It’s early April 2015, and John Marzluff and I are standing with a film crew attempting to capture some footage of a crow funeral to compliment a story they are working on about Gabi Mann. I’ve already set the dead crow on the ground, it’s placed just out from a cherry tree resplendent in springtime blossoms. After only a few moments of waiting, the first crow arrives and alights on the tree, its head cocking around to get a better look at the lifeless black feathers beneath it. I hold my breath for the first alarm call, ready for the explosion of sound and the swarm of birds that will follow it. But it doesn’t come. Instead, the bird descends to the ground and approaches the dead body. My brow knits together in surprise but, ah well, I think, the shots of it getting so close and then alarm calling will make good footage. The audience will have no questions about what it is responding to. To my continued surprise, however, the silence persists; only now the crow has drooped its wings, erected its tail, and is approaching in full strut. No, no, this can’t be, I think. But then it happens. A quick hop, and the live crow mounts our dead one, thrashing in that unmistakable manner. “Is it giving it CPR?” someone asks earnestly. Still in disbelief, John and I exchange glances before shaking our heads and leaving the word “copulation” to hang awkwardly in the air. After a few seconds another bird arrives to the cherry tree and explodes in alarm calls, sending our first bird into its own fit of alarm, followed by a more typical mobbing scene. The details of what I’ve just witnessed as still washing over me when I hear John lean over to me…”You need to start your field season tomorrow.”
What crows do around dead crows is something I’ve dedicated much of my academic life to understanding. In the course of my first study, my findings made for a nice clear narrative: crows alarm call and gather around dead crows as a way of learning about dangerous places and new predators. Although there are other hypotheses we can’t rule out, certainly danger avoidance is at least partially driving this behavior. An important detail of that original study though, is that because of the way it was designed, with a dangerous entity always near the dead crow, our live crows were never in a position to ever get very close to our dead stimulus. So the possibility that they do other things around dead crows, like touching them, couldn’t be explored.
It’s been 3 years since that day in April and during that time it has taken every ounce of my power to remain tight lipped when journalists would ask “what’s the most interesting thing you’ve learned from your studies?” Because until we were able to scientifically vet the prevalence of this behavior, I wasn’t willing to say much about it for fear of making necrophilia mountains out of mole hills. But with our findings now officially available in the journal Philosophical Transactions B, I am delighted to finally share what has been the most curious secret of my PhD: crows sometimes touch, attack, and even copulate with dead crows.
Although this statement is jarring in its own right, what really gives it power is that we know this not just from that first fateful day with the film crew, but through an experimental study testing the response of hundreds of birds over several years. That’s important because it allows us to say not just what they’re doing but possibly why they’re doing it (and at least why they’re not doing it). So how did we conduct this experiment?
First, I dove into the literature to try and see if there was any precedent for this kind of behavior in other animals. Although there have been no systematic studies, repeated observations of animals touching, harming, even copulating with their dead occur in dolphins, elephants, whales, and many kinds of primates, among some other animals. Based on this, we hypothesized that this behavior may arise from: attempts to eat it, attempts to learn from it, or a misuse of an adaptive response (like territoriality, care taking, mate guarding, etc.). To test these ideas I searched the neighborhoods of Seattle until I found a breeding adult pair and (while they weren’t looking) presented one of four stimulus options: An unfamiliar dead adult crow, an unfamiliar dead juvenile crow, a dead pigeon or a dead squirrel. The latter two stimuli being key in helping us determine if the behavior was food motivated, whereas the nature and prevalence of the interactions themselves (common, uncommon, exploratory, aggressive, sexual) helped us address the other hypotheses. In all, I tested 309 individual pairs of crows; or in other words, once again I freaked out a lot of Seattle residents wondering why there was a woman with a camera, binoculars, and some dead animals loitering in front of their house for long periods of time.
Our main findings are that crows touched the animals we would expect them to eat (pigeons and squirrels) more than the dead crows, and although crows sometimes make contact with dead crows, it’s not a characteristic way they respond. Because this behavior is risky, this seems to back up previous studies in crows that suggest that they are primarily interested in dead crows as a way of self preservation and avoiding danger.
A crow tentatively pokes at one of our dead crows
That said, in nearly a quarter of cases, crows did make some kind of contact with dead crows. Like with mammals, we saw that these behavior could be exploratory, aggressive and in rare cases even sexual (about 4% of crow presentations resulted in attempted copulations), with the latter two behaviors being biased towards the beginning of the breeding season. Importantly, the latter two categories of interactions were rarely expressed independently, and it was often a mixture of the first two; in rare cases, all three. In the most dramatic examples, a crow would approach the dead crow while alarm calling, copulate with it, be joined in the sexual frenzy by its presumed mate, and then rip it into absolute shreds. I must have gone through a dozen dead crows over the course of the study, with some specimens only lasting through a single trial. It was an issue that may have been insurmountable if not for the donations of dead crows by local rehab facilities and the hard work of my long time crow tech turned taxidermist, Joel Williams.
It’s hard to witness this behavior without wondering if maybe the crows somehow don’t recognize that it’s dead and are instead responding like they might to a living intruder or to a potential mate. So we tested that idea too, by conducting a second experiment where we presented either a dead crow or a life-like crow mount. The differences in their response was clear. They dive bombed the “live” crows and less often formed mobs, just like we would expect them to do for an intruder. They also attempted to mate with the “live” birds but in these cases it was never paired with alarm calling or aggression. So the issue doesn’t seem to be that they think it’s alive.
The fact that this behavior was rare, and often a mix of contradictory behaviors like aggression and sex, seems to suggest that none of those hypotheses I outlined earlier are a good fit for this behavior. Instead, what we think happens is that during the breeding season, some birds simply can’t mediate a stimulus (the dead crow) that triggers different behaviors, so instead they respond with all of them. This may be because the crow is less experienced, or more aggressive, or has some neurological issue with suppressing inappropriate responses. Only more experiments will help us determine what makes this minority of birds unique, and whether expressing these seemingly dangerous behaviors are the mark of the bird that is more, or less reproductively successful in the long haul.
So while there’s still much more left to be explore here, I can finally say that this is without a doubt some of the most interesting behavior in crows I’ve ever witnessed. I hope you will check out the publication here, and seek out all the other amazing work being reported in this special thanatology (death science) themed issue.