What came before bipedalism?

Chimps have a special place in our culture as “stand ins” for our early ancestors. The classic ascent of man image depicts us evolving upwards from a knuckle walking creature, whilst scientists use chimps when they want to research the behaviour of our ancestors. However an increasing amount of evidence suggests that our lineage differed from chimps in several key ways; including how we move. It would seem that our lineage never knuckle walked like a chimp (Lovejoy and McCollum, 2010).

The Australopithecines emerged around 4 million years ago and were the first adept bipeds in our family. However, their skeleton contains no evidence of them being an ex-knuckle walker (Marze, 1986). Roll the clock back even further to Ardipithecus ramidus, who lived around 4.4 million years ago and even though they were that skilled at walking upright there is still no evidence of them walking on their hands like a chimp (Lovejoy et al., 2009).

So if our ancestors didn’t knuckle walk before they became bipedal, how did we move before we started walking upright? Two different ideas have emerged.

Ardipithecus ramidus walking along a branch (the left foot), showing a possible origin of terrestrial bipedalism (the right foot)

The first suggests that bipedalism evolved as a way of walking across branches, allowing our ancestors to traverse the jungle with the greatest of ease. Evidence for this hypothesis comes from the aforementioned Ardipithecus ramidus. This species had some bipedal ability, yet still lived in the trees, suggesting that walking upright did in fact evolve as a means of travelling around the tree tops (Lovejoy et al., 2009).

The alternative hypothesis argues bipedalism evolved from swinging through the trees (like an orang-utan). Dangling below the trees like this resulted in a vertical body well suited for bipedalism; biasing our evolution in that direction when it came time for us to move on the ground. Evidence for this idea comes from apes that lived before the chimp and human families split 7 million years ago. Researchers noted that the earlier apes which swung from the trees are those with these vertical bodies. Further, the only living ape that swings from the trees is also the best at walking upright (apart from us).

An orang-utan showing off just how much better at bipedalism it is than a hunched over chimp

An upright orang-utan showing off just how much better at bipedalism it is than a hunched over chimp

Although both ideas had a fair bit of evidence supporting them, neither could gather enough to demonstrate that it was the correct answer. However, all of this changed when scientists re-examined Oreopithecus, an European ape that lived in the swampy jungles of Italy 9 million years ago (back when Italy had swampy jungles). When it was first discovered, some thought that it might be a human ancestor. Since then the vast majority of scientists have moved away from this position given that Oreopithecus has an unusually small brain and strange teeth that indicate it is not part of the human line.

Despite not being directly related to us Oreopithecus provides important information on how bipedalism in humans emerged. It appears to have swung through the trees, yet was also a relatively good biped. In other words, it indicates that bipedalism did emerge from swinging through the trees, confirming the “swinging-as-a-precursor-to-bipedalism” hypothesis (Kohler et al., 1997).


Or at least, it did until last week. New research has just been published that re-re-examines Oreopithecus and finds little evidence for bipdealism in the species. The key piece of evidence for upright walking originally cited was a curve in the lower back, known as lumbar lordosis. Humans have it too; you can feel it if you stand up and trace the bottom half of your spine. This curve helps keep the centre of gravity where it should be so we don’t topple over.

People thought Oreopithecus also had it, but this new research reveals that it actually doesn’t. There is distortion in the fossil which makes it look like the back is curved; but when this distortion is taken into account the lumbar lordosis disappears (Russo and Shapiro, 2013). This may seem like a fairly basic error to make but remember, it has happened before. Examining skeletons that have been buried for millions of years is a tricky business and sometimes mistakes are made.

And without Oreopithecus the biped we’re back to square one, with no way to tell if bipedalism arose as a result of walking on branches or dangling below them. It would seem that the origins of bipedalism are as mysterious as ever.


Köhler, Meike; Moyà-Solà, Salvador. (1997). Ape-like or hominid-like? “Ape-like or hominid-like? The positional behavior of Oreopithecus bambolii reconsidered” . PNAS94 (21): 11747–11750

Lovejoy, C. O., Suwa, G., Spurlock, L., Asfaw, B., & White, T. D. (2009). The pelvis and femur of Ardipithecus ramidus: the emergence of upright walking.Science326(5949), 71-71e6.

Lovejoy CO, & McCollum MA (2010). Spinopelvic pathways to bipedality: why no hominids ever relied on a bent-hip-bent-knee gait. Philosophical transactions of the Royal Society of London. Series B, Biological sciences, 365 (1556), 3289-99

Marzke, M. W. (1983). Joint functions and grips of the Australopithecus afarensis hand, with special reference to the region of the capitate. Journal of Human Evolution12(2), 197-211.

Russo, G. A., & Shapiro, L. J. (2013). Reevaluation of the lumbosacral region of Oreopithecus bamboliiJournal of Human Evolution.


12 thoughts on “What came before bipedalism?

  1. Since I am a lumper not a splinter I see Ardipithecus as an Austrolapithecus. As for bipedalism I see Austrolopithicus, Ardipithecus and Oreopithecus evolving from the apes and, hence, knuckle walking and upright walking together like apes do today. The reconstruction of the Ardipithecus ramidus should be more bent-over, more ape-like looking. Even in Homo erectus/habilis we still see ape-like forms. It is not until the Neanderthals that we see upright walking. There is a big gap between Australopithecus and neanderthals. If the trees were moving ‘south’ they, the hominids, still would have to adapt to changing environmental conditions. Whether they moved south with the forests would be the logical guess. The genetic mutations for upright walking would be independent of environmental conditions and come later. I am also genetically inclined not environmental, nee Neo-Lamarkian, inclined. The basis of evolutionary change is genetical.

    Date: Tue, 30 Jul 2013 11:35:51 +0000 To: ralphgironda52@hotmail.com

    • I think lumping is really the only tenable position these days. If humans and Neanderthals can interbreed then the handful of differences between, for example, ergaster and erectus probably amounted to nothing as well.

      Speaking of which, whilst Homo erectus et al. still did have some ape traits in their body, they were adept bipeds; just as good as us. In some regards they were better. H. erectus used less energy when walking than we do (although we’re more efficient runners).

  2. Hello Mr. Benton. I just started reading your blog a few weeks ago. Lots of good stuff here.

    In relation to your claim that the ancestor of bipdeal hominins was not a knuckle-walker: while this opinion is held by some researchers (such as Owen Lovejoy and Tracy Kivell to name a few) there are also plenty of other researchers who disagree (see Richmond & Strait, 2000 “Evidence that humans evolved from a knuckle-walking ancestor” and Richmond et al. (2001) “Origin of Human Bipedalism: The Knuckle-Walking Hypothesis Revisited” for a start).

    Addtionally, I’m not convinced that Ardipithecus can be used to resolve this question. There’s a really good chance it’s an closely related, but divergent hominin side branch and not necessarily a direct ancestor to later bipedal ancestors.

    Finally, remember that both chimps and gorillas are knuckle-walkers. The most parsimonious construction of the hominid family tree would suggest that bipedal hominins had a knuckle-walking ancestor too. Otherwise, knuckle-walking would have to had to evolve two separate times, once in chimpanzees and once in gorillas.

    Anyway, just my two cents. I also write about paleoanthropology on the interwebs. http://livelikedirt.blogspot.com Come on by and check it out sometime. Party on.dude!

    • Figuring out just who is related to who is notoriously difficult, so yeah a strong chance Ardi isn’t directly related to them. Nonetheless, they do at least appear to be closely related so they can still provide information on their evolution. However, this muddling of ancestry (amongst other reasons) means that Ardi alone isn’t enough to resolve this debate, which is why I ultimately concluded the origins of bipedalism are still a mystery.

      Which is my long winded way of saying “I agree with you.” I disagree a bit more when it comes to the presence of knuckle walking in our past. It seems to me that the balance of evidence is shifting away from this conclusion (although not far enough to dismiss it entirely).

      For example, few studies have managed to find knuckle walking traits in hominins and those which have are…imperfect. Richmond & Strait’s 2001 paper, for example, is based on a broken cast (which they didn’t realise was broken), leading to mistakes in their data. The anatomical traits they did correctly identify turns out not to be as strongly linked to knuckle walking as they suggested. I wrote more about it here https://evoanth.wordpress.com/2012/12/20/answers-in-genesis-v-evoanth-4-the-wrist/

      One of my lecturers led the charge against the paper, and does delight in telling how those who dared oppose him only did so because of faulty data.

      Palaeoantho on the interwebs seems to be a rare thing. We must huddle together for warmth.

    • Actually, although the both are adept knuckle walkers, the composition of the wrists in both Gorillas and Chimps are different enough to assume that knuckle walking probably did evolve twice. See Kivell and Schmitt, 2009 for more detail.

  3. Pingback: Arkarbor

  4. I think it’s a real underestimate to say that only a few studies have managed to find knuckle walking in hominins. I just dumped a couple on you here, but I have plenty more where they came from (I think you’re forcing me to write a long blog response on this topic – I’m aiming for Saturday, but I’ll let you know when its up). I do think it’s correct to say that papers that argue knuckle-walking is a shared trait among the African ape and human clade are imperfect. But I don’t really think that’s a huge problem, as nearly all papers are imperfect in some manner. That’s part of the reason there’s so much debate in paleoanthropology. That’s why we do science, to get a little bit closer to truth. I’m not sure we can ever be so perfect as to obtain some absolute final truth on a given topic, especially one as tricky as this.

    To start, Pliocene hominins aren’t always the best place to look for knuckle-walking clues. In order to learn more about the locomotive style of the shared ancestors of gorillas, chimpanzees, and humans you really have to look at their potential shared common ancestors, the Late Miocene apes. It might be hard to pin down a precise hominoid ancestor, but there are some interesting pieces of evidence which can help to inform this debate (again, I’ll lay out my arguments more specifically when I write my own blog on this subject).

    In the meantime I’ll check out your blog post on the Richmond & Strait 2001 paper.

    • You’re right, I haven’t counted the papers I don’t know how prevalent the idea we evolved from knuckle walkers is. What I meant to say is that on balance, I think there is stronger evidence we didn’t develop from a knuckle walker.

      That said, I’m not sure the evidence is strong enough to justify the complete dismissal of a knuckle walking origins. I’ll still happily read, write about and find interesting research looking at that possibility. So if you did write about that I’d enjoy it.

      Especially if it had information about a knuckle walking Miocene ape. I wasn’t aware any existed, and it would have fascinating implications (Although it might not be as scientifically sexy as a bipedal Oreopithecus)

  5. Okay. Here are my arguments as to why hominin bipedalism most likely evolved from a knuckle-walking ancestor. My post is focused on the arguments surrounding the homologous or homoplastic nature of knuckle-walking in chimps and gorillas, but the subtext is that humans evolved from a knuckle-walker. Check it out if you get a few minutes (and/or need some help sleeping): http://livelikedirt.blogspot.ca/2013/08/on-origin-of-knuckle-walking-by-means.html

  6. Pingback: Skull spotting: part 1 | EvoAnth

  7. I have been researching a new concept for 18 years; primate evolution has taken advantage of embodiment.

    Imagine if you transplanted a chimps brain into a human body? When the chimp stood up, its human body would be fully upright. Would its limbic system respond to this as a dominance display?

    Whilst upright the larger human version of the zygomatic major (smile muscle), would respond to the pull of gravity by contracting. Would the chimps limbic system respond to this as an involuntary play face?

    Now please consider this; the human limbic system is essentially a chimp.

    I am suggesting that one reason why evolution may have resulted in uprightness, is to take advantage of the limbic response to uprightness. The limbic response to uprightness and zygomatic major contraction is to immediately increase our white blood cell numbers and reward us with beneficial neurotransmitters.

    So if the body changes but the limbic system does not, our body is the embodiment of positive affect.


    Dean Abel

    • But there’s also acclimatisation to take into account. If an animal repeatedly encounters a stimulus they will gradually grow numb to it, so whatever responses our ancestors might have felt to bipedalism would wear off too fast to be of any evolutionary significance.

You evolved too. Have a say.

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

WordPress.com Logo

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

Google photo

You are commenting using your Google 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