However, this is exactly where the evidence leads.

Overall, I don’t expect that the entire anthropology community will suddenly abandon everything that has been taught for decades. However, my point is the following:

1. We see the spine anatomy of a hindlimb supported upright ape in Morotopithecus, Pierolapithecus, Oreopithecus. The data is compelling and extensive – and I have detailed it in technical raw data form in my book: Axial Character Seriation in Mammals, which republishes my Harvard PhD Thesis. The underlying patterns are extracted and synthesized in my recent PLoS ONE paper “Homeotic Evolution of the Mammals, Diversification of Therian Axial Seriation and a Morphogenetic Basis for Human Origins” and in my Neurosurgical Focus article. The context in evolutionary theory is explained in my recent book “The Upright Ape: A New Origin of the Species” which has a foreword by David Pilbeam – currently Dean of Harvard College and certainly one the most knowledgeable and experienced paleoanthropologists in the world.
2. We have evidence of an upright hindlimb supported Orrorin based on the femur and Sahelanthropus based on the skull.
3. There is no convincing fossil evidence at all of a non-bipdeal hominoid outside of the proconsulid group.
4. We have an early outgroup whose infants have innate bipedal walking (see the video Hominiform Progression). The Siamang video is interesting because of the innate bipedalism. As I point out in the video, John Fleagle has seen young siamangs of this age walk bipedally high in the canopy in Malaysia.

It is typical to say that all of this is irrelevant and misleading and should be ignored. There was a quadrupedal common ancestor for chimps and humans and the human lineage suddenly and majestically stood up about 5-6 million years ago. However, I feel that there is no a priori reason why we must ignore all of the evidence for early bipedalism.
None of the skeletal evidence can ultimately distinguish between “short bursts” and long distance bipedalism as Kambiz points out in his post. My focus is on the character state and whether the crucial anatomical basis is a shared derived feature of a hominiform clade.

It can certainly be said that the siamangs only engage in bipedalism for short bursts, but that is also true of their brachiation. Similarly, the chimps and gorillas knuckle walk and the orangutans fist walk only in short bursts. However, the important point is that chimps, gorillas and orangutans seem to locomote in diagonal posture more than 90% of the time and only occasionally deploy a short burst of bipedal walking. I would argue that they have very bad spinal architecture for bipedal walking. On the other hand, hylobatids use bipedalism 100% of the time when they locomote on the ground no matter how long the burst of activity. If a hylobatid has to travel a long distance on the ground – it does not lapse into a quadurpedal gait – it just keeps walking bipedally. There is an important difference in the role of bipedalism as deployed by hylobatids and hominines as opposed to what we see in chimps, gorillas and orangutans.

This would be a morphogenetic origin for upright bipedal walking rather than an adaptive origin. Essentially, the origin of upright posture was not driven by any ecological scenario, but rather occurred suddenly as a result of a morphogenetic mutation in the Pax genes. Various descendant forms will have lived in various environments with variously optimized versions of primary upright bipedalism on large horizontal arboreal supports and on the ground.

It is certainly easier to assert that Morotopithecus was upright and hindlimb supported – based on spinal anatomy – than to prove it was primarily bipedal or a long distance walker. However, this is where the video showing the baby siamang learning to walk bipedally is relevant. Yes, you could argue that innate bipedalism evolved independently in parallel in hylobatids and hominines, but is also reasonable to consider that since this is so unusual, that it reflects descent from a common ancestor that had this feature. Essentially – an eight month old Morotopithecus baby would do the same thing that we see in the two descendant groups (hominines and hylobatids) – the baby would innately begin to walk bipedally as it’s primary locomotor pattern.

So – if the chimp-human split did take place 6 million years ago (as the molecular data suggests), then what do we do with Sahelanthropus which many believe was a full time upright biped but which lived 7 million years ago?

If you want a slow gradual evolution of bipedalism, you need to push the chimp human split back to say 8 million years. However, there is an alternative explanation. Upright bipedalism was already the primary means of locomotion in the common ancestor of chimps and humans – Sahelanthropus is ancestral to both lines.

What defines a “human?” I have taken the position that it is a body plan (bauplan). Most of us have accepted that early Australopithecines whose brains and skulls were chimp-like, should be considered human and not ape. When you find a fossil such as Sahelanthropus that has a “chimp-like” skull from the point of view of its face and brain, but has the skull base of a human (and presumably upright bipedal post-cranial anatomy) – how can you tell from the fossil if it’s an ape or a human?

The Hennigian cladistic approach lets us say that the isolation point between the chimp and human lineages – where hybridization became impossible – is the origin point of humans. However this means that the definition is arbitrary since ape and human would pretty much look identical at that time.

Another alternative is to stick with our current definition – a hominoid whose anatomy reveals that it is primarily an upright biped is a human. I have proposed the term “hominiform” to refer to a clade of hominoids that share the Morotopithecus spinal transformation (septo-neural transposition – in which the dorso-ventral plane of the body flips from ventral to the spinal canal to a new position dorsal to the spinal canal) and the styloid process is converted into a neomorphic hominiform lumbar transverse process. The synapomorphies would include innate bipedal walking in the infants.

Among hominiforms we have primitive “eubipedal” types (most Miocene and Pliocene fossil hominiforms, the hylobatids and the hominines) and derived “metabipedal” types (lineages of chimps, gorillas and orangutans) that have abandoned bipedalism as their primary locomotor pattern on the ground.

Sahelanthropus appears to be a human species that is representative of species in the line of ancestry to both the chimpanzees and hominines.

Aaron Filler, MD, PhD

The article doesn’t mention that there are numerous other reports about sexual dimorphism in the human lumbar lordosis – about half of the studies found no statistical difference between males and females and the other half found a small difference. All of these are referenced and updated in a recent article titled, “Radiographic Analysis of the Sagittal Alignment and Balance of the Spine in Asymptomatic Subject” by Vialle and colleagues (Journal of Bone and Joint Surgery Am. 2005, 87:260-7). Hence the idea of sexual difference here is not new and although the wedging measure used by Whitcome et al produced a statistically significant difference, it is not clear the wedgeing of bones equals real differences in lordosis and it is not clear that the sexual difference will hold up in larger studies. They don’t tell us why the other studies showed no sexual difference.

Nonetheless, the story makes the same point I have made in my books, Axial Character Seriation in Mammals and The Upright Ape: A New origin of the Species, as well as and in two recent articles – one on the evolutionary origins of back pain – in which I discuss the evolution of lordosis in the online journal Neurosurgical Focus, and the other, my article in the online journal PLoS ONE.

Most mammals have a lumbar spine whose architecture tends to passively resist the force of gravity when positioned in a horizontal or diagonal position. In humans and such ancestors as Morotopithecus, however, the spine mechanics are reversed so that quadrupedal stances are very tough on the spine. Only fully upright stance is comfortable. In the Neurosurgical Focus article, I also point out that the backward projection of the hip bone – the iliac crest (or more specifically the ‘posterior superior iliac spine’) gradually moves farther and farther posteriorly so that the back muscles can resist the weight of young carried in front. This provides another basis for understanding the time of onset of the lordosis. However, in the Neurosurgical Focus article, I also show how the conversion of the longissimus muscle attachment onto the lumbar transverse process would have been sufficient to allow for carrying infants in the Morotopithecus configuration as far back as 21.6 million years ago.

These findings further support the point that position that bipedal walking – full time bipedalism – may be very ancient in the hominoids. It also may explain why various hominoid lineages such as those leading to modern orangutans, chimps and gorillas abandoned upright bipedalism in favor of a re-engineered lumbar spine that could better resist the weight of the late term fetus during long distance travel in horizontal quadrupedal and diagonal postures.

Aaron Filler, MD, PhD
Institute for Spinal Disorders, Cedars Sinai Medical Center