Capuchin monkey (Sapajus libidinosus) stone flaking

If my academic career is remembered at all, it may well be because of a time I was wedged awkwardly among boulders, halfway up a hillside in the dry heart of Brazil. The day that I found out just how similar monkey tools could be to those of our ancestors.

The rest of my team had moved on, further up the narrow valley, searching for rock-art and hopefully some local wildlife. I’d stayed behind because, apart from my dislike of climbing high, steep slopes made entirely of loose pebbles and thorn scrub, I’d noticed a few stone pieces scattered over a small area of dry dirt and fallen leaves. They hadn’t been there long.

Those stones looked like they’d been deliberately, but inexpertly, broken. As if an ancient human had wandered by just before our survey team showed up, knapped out a few quick flakes and then shambled off again. Thing is, we were a few million years too late, and on the wrong continent, to meet our ancestors. What we did have in the area was wild capuchin monkeys.

And then, as I sat there trying to piece together the stone flakes and work out how they’d ended up that way, a few of the monkeys turned up. And showed me exactly how they’d done it.

A cracking sound

I was in the Oitenta region of Serra da Capivara National Park (SCNP), Piaui. The park is a puzzle of sandstone and conglomerate plateaus, cut by steep valleys that wind down to open plains. The particular set of boulders I was crouched among, watching the monkeys at work, was elevated about 20m above the plain to the south.

This part of Brazil has been inhabited for many thousands of years by modern humans. The vivid rock-art and dense archaeological sites created by those indigenous groups led in part to SCNP being added to the UNESCO World Heritage List in 1991. For millions of years before Homo sapiens showed up, however, capuchins were the region’s top primate. The two species have history: you can find capuchins (including a baby riding an adult’s back) among many other animals in the local rock art record. This photo is courtesy of my colleague and leading capuchin expert Tiago Falotico:

We don’t know exactly when capuchins started using stone tools in SCNP, but excavations that I led of their activity areas have shown that it goes back at least 3000 years. And I’m confident that when we search for deeper evidence we’ll find it extends much further back than that.

This simian affinity for rock technology didn't escape the attention of Brazilian scientists. In 2007 Antonio Moura of the Universidade Federal da Paraíba published a report of an unusual behaviour at SCNP, which he termed ‘stone banging’. During fieldwork from 2000-02, he saw that capuchins would sometimes pick up a heavy quartzite stone and bang it hard against a second one that was still attached to exposed conglomerate bedrock. The monkeys often did this while making threatening or aggressive motions, so Moura hypothesised that the capuchins were making a noise to deter him, or other predators, from getting too close. The banging did also occasionally cause loose stones to fall from the outcrop, posing a risk to following fieldworkers.

If you’re not exactly sure what a quartzite conglomerate looks like, here’s a photo I took of one with a couple of adventurous capuchins scaling it. Each capuchin is about the size of a regular house cat:

Loud displays make sense as a possible way to scare off predators—although whether or not it actually does we don’t know for sure— but Moura also noted that almost a third of the banging displays were apparently non-aggressive. Those were more puzzling. What were these drumming monkeys up to? Was this percussion just for fun, or attention, or were they after something more concrete? A possible clue came from behaviour that Moura saw immediately afterwards:

on 2 distinct occasions the alpha male (Boludo) hit a stone against a rock surface and stayed smelling the place.

Moura did his own banging experiments and found that the struck rock gave off a strong, sharp smell, like gunpowder. He didn't pursue this behaviour much further, and as an ethologist he was naturally less interested in the stones themselves than an archaeologist like me would be, when I showed up over a decade later. But that hot afternoon in late September 2014, it seems that stone smelling (and even stone licking) was what brought me and the monkeys together.

Lithic jigsaws

Let’s set the scene a little. In the first photo below, you can see two grey stones lying to the right of my 10cm scale. Around them are outcrops and the largely dry vegetation that characterises the area at that time of year. In the second photo I’m holding those two stone pieces together, and they fit perfectly. The smaller one had been struck neatly off one side (the left side in the picture) of the larger one. It wasn’t necessarily a single intentional or well-placed blow that caused the split, because the larger piece had lots of little whitish chip and scuff marks where it had been pounded against something hard. It just happens that one blow landed in the right place to fracture that big flake off the side:

There are a few reasons why this kind of break might happen, and rocks simply tumbling down a cliff thanks to gravity is one of them. So there was no definitive proof of monkey involvement. But literally three minutes after I took that second photo, I heard the sharp crack of stone-on-stone coming from behind a boulder just a few dozen metres from where I stood.

I crept as quietly as possible (i.e., not that quietly at all with dry leaves underfoot and a severely overloaded backpack) around the far side of the boulder. There, a male and a younger capuchin were happily banging away at an outcrop with their own handheld stones. They were wary of me—the capuchins at SCNP are habituated to human presence but will flee if they get spooked for any reason—but as I huddled myself down between the rocks they soon dismissed me as being anything of interest.

I had with me my little Sony handycam, as well as a simple Casio digital camera, and I alternated between filming and photographing the monkeys at work. On a low conglomerate ridge, each of the capuchins was smashing loose cobbles repeatedly down onto embedded, rounded rocks. They would then dip their head down and sniff or lick at the fine dust that was produced, even going as far as to put small slivers of broken rock into their mouth.

The journal Nature made a video summarising this work when we published it with them in 2016. The video goes beyond the behaviour into the stone flake side of things, and we’ll get to that soon, but it nicely shows the younger capuchin keenly licking away:

The fact is, we still don’t know why they’re ingesting the dust. It could be that it helps with internal parasites, or it provides some kind of nutrients, or even that the lichen growing on some of the cobbles has itself a benefit to the monkeys. We hope in future to be able to test monkey health against the times that they do this sniff/lick activity, but that’s a whole different project.

Breaking the rules

So what about those broken stones? By forcefully banging their handheld tools down onto raised cobbles, the monkeys are actually carrying out a kind of reverse of what our hominin ancestors did. Our ancestors would hold a heavier stone in one hand—the core—or steady it on a support, and then strike down with a hammerstone to detach sharp-edged flakes from the stationary core. Despite the fact that the monkeys ended up breaking their hammerstone rather than the thing they were hitting, the end result was the same. Flaked stone tools.

This is the best photo I got of the moment that a hammerstone broke. We used this one in our published report, but it always makes me so nervous with those tiny, white razor-blade slivers of stone flying up around the monkey’s eyes. None got in, thankfully.

Most of the stones they used were single cobbles of quartzite, although the older monkey also had a try at using a big chunk of broken-off conglomerate as a hammer too. He first selected it from those lying naturally around the site, and then gave it a full-bodied go:

The whole cracking event took only half an hour or so, before the monkeys moved away and up the cliff. I extracted myself from my hiding spot, and went over to survey the damage.

It was immediately clear that this wasn’t a one-off use of this particular site. Several of the embedded cobbles on top of that rock ridge had been worn down by strikes, each one removing a little more of the cobble and leaving behind small circular scars and pits. This one, which is about 6cm wide, is the same one that was hit by the monkey in the broken-stone photo above:

Can you see all the little chips and flakes of stone around the cobble’s base, as well as the lichen growing on its surface? One of those is the most plausible target of the monkey smash-and-sniff assault.

Many of those small flakes would be sufficient to cut through thin skin or scrape away the outer surface of a fruit. Yet we’ve never seen the monkeys use them for those or any other tasks. They’re not alone in that—no ape has used a sharp cutting tool in the wild either, although various captive animals have been trained to do so. While they make essentially the same rudimentary stone technology as a hominin some three million years ago, the SCNP capuchins don’t seem to have taken the next step and begun using those stone flakes.

To help analyse the stones, I brought hominin tool expert Tomos Proffitt into my Oxford team. Through painstaking work, Tom showed that the capuchin flakes had the same features you’d find on human-knapped stones, including distinctive characteristics of being struck by a stone hammer. He also went beyond my initial fitting together of some of the pieces to show that these flakes were broken off the capuchin hammers one after another, just as they did when our early ancestors gradually reduced a core into a series of sharp-edged tools.

Here’s one of Tom’s refitted hammer stones/cores, made up of multiple pieces that I retrieved from the site:

And here’s the same hammerstone as it was being deconstructed by one of the capuchins while I (and the juvenile) watched:

Stepping stones?

Tom and I, along with Tiago and the other members of my Primate Archaeology group, published this work in a paper we titled ‘Wild monkeys flake stone tools’. In part this title is the result of trying to use simple everyday words rather than technical ones (the whole title has the same number of syllables as ‘archaeological’), but it was also because that’s what the monkeys were doing. Their tools—the hammerstones—were being damaged seemingly as a way to extract something from the quartzite or its surroundings. And that damage was repeatedly producing stone pieces that have all the hallmarks of deliberate tool-making that we see in the earliest record of our lineage.

What that simple title misses, of course, is nuance. The capuchins aren’t deliberately making flakes to use as tools, which we know our ancestors were doing because they went on to leave cut marks from those tools on bones of the animals they ate. And while the technique the monkeys used closely resembles the ‘passive hammer’ style reported from the very earliest hominin tools that we currently have—from Lomekwi in Kenya dated to around 3.3 million years ago—the monkeys didn’t then pick up and take away the flakes they made. They thoughtfully left them behind for me.

It would be easy to suggest that capuchins are in a kind of proto-stone age, one step away from launching their technological careers into a human-like trajectory. Who knows, perhaps they are. But we don’t have evidence for that. What makes this capuchin work more useful, I think, is in broadening our imaginations for the kinds of activities that could have nudged our own lineage onto our particular pathway.

The idea that our ancestors might have started breaking stones because they liked the taste, or needed the nutrients, isn’t part of standard archaeological thinking. Yet the capuchins have created exactly the right conditions for an enterprising young monkey to start using cutting tools. The flakes are there, just waiting (well, I took a few away, but there are plenty more), and because they come back to the same site over and over again we can imagine how the monkeys might learn about sharp edges just through trial and error.

It’s even tempting to think that we could provide our own nudge to these industrious primates, perhaps giving the Oitenta monkeys something that needs cutting to get a reward, to try and coax them to begin using the flakes. But that would be poor science and ethically questionable, changing the natural behaviour of wild animals just for our own curiosity.

Besides, we’ve barely started to understand this activity. We know a bit about what they do and where, but we don’t know nearly enough about the who, why, and when. Changing the monkey behaviour, or attempting to, would not make sense when we barely know what’s happening in the first place. I can attest first-hand that what’s needed now is patience, a camera and a notebook, and a willingness to overlook cramp and insect bites as you wedge yourself between some boulders. And if you’re lucky, Nature will show up with just the right tools to surprise you.

Sources: Proffitt, T. et al. (2016) Wild monkeys flake stone tools. Nature 539:85–88. || Moura, A. (2007) Stone Banging by Wild Capuchin Monkeys: An Unusual Auditory Display. Folia Primatologica 78:36–45. || Falotico, T. et al. (2019) Three thousand years of wild capuchin stone tool use. Nature Ecology & Evolution 3, 1034–1038. || Harmand, S. et al. (2015) 3.3-million-year-old stone tools from Lomekwi 3, West Turkana, Kenya. Nature 521:310-315.

Main image credit: Michael Haslam || Second image credit: courtesy Tiago Falotico || All other images: Michael Haslam & Proffitt et al. (2016)