Did homonins learn to walk upright in trees before walking upright on the ground?

Thorpe and colleagues, publishing in the journal Science, have performed a study of orangutan movement that is turning the traditional view of the evolution of walking on its head.

The traditional view is that walking upright on the ground -- terrestrial bipedalism -- was preceded by a phase of knuckle walking on the ground. It was speculated that this phase would closely resemble how we see modern gorillas walking on the ground today.

However, Thorpe and colleagues, by observing orangutans moving through trees, are arguing that there was another way that terrestrial bipedalism could have developed: by becoming bipedal in the trees first. In this model, the animal uses his or her arms to stabilize the body while walking upright in the trees. This form of bipedalism could have given way directly to terrestrial bipedalism. Money quote:

Locomotion on flexible branches is safer if supported from above as well as from below. The advantage of hand-assisted bipedality is that the hand assistance ensures maximum safety while the bipedalism enables an unloaded hand to reach out for feeding, weight transfer, or balance in the peripheral branches, where the majority of preferred foods are situated and where primates must cross between tree crowns. Effective gap-crossing behaviors are highly advantageous for primates, because crossing rather than circumventing gaps in the canopy can dramatically reduce the energetic costs of travel, especially where a change of height within the tree would otherwise be required. Hand-assisted locomotor bipedality, adopted under these strong selective pressures, seems the most likely evolutionary precursor of straight-limbed human walking. This proposal is in accord with the conclusions of a comprehensive assimilation of studies of fossil morphology, computer modeling, human skeletal development, and experimental studies of human exercise physiology, suggesting that, despite differences from humans in body form (such as the retention of long arms), the best-known early hominin, Australopithecus afarensis, would have been a quite efficient upright biped, over short distances. (Emphasis mine. Citations removed.)

This has not only reinvigorated the debate about the origins of bipedalism. It has brought into question whether we can use anatomical evidence of bipedalism as a defining feature of early homonins. O'Higgins and Elton, in a review of the paper also in Science, comment:

Central to these debates is whether bipedalism arose in the trees and was taken to the ground, or whether it arose from an ancestor that was already terrestrial. The orangutan data presented by Thorpe and colleagues strongly suggest the former, and could also explain how hominin bipedality arose without needing to go through the stage of inefficient "bent-hip, bent-knee" bipedalism typical of modern chimpanzees. Crucially, the orangutan model also illustrates the way in which large-bodied primates could evolve straight-limbed bipedalism in arboreal contexts.

A number of fossils contemporary with the likely split of the chimpanzee-bonono and human clades, between 4 million and 8 million years ago, have been claimed to show anatomical evidence of upright posture and bipedal walking. These include Sahelanthropus tchadensis, Orrorin tugenensis, and two species of Ardipithecus. Although there is no general agreement on the locomotor and taxonomic affinities of these fossils, one possibility may well be that they are evidence of different ways of shifting from the ancestral type of hand-assisted arboreal bipedality proposed by Thorpe and colleagues. In later hominins, there is also evidence for locomotor diversity, within and between lineages. Limb proportions, for example, differ in Australopithecus afarensis and Au. africanus, and there is a range of foot morphologies in hominins from around the same time period. Thus, bipedal walking might have evolved independently in various early hominins. This could have occurred if multiple lineages originated from an earlier arboreal ancestor that used hand-assisted bipedalism. If that was the case, can anatomical evidence for bipedalism really be used as a crucial defining feature of hominins?

With the orangutan model, Thorpe and colleagues present a plausible and elegant argument in favor of the emergence of bipedalism in an arboreal rather than terrestrial context. In doing so, they have reinvigorated the debate over the emergence of behaviors preadaptive to bipedalism, and have shifted the focus back into the Miocene. A prediction of their model is that diversity of locomotor behaviors, including bipedalism and knuckle walking, could have arisen among descendants of an arboreally bipedal large ape. We must now question whether morphologies that indicate bipedalism can be used to identify hominins at the base of their radiation. This then raises the issue of whether we can unequivocally identify any traits that are truly diagnostic of early hominins. (Emphasis mine. Citations removed.)

Hat-tip: Eurekalert.