The human family evolved in Africa, but left that continent several times. The earliest of these is dated to ~2 million years ago and associated with the first real "human": Homo erectus.
However, fossils from India dating to 2.6 million years ago indicate there may have been an even earlier migration. This early date would suggest that the migration took place before the evolution of Homo erectus.
It might have been one of the earlier members of the Homo - perhaps even the recently discovered Homo naledi - or maybe even a pre-human Australopithecus, like the famous Lucy fossil.
Or alternatively these Indian bones indicate nothing of the sort.
The bones of migration
The bones in question come from near Chandigarh, India, in the shadow of the Himalayas. This site has yielded more than a thousand fossilised bovid bones.
Of the >1400 bones recovered, three are especially interesting because they appear to have cut marks on them. These cut marks would indicate they were damaged by hominins trying to butcher the animals with stone tools. Which isn't particularly special, it's something we still do to this day.
What makes these three bones interesting is that they've been associated with rock layers dating to 2.6 million years ago. This would vastly predate the first known migration out of Africa, indicating our ancestors were spreading around the world much earlier than we thought. In fact, this date is so early it predates the first recognisable "human" ( Homo erectus) who only appears to be ~2 milion years old.
Or at least, that's what would make them interesting if we could be sure they were 2.6 million years old. In actual fact these bones were found on the surface, without any "made 2.6 mya" labels attached. The claim they're super old comes from the fact that they have a similar discolouration to other fossils from 2.6 million year old rock layers at the site. Also, some of the bones have sediment clumps stuck to them that also appear to have come from these layers.
If you think that sounds a bit sketchy, you'd be right. This sort of evidence is hardly a reliable way for confirming how old something is.
Crocodiles can't make tools
The other issue with bones from the surface is that they're, well, on the surface. As such they're in harms way. Various natural forces could potentially damage the bones, creating the false impression of tool cut marks.
To try and rule this possibility out the authors got a hold of a bunch of modern bones and smashed them up in all sorts of exciting ways. In theory, these could then be compared to the fossil bones to see if any of this natural damage could produce a false positive.
In particular, they focused on "accidental" damage caused by modern people, real stone tool use, and crocodiles (as they were living in the region and their teeth can produce tool-like marks). Of these three possibilites, the damaged to these bones seemed most consistent with genuine stone tool use. Crocodiles weren't using them as toothpicks. They were also able to eliminate a few other possibilities (like being trampled by other animals) based on previous research. After all of this, they could ultimately conclude
After eliminating all natural and animal causes, we can assume they were made by a quartzite cutting edge.
Now, the smart ones amongst you might have noticed a problem. Namely, that 3 - 6 alternative causes for bone damage don't represent "all natural causes". For example, these bones seem to have spent a fair bit of time in the water; yet they could only eliminate water as a possible cause of damage for one of the bones.
Don't get me wrong, the evidence does seem to be most consistent with tool marks. But the relatively narrow scope of the experimental data raises a red flag. Something about all this smells fishy, even if it isn't the crocodiles.
Bad bones (and tools)
Clearly, I think more evidence is needed before we consider the possibility of a pre- Homo erectus migration seriously. And I don't think my demand for more evidence is unfair. There are countless other examples of these sorts of "surface finds" that seem significant, but ultimately lead nowhere.
In fact, many of these failures are cited in this new research as precedence for really old tools (and really old migrations). In reality, they're actually a good reason to treat theseclaims with great caution.
Perhaps the most famous example - and one I've talked about before - are a set of bones with cut marks from Dikika, Africa. Just like these Indian bones they were recovered from the surface, with no reliable way to tell their age. Just like these Indian fossils, the interpretation preferred by the discoverers was that these were cut marks from tools. Except they weren't. Subsequent re-analysis identified that these marks were actually more consistent with trampling damage than tool damage. Despite that, people continue to hold up these bones as evidence of super early tool use (including this new Indian paper).
This paper also points to a set of tools from Pakistan that suffer from a similar set of problems. They were also found on the surface, with no real way to reliably determine how old they were (although that didn't stop the initial discoverer from trying to claim they were 2 million years old). As such, skepticism continues to persist about these tools. This caution could have been alleviated if any follow up evidence confirming there age had been found. Yet it's been nearly 30 years since those were originally found and we're still no closer to knowing how old they really are.
So yeah, when someone turns up trying to rewrite the story of human evolution with surface finds you should be skeptical. They might not actually be wrong, but without more evidence we can't be sure they're right either. People didn't widely accept that tools were being used during the same period as the Dikika bones until an actual contemporary archaeological site was discovered (and even then, that doesn't stop the Dikika bones actually valid).
tl;dr
The human family evolved in Africa, but left that continent several times. The earliest migration is dated to ~2 million years ago, but new fossils from India could push that date back even further. Or it could be nothing of significance. On balance actually, probably nothing of significance. But that won't stop people citing it for years to come, ensuring it frustrates me for just as long.
References
Bar-Yosef, O. and Belfer-Cohen, A., 2013. Following Pleistocene road signs of human dispersals across Eurasia. Quaternary International, 285, pp.30-43.
Dennell, R.W., Rendell, H. and Hailwood, E., 1988. Early tool-making in Asia: two-million-year-old artefacts in Pakistan. Antiquity, 62(234), pp.98-106.
Domínguez-Rodrigo, M., Pickering, T.R. and Bunn, H.T., 2010. Configurational approach to identifying the earliest hominin butchers. Proceedings of the National Academy of Sciences, 107(49), pp.20929-20934.
Malassé, A.D., Moigne, A.M., Singh, M., Calligaro, T., Karir, B., Gaillard, C., Kaur, A., Bhardwaj, V., Pal, S., Abdessadok, S. and Sao, C.C., 2016. Intentional cut marks on bovid from the Quranwala zone, 2.6 Ma, Siwalik Frontal Range, northwestern India. Comptes Rendus Palevol.
