It just so happens that I’m preparing to teach a class on human interactions with large carnivores, so this has been on my mind a lot recently!

Relationships between humans and the creatures that sometimes prey on us are challenging to study, for a number of reasons. In part because acts of predation on humans break the (admittedly artificial/recent/largely Western) divide between “nature” and “culture,” any individual occurrence is both categorically disturbing and can be approached from a huge variety of disciplinary angles. These approaches sometimes complement one another, but they may also jar or lead to very different conclusions--in part because the nature of the evidence from each field is so different, and in part because the topic quickly becomes charged with an array of strong fears, beliefs,and preconceptions. In this answer, I’m going to cover some of the types of evidence that can be used to think about historical acts of predation on humans. Then, I’m going to attempt to focus in on the context you’re asking about specifically, and try to demonstrate why it’s difficult to get anything like a clear picture.

Paleontology demonstrates that humans and our hominid relatives have experienced predation throughout our evolutionary history. This is attested by an abundance of fossil remains that display the marks of predatory animals: eagle talons on the Taung child, an infant Australopithecus from c. 2.8 million years ago; hyena munch marks on the Homo erectus assemblages from the Zhoukoudian caves (very roughly 500,000 years ago, though the dates are contested). But the paleontological record is partial. As far as I know, neither bears nor wolves are directly implicated in predation on fossil hominids. However, this may be an absence of evidence rather than evidence of absence.

In seeking to recover the details of prehistoric human interactions with large predators, anthropologists have sometimes turned to modern indigenous groups that continue to live in close contact with these creatures. There are problems with this approach, of course--no modern indigenous groups perfectly replicate prehistoric societies, many live in different environments (often marginal and/or degraded) than those they inhabited even in the recent historical past, detailed death records can be hard to come by in these communities, and carnivore populations (and, arguably, behaviors) are different now than they were historically. Still, what evidence we do have suggests that people living in close proximity to large carnivores do occasionally become their prey. These events only rarely occur within towns or villages; they are more likely to happen in fields, pastures, woodlands, or transitional areas. Predators can target children, who are small and often uncautious; women, who are often smaller than men and sometimes engage in activities that can put them in dangerous environments (foraging in dense thickets, washing clothes near crocodile habitats, being around children, etc.); and men, who can be targeted while hunting (sometimes by the same animal that they are currently stalking) or while gathering wilderness products like wood, honey, or pearls. Of course, every society divides labor differently, and these descriptions aren’t meant to be essentializing--these are just the kinds of situations I’ve seen cited in the literature! Everyone is more in danger at night and/or while relieving themselves. Enclosed indoor toilets are a recent, and in places still rare, luxury.

These lines of inquiry also interface with ecological insights. The general impression from these studies is that animal predation on humans is sometimes linked to the predator’s inability to find wild prey, either because of factors like ecological collapse or epidemic diseases among ordinary prey species, or because of injuries that the individual predator suffers which make it less able to track and kill animal prey. When humans colonize areas that have previously had no or minimal human intervention, they can be subject to predation--either because they are curtailing predators’ ordinary food supplies, encroaching on predator territories, and/or because the animals are not habituated to human presence and do not have a developed fear of them.

However, as detailed as some of the data on these interactions is, it is overwhelmingly modern (predominantly 19th century and later). For earlier periods, we are left with--as you say--historical records, myths, and folklore. Historical records do occasionally record animal attacks on humans. However, there is a significant bias towards the upper strata of society in these accounts. So for instance, it’s recorded that King Favila of Asturias (reigned 737-739) was killed by a bear, probably while hunting. If he was indeed hunting the bear at the time, he was engaging in a highly dangerous, elite activity, and his death may have more in common with those of men slain by wild boars while hunting them, than with the victims of predation. Because those most in danger of routine predation tend to live in rural and/or marginal areas, their deaths are less likely to be recorded. This is particularly true in the premodern world.

Mythology and folklore are difficult historical sources for a host of reasons. In this particular case, issues include the often-wide gap between a story’s setting and its composition; the tendency of myths to focus on extraordinary rather than ordinary events; and the frequent invocation of magic and metaphor. So while there is a fair amount of wolf predation on humans in the Völsunga saga (written in late 13th century Iceland, set in a nebulous 5th century-ish north-central Europe), it is all carried out by werewolves. Werewolves can be understood as “the wolf within man,” and seen as metaphors for human violence and sadism; the case of Peter Stubbe, in late 16th century Germany, is exemplary here. But they may also reflect a tendency for humans to anthropomorphize animals that prey on people; to ascribe human cunning and malice to them, and to seek revenge on them in ways that actualize them as historical agents. One of the most-studied instances of canid predation on humans is the 100 or so deaths that occurred in the Gévaudan region of south-central France between 1764 and 1767; the victims were mostly children and adolescents tending sheep or cattle. Both contemporary chroniclers and later researchers (and novelists, folklorists, internet commentators, etc.) have sometimes seen human criminality behind these events, either in the form of supernatural shapeshifters or of serial killers acting “like” (or in the disguise of?) wolves. There’s no solid evidence that this was the case, but it demonstrates how even in comparatively well-documented cases, animal predation on humans brings up a complex tangle of questions.