn’t a particularly useful tree climbing trait. That’s not to say that they didn’t climb trees, just that they’d sacrificed some of this arboreality in favour of a dextrous hand, a fact AiG glosses over.

      Their cherry picking of data doesn’t end there. The papers notes how sediba walked with a strange gait, then note how humans who walk with such a gait suffer from muscle strain etc. AiG cites both of these fact as evidence against it walking upright but conveniently omits the very next line from the article which goes “although hyperpronation can have pathological consequences in modern humans, we are proposing here that the skeleton of Au. sediba reveals a suite of anatomies that are adaptive for…this kind of walking”

      So by only citing certain aspects of their anatomy, AiG can make sediba look like a regular arboreal, non-bipedal ape. Big whoop, you can do the same thing to humans. Except if you cherry picked humans like that, everyone would call you a liar. Can we say the same thing about AiG because they’re doing it to sediba?

  5. It would be very interesting to have a interactive web site/page where all known “finds” and present digs in progress are indicated, on a map, of Europe, Africa and Asia. On which, if you could enter a dig site with you arrow (on the computer), you have access to basic data retrieved on that site. Then, for example, if Lucy’s name came up- where she was uncovered, one could click on her and on specifics about her. Selected archeologists should then be requested to enter and update data as it is clarified and once processed ?
    It seems now that even western chimps ( 300cc brains) use tools. As well as starlings with 5 cc (?) brains. Most probably it will not be size, but ability, deduced from “cultural” discoveries to come, that will distinguish Homo from Hominid. Hominid from Ape? To put it simply.
    Clearly, even studies in primate and human behavior are helping to clear our minds. It is one very large mosaic to be pieced together.

    • That would certainly be an interesting and useful resource. If you make the website I’ll happily start plugging in data. I’ve already gathered a bit for research I’m doing.

      As for brain size’s relationship to tool sue, it’s worth noting that brain size relative to body size seems to be the key, not just absolute brain size. For example, although crows have a small brain, compared to their body it is relatively the same size as a chimps. Crows are also the most adept tool users in the bird family.

  6. In reply to Adam at 10.08 am, I’m now sending my comments to AiG, together with a link to this blog post (hope that’s OK).
    If A sediba managed to walk bipedally some of the time, despite the hyperpronation issue, the AiG claim that A afarensis must have been a knuckle-walker looks increasingly untenable.

  7. Pingback: Australopithecus sediba: is she or is she not a human ancestor? |

    • Their ultimate conclusion rests on the old “common features evidence of common designer” claim. That’s an unfalsifiable hypothesis making it, and their refutation of sediba as part of the human family, unscientific.

You evolved too. Have a say.

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s