When did we start wearing clothes?

Human head lice live in hair. Clothing lice live and lay eggs in clothing, feeding on the body. They are two forms of one species, and the second cannot exist without the habit.

So the moment clothing lice diverged from head lice is a lower bound on when people began wearing clothes regularly enough to support a parasite specialised for it.

The approach goes back to a 2003 study of louse sequences, which dated the origin of the clothing louse to roughly 70,000 years ago.

Its authors read that estimate as tying clothing to the spread of modern humans out of Africa and into cooler regions, which the archaeology then available did not contradict.

That first study already found more diversity in African than in non-African lice. What changed was the date. Its authors corrected it to about 107,000 years in 2004, and a 2010 study using three nuclear genes and a different statistical model placed the split before the migration out of Africa, which made clothing look less like a response to leaving. The method survived; the story attached to the first number did not.

A 2010 study reanalysed the published sequences of four louse genes and put the divergence at least by 83,000 years ago and possibly as early as 170,000. Those are its words. In its own table 83,000 is the single most probable value and 170,000 the median, and the 95 per cent interval runs from 29,000 to 691,000 years.

The authors place the habit with anatomically modern humans in Africa, before they spread into colder latitudes, and note that the median falls at the start of an ice age, about 190,000 to 130,000 years ago.

Both studies calibrate their clock on the split between human and chimpanzee lice, taken to match the split between their hosts about five and a half million years ago, and the answer carries the uncertainty of that and of the mutation rate. The 2003 study put the origin at not more than about 72,000 years ago, give or take 42,000, before its correction; the 2010 interval is wider still.

There is also a disagreement about what is being dated. The dating studies model one split. A 2020 review by researchers in Marseille argues that clothing lice keep arising from head lice where infestations are heavy, so that the split is not a single event. The same review still reports the 2010 date as evidence of early clothing.

The result is therefore best read as an order of magnitude, and as evidence that clothing is old rather than as a date.

Tools that may have scraped hides appear about 780,000 years ago, though hides had other uses and scrapers could have served for meat. Bone tools from a cave in Morocco, dated 120,000 to 90,000 years ago, were probably used on leather and fur, beside carnivore bones with skinning marks, and their excavators read them as highly suggestive, if indirect, evidence of the earliest clothing. Eyed needles, which point to fitted garments, appear about 40,000 years ago.

Those mark different things: working a hide, and fitting a garment. The lice date something between the two, the regular wearing of clothes, and the Moroccan tools fall inside the range the 2010 study gives for it.