The lethality factor
The "lethality factor" has a lot to do with people's attitudes and behaviors, and a bit less to do with whether they have access to poison, traps, or large-caliber firearms. People who are hostile to bears, intent upon doing them harm, and with a gun are perhaps the most lethal of all people to grizzlies--and the primary cause of brown bear extirpations worldwide.
Technically, "lethality" is the likelihood that a bear will end up dead as a result of encountering a human...the odds that death will occur given that an encounter has happened. As illustrated by the figure at left, the total number of grizzlies killed by humans is axiomatically a function of how often bears run into ("encounter") people and the odds that these encounters will be lethal to the involved bear. Both encounter rate and lethality can drive how many bears die, potentially in ways that allow for trade-offs. In other words, if people are benign, well-behaved, and unarmed (i.e., not particularly lethal), grizzlies will be able to persist even if they are encountering people at a high rate. This kind of circumstance typifies a National Parks. Conversely, if people have a bad attitude and are armed to act on it (i.e., highly lethal), a low rate of encounters, if sustained, can drive a population of grizzlies to extinction. Such was the case during the late 1800s and early 1900s in North America (see Extirpations).
The lethality of humans to grizzly bears is clearly driven by culture and cultural narratives. The ethos of domination and use and the accompanying narrative of Manifest Destiny that Europeans brought with them to North America is a prime example of a combination that was highly lethal to grizzlies. But there are other factors that as clearly drive lethality. An obvious one is the extent to which people experience personal losses from grizzly bears. Lost livestock or agricultural crops are a common source of aggrievement, with damage to crops (for example, oats) more common in Europe and Asia than in North America. Familial idosyncracies are also a variable. People raised in families that esteem hunting, including the hunting of bears, may kill bears, but not necessarily harbor overt feelings of animosity. And, finally, people who don't have access to firearms or poisons may not be particularly lethal, despite disliking or even loathing bears. It is easy to imagine such circumstances in authoritarian states that closely regulate possession of guns, or empoverished developing countries where people can't afford such luxuries.
More about lethality & what goes on in peoples' heads
The figure above contains some relatively concrete as well as some pretty abstract information, all of which pertains to a key driver of human lethality: The ways that people symbolically construct grizzly bears. I have concluded that symbolic constructions have a lot to do with how people imagine bears and not much to do with the real animals. At the extremes, grizzlies can either be demons or close relatives. Grizzlies can also be symbolically identified with the extent to which a person might loathe the federal government (if the federal government is charged with conservation of bears) and environmentalists who voice an alien and scary worldview (especially if they are promoting the protection of bears over exploitation and use of nature). Which can lead to the displacement of fear and hatred onto grizzlies, even when the real objects are other people. Which can lead to the symbolic discharge of accumulated angst through the act of killing a grizzly bear. In the case of grizzlies in the lower-48 states, such killing is illegal and goes under the billing of "maliceous."
Getting back to the figure above: The bars at top are marked to show the proportional causes of grizzly bears deaths in three different areas of northwestern Montana: the Cabinet-Yaak ecosystem, and the West and East sides of the Northern Continental Divide ecosystem. I've highlighted two causes. Orange denotes the proportion caused by hunters intent on legally killing black bears (Ursus americanus) who instead "mistakenly" killed a protected grizzly. The burgundy denotes the proportion attributable to maliceous killing--or poaching. The Cabinet-Yaak ecosystem is remarkable. The majority of known grizzly bear deaths are due to poaching and mistaken identifications. Most of the remainder (in gray) are bears dying from unknown causes, which is often a euphamism for poaching that was not conclusively documented. This starkly contrasts not only with the Northern Continental Divide ecosystem, but with other grizzly bear ecosystems as well. So, what's going on with the people living in the Cabinet-Yaak ecosystem?
I tested a potential answer to this question by looking at factors that might underlie peoples' attitudes and behaviors, especially as they relate to grizzly bears. These factors pertain to degree of political conservatism (the Obama vote), age, extent of poverty, lack of education, economic reliance on agriculture, and the extent to which timber harvests have declined in recent decades, all of which likely positively correlate with a hostile or intolerant attitude. Of these, lack of eduction, extent of poverty, and decline in timber harvests--along with crime rate--potentially indicate high levels of social distress. Social distress is relevant because of the extent to which the entailed angst is often displaced onto "alien others," of which grizzly bears, environmentalists, and the federal government are prime candidates. Political conservatism can also fuel this dynamic.
I assembled information on all of these factors for the counties encompassing each of the three northwestern Montana grizzly bear ecosystems, and then plotted each factor along its own axis in a figure called a radar diagram. I show the resulting diagrams in dark pink for each of the ecosystems above. A key point here is that I plotted the information for all of the factors so that a higher value was more likely to positively correlate with negative attitudes towards grizzlies. Which means that the larger the pink "rose"--or plotted constellation of values--the greater the likelihood that people, on average, harbor a negative attitude towards grizzlies. Notice how much larger the dark pink rose is for the Cabinet-Yaak compared to the East and West sides of the Northern Continental Divide (parenthetically, I've reproduced the Cabinet-Yaak rose in light pink behind the other plots to highlight the contrast).
None of this is conclusive, it is probably no accident that the indicators of widespread social distress are so much more pronounced in the Cabinet-Yaak area than anywhere else. I can at least conclude from this that my speculations about symbolic drivers of human lethality to grizzlies are consistent with--or not contradicted by--the information that I assembled. Based on this I would tentatively conclude that people are quite lethal (on average) to grizzlies in the Cabinet-Yaak area, probably because of displaced unresolved angst organized around a demonic contruction of grizzlies. Which might partly explain why recovery of the grizzly bear population in this region has been so difficult.
Evidence of lethality effects
Regardless of what credence you might give my argument immediately above, the question still remains: Is there evidence of the importance of human lethality to the survival of grizzly bear populations? The rapid extirpations of grizzly bears from western North America by newly-arrived European is perhaps self-evident proof of a lethality effect. But there are other lines of evidence supporting the importance of human lethality to survival of brown and grizzly bear populations.
The map at right shows a result that is germane to this issue. The map was taken from an analysis that looked at spatial drivers of grizzly bear extirpations in the western US between 1850 and 1970. The various shades of green delineate areas within which grizzly bears had different odds of surviving between 1970 and 2000...but assuming that human lethality during this interval was the same as it was between 1920 and 1970. The darker the green the higher the odds of population persistence. Under this scenario, grizzlies would have survived in the Yellowstone region only in the heart of Yellowstone National Park and barely at all in places such as the Glacer National Park and the Bob Marshall Wilderness Area in the Northern Continental Divide. The basic idea is that this map resulted from projecting historical drivers of extirpation foreward in time, most notably human population sizes--without any allowance for whether people were less likely to kill grizzlies.
But the assumption of constant lethality clearly failed. Grizzlies fared much better during 1970-2000 that was predicted. Some factor had changed that was not considered by the model. And, in fact, such was the case. Grizzly bears in this part of the world were afforded protection of the U.S. Endangered Species Act. These protections led to the end of sport hunting, the removal of garbage and other foods that might attract grizzlies to people, and even the ending of livestock grazing in places. Penalties for illegally killing bears also increased along with oversight of human activities that might otherwise harm grizzlies. All of this led to a substantial reduction of human lethality and a stabillization and even increase of grizzly bear populations, resulting in grizzlies in many more places than was predicted. Parenthetically, the Selway-Bitterroot area of central Idaho requires some explanation. As it turns out, grizzlies were extirpated in this ecosystem well before 1970, largely because the availability of spawning salmon concentrated grizzlies along rivers in the lowlands where they were highly vulnerable to humans--and despite the existence of extensive wilderness in the surrounding uplands (see Extirpations). However, according to the argument I offer here, grizzlies might have fared quite well between 1970 and 2000 if they had survived to experience the benefits of reduced human lethality.
Additional evidence for the effect of human lethality comes from looking at differences between Eurasia and North America in the joint distributions of humans and Ursus arctos (of which brown and grizzly bears are genetic variants).
The map at left shows the more-or-less current distribution of people and grizzly/brown bears, differentiating areas by whether brown bears are present and by human densities. Areas with no bears and high, moderate, or low densities of people are denoted, respectively, by dark burgundy, light burgundy, and gray. Green denotes areas with brown and grizzly bears; dark green indicates the presence of few people, light green the presence of moderate to high human densities.
The regions of this map that are most relevant to the topic of human lethality are those colored light green--where we jointly have Ursus arctos along with non-trivial numbers of people. Notice the virtual absence of such areas in North America in contrast to the extent of this overlap in Eurasia. The point being: there are large parts of Eurasia where brown bears have survived despite being around lots of people, unlike in North America, where essentially any exposure to people has apparently resulted in the extirpation of grizzlies. Which begs the question: Why?
I strongly suspect that this difference between the continents is largely a result of differences in how lethal humans were to bears during the last two centuries. Europeans arrived and spread like a lethal tidal wave in North America, descimating pretty much all life on the continent. The ethos they brought was especially destructive. By contrast, I suspect that brown bears and people were better able to work things out over a sustained period of time in Eurasia, resulting in gradual behavioral and attitudinal adjustments on the part of both species. Hence, there are lingering, and in places increasing, areas of overlap of relatively dense human populations with brown bear range, probably as a direct result of lesser lethality on the part of the involved people. This for an amalgam of speculative reasons that I describe above as drivers of human deadliness.