Were Native People Keystone Predators? A Continuous-Time Analysis of Wildlife Observations Made by Lewis and Clark in 1804-1806

Kay, Charles E. 2007. Were native people keystone predators? A continuous-time analysis of wildlife observations made by Lewis and Clark in 1804-1806. Canadian Field-Naturalist 121(1): 1–16.

Full text [here]

Note: the The Canadian Field-Naturalist is about a year and a half behind in actual publication, so although the journal issue in which this paper appeared is dated Jan.-March 2007, the publication was available only earlier this week. Full text with cover is [here]. The cover of the issue features the C.M. Russell painting entitled “When Blackfeet and Sioux Meet.”

Selected excerpts:

Abstract: It has long been claimed that native people were conservationists who had little or no impact on wildlife populations. More recently, though, it has been suggested that native people were keystone predators, who lacked any effective conservation strategies and instead routinely overexploited large mammal populations. To test these hypotheses, I performed a continuous time analysis of wildlife observations made by Lewis and Clark because their journals are often cited as an example of how western North America teemed with wildlife before that area was despoiled by advancing European civilization. This included Bison, Elk, Mule Deer, Whitetailed Deer, Blacktailed Deer, Moose, Pronghorn Antelope, Bighorn Sheep, Grizzly Bears, Black Bears, and Grey Wolves. I also recorded all occasions on which Lewis and Clark met native peoples. Those data show a strong inverse relationship between native people and wildlife. The only places Lewis and Clark reported an abundance of game were in aboriginal buffer zones between tribes at war, but even there, wildlife populations were predator, not food-limited. Bison, Grizzly Bears, Bighorn Sheep, Mule Deer, and Grey Wolves were seldom seen except in aboriginal buffer zones. Moose were most susceptible to aboriginal hunting followed by Bison and then Elk, while Whitetailed Deer had a more effective escape strategy. If it had not been for aboriginal buffer zones, Lewis and Clark would have found little wildlife anywhere in the West. Moreover, prior to the 1780 smallpox and other earlier epidemics that decimated native populations in advance of European contact, there were more aboriginal people and even less wildlife. The patterns observed by Lewis and Clark are consistent with optimal foraging theory and other evolutionary ecology predictions.

Introduction

It has long been postulated that native people were conservationists who had little or no impact on wildlife populations (e.g.; Speck 1913, 1939a, 1939b). Studies of modern hunter-gatherers, however, have found little evidence that native people purposefully employ conservation strategies (Alvard 1993, 1994, 1995, 1998a, 1998b; Hill and Hurtado 1996), while archaeological data suggest that prehistoric people routinely overexploited large-mammal populations (Broughton 1994a, 1994b, 1997; Jones and Hilderbrant 1995; Janetski 1997; Butler 2000; Chatters 2004). Elsewhere, I have proposed that native people were keystone predators, who once structured entire ecosystems (Kay 1994, 1995, 1997a, 1997b, 1998, 2002).

To test these competing hypotheses, I performed a continuous-time analysis of wildlife observations made by Lewis and Clark on their expedition across North America in 1804-1806 because their journals are often cited as an example of how the West teemed with wildlife before that area was despoiled by advancing European civilization (Botkin 1995, 2004; Patten 1998: 70; Wilkinson and Rauber 2002; Nie 2003: 1). Lewis and Clark were the first Europeans to traverse what eventually became the western United States, and many of the native peoples they met had never before encountered Europeans. In addition, historians universally agree that Lewis and Clark’s journals are not only among the earliest, but also the most detailed and accurate, especially regarding natural history observations (Burroughs 1961; Ronda 1984; Botkin 1995, 2004). Thus, the descriptions left by Lewis and Clark are thought by many to represent the “pristine” state of western ecosystems (Craighead 1998: 597; Patten 1998: 70; Wilkinson and Rauber 2002; Botkin 2004). Botkin (1995: 1), for instance, described Lewis and Clark’s journey as “the greatest wilderness trip ever recorded.” …


I developed three measures to quantify the wildlife observations recorded by Lewis and Clark in their original journals, which have recently been re-edited and republished (Moulton 1986, 1987a, 1987b, 1988, 1990, 1991, 1993 – hereafter cited only by volume and page). First, game seen. If Lewis and Clark reported old sign of a species, that was assigned a value of one, fresh sign a two, and if they actually saw the animal, a three. …

Second, game killed. On each day, Lewis and Clark recorded the exact number of animals that were killed to provision their party. …

Third, herd size. If Lewis and Clark reported sighting large numbers of a particular animal, a value of ten was assigned to that species on that day. A value of ten was also assigned if Lewis and Clark reported killing 10 or more of one species on a single day. I then added game seen, game killed, and herd size values for all species on each day to obtain a daily measure of wildlife abundance. Again, this was done for all 863 days of the expedition.

I also developed a similar convention to quantify the relative abundance of native people that Lewis and Clark encountered each day of their journey. …

Results

Lewis and Clark’s observations show an inverse relationship between wildlife and native people (Figure 1). Wildlife was abundant only where native people were absent, and if it had not been for the presence of aboriginal buffer zones between tribes at war (Hickerson 965; Steffian 1991; Martin and Szuter 1999a, 1999b, 2002, 2004; Farr 2001; Laliberte and Ripple 2003), there would have been little wildlife anywhere in the West.

FIGURE 1. Relationship between the abundance of native people and the abundance of wildlife as observed by Lewis and Clark in 1804-1806. Plotted are the mean daily abundance of all wildlife species and the mean daily abundance of native people by trip segments – segments 1, 2, 54, and 55 were excluded because those areas were near European settlements. Line fitted using a smoothing spline with cross validation (Mathsoft 1997: 158-167). X and Y axes are offset. Note that there are no data points in the upper right as might be expected if cultural beliefs fostered conservation. Clearly, it made little difference what native people believed, or said they believed. Instead, aboriginal hunting followed predictions derived from optimal foraging theory and other evolutionary ecology models.

Discussion: Optimal-foraging theory

According to optimal-foraging theory, high-ranked diet items are more susceptible to overexploitation than lower-ranked items (Smith 1983; Stephens and Krebs 1986; Smith and Winterhalder 1992; Butler 2000). Theoretical considerations and studies of modern huntergatherers both indicate that large mammals are the highest-ranked diet items, and that, in general, the larger the animal, the higher its rank (Smith and Winterhalder 1992; Hill and Hurtado 1996). Moreover, if risk to the hunter or travel distances are great, only the highest-ranked diet items should be pursued (Smith and Winterhalder 1992). Thus, optimal-foraging theory would predict that when native people entered aboriginal buffer zones, they should have concentrated their hunting on the larger species, such as Bison and Elk, causing those species to decline accordingly. This would also imply that native people lacked any effective conservation strategy regarding these prey items. This pattern was, in fact, observed by Lewis and Clark as they left various native peoples and entered buffer zones, first Whitetailed Deer increased, then Elk and then Bison. Conversely, as Lewis and Clark exited a buffer zone, Bison disappeared first, followed by Elk, while some Whitetailed Deer were usually able to escape native hunters (Figure 9). Furthermore, Lewis and Clark noted that native hunters preferred to kill female ungulates (3: 61, 270) due to that sex’s higher fat content. Now, killing females runs counter to any conservation strategy (Kay 1994, 1997b, 1998; Kay and Simmons 2002).

Alvard (1998b, 2002) recently reviewed the conditions under which evolution by natural selection might favor resource conservation by humans. In short, conservation will be favored by evolution only if the resource is economical to defend. For instance, if 1000 kcal are spent defending a resource, but less than 1000 kcal are derived from that resource, evolution will not favor conservation. For a variety of reasons, including competition from carnivore predators, large mammals were seldom, if ever, economical to defend (Kay 1994, 1998, 2002). Instead the logical, rational thing to do was to kill-out the large mammals as quickly as possible and then move on to other resources, which is exactly what aboriginal people did (Kay 1998, 2002). Counter-intuitively, once that was accomplished, native populations actually increased because people were forced to consume lower-ranked, but more abundant diet items (Hawkes 1991, 1992, 1993). There is also an evolved discount rate that acts to negate a wide range of possible conservation practices (Rogers 1991, 1994). …

Even within buffer zones, though, wildlife was not as abundant as one might think, because the animals were predator, not food, limited (Kay 1998, 2002). Food-limited ungulates invariably destroy berry-producing shrubs and woody riparian vegetation due to repeated browsing, and once willows (Salix spp.), cottonwoods (Populus spp.), and aspen (Populus tremuloides) decline, so do associated species like Beaver (Castor canadensis) (4: 189-190), which are dependent upon those plants for food (Kay 1998 and references therein; Nietvelt 2001). Lewis and Clark, however, reported that riparian thickets were common in buffer zones, as were Beaver and berry-producing shrubs (e.g.; 4: 70, 145-146, 189-190, 247, 278, 332, 374, 391-392, 399, 414, 419, 428, 435, 451; 5: 14, 42, 46, 59). In addition, Lewis and Clark noted that Whitetailed Deer often had twin fawns or triplets, and that even lactating deer were fat (4: 165), which would not have been physiologically possible if ungulate populations had been food-limited. Thus, carnivore predation and occasional hunting by native people (4: 232) kept buffer zone ungulate populations well below what the habitat could otherwise support (White et al. 1998; Kay 2002). …

A number of investigators have cited Lewis and Clark’s descriptions of abundant wildlife without realizing that those accounts only apply to the center of buffer zones (Craighead 1998; Wilkinson and Rauber 2002). … During pre-Columbian times, there may have been no more than 4000-5000 grizzlies in all of North America because the bears were simply large packages of fat meat that native hunters killed at will (Hallowell 1926: 31-37; Birkedal 1993). Similarly, there never were 60 million Bison on the Great Plains (Seton 1929; Roe 1951), as is commonly believed (Shaw 1995; Geist 1996; Kay 2002). …

It has long been known that native people in the Americas had no immunological resistance to European diseases, but only recently has it been learned that those diseases had a significant impact on native people prior to direct European contact (Dobyns 1983), or how this, in turn, caused abnormal increases in wildlife populations (Neumann 1985; Preston 1996, 1997, 2002; Kay 1998, 2002; Kay and Simmons 2002). European diseases, for instance, preceded Lewis and Clark. The smallpox epidemic of 1780 was especially devastating (Boyd 1985; Trimble 1985), and its aftermath was noted by Lewis and Clark (2: 478-482; 3: 285, 295, 311-312; 6: 81-82, 285, 308). In 1804-1806, Lewis and Clark found four Mandan villages along the middle Missouri but observed that there had been 12 prior to the 1780 epidemic. Similarly, Arikaras villages were reduced from 32 to 2 (Ahler et al. 1991: 57). Thus, if Lewis and Clark had journeyed west in 1775 instead of 1804-1806, they would have met more native people and correspondingly there would have been even less wildlife (Geist 1998: 4-5; Kay 1998, 2002). Furthermore, European diseases may have decimated native populations throughout western North America as early as 1550-1600 (Ramenofsky 1987; Campbell 1990; Kornfeld 1994: 198; Preston 1996, 1997, 2002), which suggests that pre-Columbian wildlife populations were likely much lower than even what Lewis and Clark experienced. Butler (2000), who studied resource depression in the Columbia Basin, reported that highranked diet items, such as ungulates, increased only after epidemic diseases decimated native populations ca. 1550. Similarly, Chatters (2004: 72-73) reported that Bison numbers increased only when native populations declined.

Conclusions

Contrary to prevailing paradigms (Lyman and Wolverton 2002; Moore 2002; Wilkinson and Rauber 2002), native people controlled the distribution, abundance, and behavior of wildlife, and large mammals were common only in boundary or buffer zones between warring tribes (Martin and Szuter 1999a, 1999b, 2002, 2004; Farr 2001). It is also clear that Lewis and Clark recognized this phenomenon, for Clark (8: 328) “observed that in the country between the [Indian] nations which are at war with each other the greatest numbers of wild animals are to be found.” This pattern can only be explained if native hunters pursued an optimal-foraging strategy and did not employ any effective conservation measures (Alvard 1998b, 2002). Only twice did Lewis and Clark report high wildlife values and encounter large numbers of native people on the same day. In both cases, native hunters were killing as many animals as possible (3: 176, 253-255). Moreover, Lewis and Clark were only able to complete their journey because of the food, horses, and above all else, knowledge that they received from native people. There were no unnamed streams, there were no unnamed mountains, and there was no wilderness (Kay and Simmons 2002). As noted by Lewis and Clark, the West was even more densely populated prior to the smallpox pandemic that decimated native people in 1780. …

[T]hese data support the hypothesis that native people were keystone predators, who once structured entire ecosystems (Kay 1998, 2002; Kay and Simmons 2002)… [E]cologists need to abandon the myth of once abundant wildlife and instead recognize that unhunted ungulate populations are outside the range of historical variability.

 
  • Colloquia

  • Commentary and News

  • Contact

  • Topics

  • Archives

  • Recent Posts

  • Meta