Rewilding Pragmatism

Or, What an African Safari Can Teach America


In recent years conservationists have proposed “rewilding” abandoned areas of North America and Europe with wooly mammoths, mastodons, and giant armadillos. But are such approaches realistic? In a new Breakthrough Journal piece, Martin Lewis argues that the decoupling of economic growth from natural resources in the US and other wealthy countries offers a big opportunity — but one that can only be realized through pragmatism. One model is South Africa's Kruger National Park. There, large populations of elephants, cheetahs, antelopes are increasing thanks to flexible management and good relations with neighboring communities. Could such an approach be adopted in California, where conservationists hope to reintroduce the grizzly bear? Photo Credit: David Strydom / Getty Images

Summer 2015 | Martin Lewis,

Perhaps it is no coincidence that at the same moment that scientists have concluded that we are now living in the Anthropocene, the age of humans, there has been a resurgence of interest in rewilding, the large-scale restoration of nature and the reintroduction of plants and animals (particularly large carnivores) by people to areas where they once thrived.

While the concept of rewilding is not new –– conservation leader Dave Foreman coined the term 20 years ago –– today, rewilding is a hot topic. In northern Siberia, one of the most extensive rewilding experiments is taking place in Pleistocene Park, where scientists are attempting to restore the area’s ecosystems to support vast populations of bison, wolves, and even mammoths. Stateside, environmental icon Stewart Brand leads the nonprofit Revive & Restore, which is attempting to bring back extinct species through genomic technology.

Rewilding occupies an unusual niche within contemporary environmentalism, a rare instance of pervasive optimism amidst a discourse otherwise dominated by elegies for nature lost and warnings of impending catastrophe. British journalist George Monbiot has championed rewilding as a beacon of positive environmentalism that “introduces hope where hope seemed absent.” Writing in Wilderness Comes Home, Bill McKibben observed that the prospect of rewilding has moved from “hazy hallucination” to “clear and prophetic vision.”1

While in part a reaction to the diminishing number and diversity of animal species — some estimates suggest that by the end of this century, up to half of the world’s species will go extinct — rewilding almost certainly reflects a recognition that in many places, rewilding is already occurring. Thanks to many decades of rising agricultural yields and the transition from wood to coal and other modern energy sources, many parts of the world have begun to reforest and leave more room for wild nature. Some of this recent rewilding has come through painstaking efforts on the part of conservationists, such as the reintroduction of the wolf to Yellowstone National Park.2 But much more of it has been spontaneous, such as the reestablishment of breeding populations of mountain lions in Nebraska and the Dakotas and the incipient spread of the species into the Ozark Plateau and other Midwestern forestlands.3

Unsurprisingly, different thinkers and scientists have competing and dissimilar visions of what the term rewilding entails and how it might be accomplished. Some advocates, such as the conservation biologist Michael Soulé, stress the creation of massive wilderness areas, in which virtually all aspects of modernity are erased from the landscape.4 Others, like Brand, are more concerned with wildlife restoration than with roads and other infrastructural intrusions.

Nonetheless, rewilding remains mostly undeveloped as a practical enterprise — more often advanced as a prophetic vision than as a set of actual proposals.5 While technological and economic developments that reduce the need to extract resources and energy from ecosystems offer the possibility for restoration of large-scale habitats and the reintroduction of locally extinct species, they offer no guarantee. Deep restoration based on the reintroduction of relatively complete faunal assemblages will require both concerted political action and prolonged cultural adjustments. This essay offers a pragmatic vision for rewilding American landscapes, eschewing purist dreams of fully restored pre-Anthropocene nature for more practical measures to restore wildlife diversity and abundance.


As a first principle, rewilded lands should not necessarily be mistaken for wilderness. The wilderness ideal, which has animated conservation efforts for many decades, now constitutes not simply roadless areas or even pristine zones barely touched by human activities, but rather the notion of self-willed lands,6 a definition supposedly rooted in the etymology of the word. The mission of Rewilding Europe, for instance, is to leave more room for nature to “do its job,” free from human interference and management.

Whatever its etymological justification, the concept of “self-willed land” is both deceptive and unhelpful. Even old-school conservationists recognize that wilderness requires some management, if only to extirpate invasive species and otherwise undo the effects of the previous human presence.7 Pervasive human impacts, which have molded landscapes in myriad ways, cannot simply be ignored or wished away. The notion of “self-willed land” in practice could even impede the preservation of biodiversity, because spatially limited habitats not infrequently experience fully natural local extinctions. Some human management of most natural areas is a necessity, if not a moral obligation.

Human management, in turn, necessarily entails debate and deliberation. In regard to active rewilding, controversies extend even to the taxonomic limits of potential reintroductions. It is one thing when a species has been locally wiped out but remains extant elsewhere, such as grizzlies in California or bison in Siberia, but quite another when the species is gone altogether. In such cases, should a closely related species in the same genus be viewed as a suitable substitute? What about an animal in a different genus of the same family that is nonetheless of similar appearance and occupies the same niche? Here the domains of taxonomy and ethics come into unaccustomed juxtaposition.

In the western United States, such controversies currently pivot around the status of the horse.8 At present, two wild members of the horse family (the Equidae) inhabit portions of the region, Equus ferus (mustangs, or wild horses) and Equus africanus (wild burros, or donkeys). In the mainstream eco-narrative, equids are exotic animals that did not exist in the pristine North American environment encountered by the first European interlopers; the wild horses and burros that currently roam some parts of the Western Intermontane region are instead mere feral escapees from the domesticated realm. We are thus told by many that wild horses should be removed or at least greatly reduced in number and range, as this would allow more room for indigenous species.9

But from a long-term perspective, horses are among the most indigenously North American of all large mammals, as most of their evolutionary history took place on this continent. Animals such as elk, in contrast, are relatively recent immigrants from Eurasia, having passed over the “land bridge” of Beringia as it opened up during glacial periods. Ironically, such newcomers survived the faunal debacle at the end of the last glacial period better than the more deeply rooted natives.10

Rewilding critics might object to the reestablishment of horses in North America on the grounds that the two feral equines are not the same species as those that were present in the Pleistocene epoch that ended some 12,000 years ago. But while the precise taxonomy of recently extinct North American equids is complex and controversial, recent genetic research indicates that the pan-North America Yukon horse, Equus lambei, would more appropriately be placed within the same species as the Eurasian horse, Equus ferus (the domesticated subspecies being Equus ferus caballus).11 As humans almost certainly gave the death-blow to North American horses a short time ago (in evolutionary terms),12 they should hardly be faulted for bringing them back; in contrast, this act might more rightfully be seen as one of redemption.

If we embrace a thoroughgoing evolutionary-environmental ethic, we would want to restore as much wildlife as possible that existed prior to the faunal debacle at the end of the Pleistocene. The ecosystems of North America were created through the coevolution of all living organisms over the course of millions of years. The relatively recent removal of most large animals threw those systems into disarray.

As a result, some lingering plant species have been slowly withering away over the past 12,000 years due to the deaths of their crucial partners, such as the mammoth- and sloth-disseminated Osage orange.13 The revival of evolutionarily intact ecosystems would thus require reintroductions on a massive scale. Crucially, if we want to see the revival of the patchy, complex vegetational arrays that were maintained through the relationship between plants and large herbivores, we would need the return of Proboscideans (animals in the elephant order).14

Accomplishing this program, however, would be anything but easy. It might be possible to resuscitate the wooly mammoth with current technology, but restoring the Columbian mammoth, the dominant Proboscidean over most of North America, is a different matter, given the absence of frozen carcasses.15

But even if the Columbian mammoth were to prove irrecoverable, the possibility of returning Proboscideans to temperate North America is not thereby lost. We could, for example, simply introduce the Asian elephant, an animal in a different yet closely related genus. Alternatively, after the successful revival of the wooly mammoth we could artificially select for sparser coats, greater stature, and so on, thus eventually breeding an animal that might as well be a Columbian mammoth. 

The possibility of introducing elephants to North America to fill a key vacant niche returns us to the ethical-taxonomic dilemma raised above in regard to the horse. Today’s horses and donkeys at least belong to the same genus as their extinct North American counterparts, but that is not the case in regard to the Asian elephant. Even though they belong to different genera, mammoths and Asian elephants still sit near each other on the taxonomic tree. In fact, the living Asian elephant is more closely related to all of the extinct North American mammoth species (genus Mammuthus) that it is to the two living African elephant species (genus Loxodonta). Asian elephants and mammoths also fulfill similar ecological functions.

Decisions about what species are appropriate for reintroduction and which are not ultimately require parsing taxonomic distinctions that involve making judgments rather than simply encountering the facts of nature. The deep rewilding of the United States and other countries thus always depends on contested timescales and taxonomic issues, extending well beyond the simple reintroduction of locally extinct species that were present when the first Europeans arrived.

The signal difference is that of timescale. What some would see as an example of pristine nature deserving of complete protection looks to others like a severely degraded landscape stripped of essential keystone species, which should be reintroduced forthwith, to the furthest extent possible. Rewilding proponents like Stewart Brand think in terms of evolutionary time, where spans of millions of years matter and in which a 10,000-year episode is a mere instant. More conventional conservationists and ecologists often look no further than the arrival of Europeans on foreign shores, regarding the environments that they encountered as existing in timeless equilibrium.

In the end, environmental opposition to the reintroduction of species that were not present at some recent and semi-arbitrary date, such as 1492, reveals a kind of crypto-creationist thinking, and one bespeaking “young-earth creationism” at that. Kirkpatrick Sale’s highly successful 1990 book, The Conquest of Paradise, perfectly encapsulates this view, depicting pre-Columbian North America as a literal paradise, one that “gave off the aspect of an untouched world, a prelapsarian Eden of astonishing plentitude” and in which nature functioned “to all intents and purposes in its original primal state.”16


Releasing either Asian elephants or revived and modified wooly mammoths into gargantuan American wildlife reserves would admittedly be an extreme maneuver. One must certainly recognize the existence of possible problems, both anticipated and unforeseen. Images of megafauna rampaging through the streets of San Francisco or Denver are easy enough to conjure. The extinction of the mammoth’s prime North American predators — Smilodon, a saber-toothed cat, and Homotherium, a scimitar-toothed cat,17 animals that are likely irrecoverable — raises the specter of long-term overpopulation as well.

Yet we manage similar conflicts reasonably well in some parts of the world, notably in South Africa’s game reserves — the premier example of which is Kruger National Park in the northeastern corner of the country. Kruger is vast, more than twice the size of Yellowstone, and its wildlife abounds. Although a few species are in trouble, many have seen significant population expansions over the past half-century.18 Although Kruger is an assiduously managed environment, it is managed for wildlife abundance and diversity, which sometimes entails a lessening of human interference. A recent decline in sable and roan antelopes, for example, has led to the experimental reduction of artificial waterholes, as these species are perhaps being outcompeted by thirstier herbivores.19 As of 1994, Kruger officials quit culling elephants, which are now increasingly common and perhaps even overabundant, in great contrast to most other areas of elephant habitat in the world today.20

But if Kruger is a wildlife haven, it is hardly a wilderness, crisscrossed as it is with roads and dotted by lodging facilities. Demand for accommodation much exceeds supply, as Kruger is immensely popular, especially among South Africa’s Afrikaner population. To get a hut (with basic amenities) at one of Kruger’s coveted bush camps generally requires predawn queuing a year to the day before one’s scheduled visit. South Africans’ passion for the park is essential for its perpetuation, ensuring governmental support and funds for wildlife monitoring, poaching interdiction, maintaining the perimeter fence, and so on.

Demand for accommodations within Kruger National Park much exceeds supply, and South Africans’ passion for the park is essential for its perpetuation, ensuring governmental support and funds for wildlife monitoring, poaching interdiction, maintaining perimeter fences, and so on. Photo Credit Kruger Park Bookings

Kruger is a national park, a public holding that is open to all — or at least all who can afford the trip and the relatively low entry fee. But adjoining Kruger are a number of private reserves, many of which were at one time marginally viable cattle ranches,21 that cater to an elite, global clientele. Wildlife is carefully managed in such reserves, and sedulously habituated to humans riding around in vehicles in order to make the observation of animal behavior easy and assured. Poaching, needless to say, is vigorously and fairly effectively opposed. But even though they typically cover tens of thousands of acres, such reserves are not large enough to be self-sustaining in regard to the most land-demanding species, such as wild dogs (Lycaon pictus).22 The solution is an ongoing movement to remove the fences separating such reserves from Kruger and from each other. Plans have also been hatched to create corridors linking the resulting “Kruger+” to nonadjacent reserves. The UNESCO-sponsored Kruger to Canyons Biosphere project located in the same general area would, if completed, cover an impressive 4.8 million hectares of land.

While the greater Kruger ecosystem is open to human observers, it is also carefully disconnected from the human economic and residential realm. Although some inner fences are coming down, barriers with the outer world are carefully maintained, because the system would collapse in the face of elephants marauding through nearby sugarcane fields and citrus groves, much less if lions were to start preying on cattle — and people. But within the barrier, a large and expanding zone of prime habitat under a combination of public and private management is relatively secure and highly popular.

If Kruger National Park is a wildlife haven, it is hardly a wilderness, crisscrossed as it is with roads and fences and dotted by lodging facilities. Adjoining Kruger are a number of private reserves, many of which were at one time marginally viable cattle ranches, that cater to an elite, global clientele. Wildlife is carefully managed in such reserves and habituated to human activity. Photo Credit CyberCapeTown

In a well-designed comparative study of wildlife conservation in South Africa and North America, Daniel Licht, Brian Kenner, and Daniel Roddy argue that the South African method is “more effective in conserving biodiversity,” largely because it relies on a more pragmatic approach to natural areas management.23 These authors lay out the signal differences between the two systems in concise form, noting that South African efforts have been: more focused on the conservation of wildlife and less on the preservation of scenery; much more commercial in orientation; more likely to reintroduce species, including apex predators; more reliant on a hands-on approach to management; much more willing to use boundary fences; more eager to pursue public-private partnerships; more willing to work with local communities, seeking their input and sharing with them revenues and other benefits; and more encouraging of ecotourism while simultaneously much more restrictive on the activities of tourists who venture into natural habitats.

A comparative study of wildlife conservation in South Africa and North America revealed that the former's efforts are much more focused on the wildlife abundance and hands-on management of species, such as the lilac-breasted roller and cheetahs above, rather than the preservation of scenery. Photo Credit Flickr User Arno Meintjes (left) and Flickr User Diriye Amey (right)

As the South African experience demonstrates, private land ownership does not run intrinsically counter to the restoration of nature on a truly grand scale. For such hopes to be realized in the United States, however, wildlife would have to gain a greater purchase on the public imagination than it currently holds; Americans, in other words, would have to become more like Afrikaners in this regard. Of course, it would help if the United States contained the megafaunal diversity of Kruger, a problem that deep rewilders are endeavoring to rectify.


Deep rewilding does not require that we forgo other rural landscapes that many people are deeply attached to, such as bucolic farmlands, vineyards with their associated wineries, and zones of boutique agricultural production. Even if an admittedly unlikely coalition of ecomodernists and conservation-minded property owners and developers could somehow successfully push forth new policies for large-scale wildlife restoration, farmlands on the metro fringes would not thereby revert to wildlife habitat. An increasingly prosperous, technologically oriented society will likely maintain and intensify its demand for rustic landscapes and craft-oriented agricultural and horticultural products.

We already see this dynamic playing out in the affluent and technologically sophisticated San Francisco Bay Area, where the emergence of distinct zones of agricultural production and nature preservation is occurring. Technologically intensive, high-yield cultivation dominates most of California’s Central Valley, one of the world’s premier agricultural heartlands. Closer to the urban core, especially in Napa, Sonoma, and northern Marin counties, artisanal farming in tourist-oriented landscapes has become more prevalent.

But elsewhere in the same general area, prime potential locations for deep rewilding remain widespread. As it happens, the Bay Area is ensconced within a set of steeply sloped hills and mountains, many of which are sparsely inhabited and are currently undergoing a degree of spontaneous rewilding. Owing to economic growth, intensification of agricultural and forest production, and changing consumer preferences, most of the terrain here is no longer needed for timber and beef production and can thus find a higher calling as wildlife habitat and hiking land. In these semiwild uplands, public parklands as well as natural habitats protected by The Nature Conservancy and other private organizations have dramatically increased in size and scope over the past several decades.

The habitats of these semiwild uplands are quite varied. To the south and west of Santa Clara County’s Silicon Valley, for example, lie the lush redwood-clad Santa Cruz Mountains, where mountain lions are no longer uncommon and black bears may be slowly reestablishing themselves. To the east lies the rugged, semi-arid, and seldom-visited Diablo Range, where tule elk, a key (sub)species of pre-Gold Rush California, have recently been reintroduced and are apparently expanding. Establishing a corridor between these two refuge zones is a high priority for wildlife advocates.24

The rugged, semi-arid, and seldom-visited Diablo Range – where tule elk now thrive after a successful reintroduction – is a high priority for wildlife advocates and could be the future home of the California grizzly. Photo Credit Martin Lewis

But reintroducing tule elk to the Diablo Range and allowing black bears to repopulate the Santa Cruz Mountains from their Big Sur redoubt are merely the first steps along the pathway to real rewilding. Essential is the reintroduction of the grizzly bear, which, after all, is the symbol of the state, even though California’s last grizzly was slaughtered in 1922. Every few years, proposals for returning this magnificent mammal to the state are mooted, but they are inevitably shot down, often on faulty ecological grounds.25

Unlike black bears, which are forest animals, grizzlies (“brown bears” to biologists) are primarily creatures of open country. This preference, based on feeding habits, meant that California’s grizzlies were most at home in the lowlands and hill country, rather than in the high mountains where most of the state’s national parks and wilderness areas are located. The oak-studded valleys and uplands of the Diablo Range would make fine grizzly habitat. The Diablo Range presents an admittedly harsh landscape in the summer — hot, dusty, and bone-dry — but it shows a much more gentle face during the rest of the year, particularly in the long and verdant spring. Yet few residents of the state give it any value or even acknowledge its existence, focusing their recreational and geo-aesthetic attention instead on the wetter western Coast Ranges and especially the Sierra Nevada. Perhaps if a large swath of Diablo Range were returned to the grizzly, attitudes would begin to change.

But even if the way could be cleared for grizzly reintroduction, opposition would persist. Many fear the return of this particular beast — and for good reason. In the productive oak woodlands of California, grizzlies can grow to a prodigious size. The largest recorded specimens weighed more than 2,000 pounds, more massive even than the salmon-fattened bruins of southern Alaska. Even some ardent wildlife enthusiasts would be wary of reintroducing an animal that can kill a person with a mere paw-swipe. And if rewilding continues beyond Ursus arctos, similar dangers from other species could be encountered in the future.

In the coming decades, will Californians embrace the reintroduction of the grizzly bear? Photo Credit Wikimedia

To accommodate the California grizzly, something equivalent to the Kruger fence would therefore be necessary, a barrier impassable to large, dangerous, and potentially destructive animals. Such an enclosure would separate an anthropocentric zone, where nature is valued but people are given priority, from a biocentric realm, where people can visit, observe, and when necessary, manage, but where biological abundance and diversity trump other considerations. If reasonable accommodations could be reached with ranchers and other local stakeholders, I see no reason why such a fence, some decades down the road, could not encircle most of the Diablo Range and eventually more than that, excluding some peripheral parks devoted to more risk-averse hikers and picnickers.

Imagine, then, a greater San Francisco Bay Area in the not-so-distant future, modeled around the ideas spelled out above. At the risk of gross simplification, its heartland would be composed of high-density cities (San Francisco, Oakland, San Jose) surrounded by a zone of microcities situated near transportation hubs set within a more traditional but still intensified suburban matrix. This urban zone could support a significant amount of wildlife, ranging from pigeon-hunting peregrine falcons nesting on high-rises to diverse assemblages of small mammals, birds, and invertebrates in backyard and apartment-complex gardens. Its prime purpose, however, would be to house people and their businesses and industries.

Most of the rural portions of the Central Valley, to the contrary, would remain devoted to high-intensity agriculture and agro-industry, although I would certainly hope to see its existing wildlife refuges expand, along with their wetlands. In the lowlands and gentle hills of the North Bay and Monterey Bay, an Arcadian landscape of upscale, artisanal farms and processing facilities might prevail. In Mendocino County, wineries and olive groves may well compete with chic cannabis gardens. In the more rugged hills, recreation-oriented wildlands with scattered dwellings would prevail. And perhaps, in the Diablo Range, we will be able to watch the California grizzly recolonize some of its old haunts.


Even in a rewilded world, there will always be consequences and trade-offs. The restoration of dangerous and potentially destructive animals, as we have seen, calls for the creation of firm wildlife barriers. Diverse and abundant wildlife would of course live on the human side of such a divide, but not species that routinely kill us or rampage through our fields, as lions and elephants are wont to do. In the megafaunal domain, on the other hand, people would be encouraged to visit and a few might even dwell within, although they would be advised to surround their homes with electrified fences like those in the private game reserves near Kruger. Humans would also have to actively manage such wildlife lands, particularly in the initial stages of restoration.

Beyond the knotty questions of which species we reintroduce, communities will have to grapple with just how much wildlife they want. How many bison or grizzlies or elephants constitute a healthy population? How large should wildlife reserves ideally be? Certainly the dream shared by ecomodernists and traditional conservationists alike is one in which wild places cover a substantial portion of the earth. It is also widely acknowledged that they should eventually be interconnected, allowing the free dispersion of genetic material, as emphasized by Michael Soulé and other conservation biologists. My own desire for the distant future is a 50-50 split: half of the world, and half of its potential primary productivity, for humankind; and half for nature without us. But determining how much we rewild and how much nature we spare is outside the bounds of this essay.

In the end, successful, broad-based rewilding probably requires the abandonment of the idea that it should be imposed on the basis of grand schemes concocted by conservation biologists or government agencies. Such schemes can certainly play a role in rousing the imagination, but not as blueprints that we are compelled to follow. In the more ad hoc, from-the-ground-up vision presented here, the stewardship of wildlife and wild areas would necessarily vary tremendously from place to place, reflecting local cultural predilections as well as demographic and economic realities. In large wilderness preserves located in remote areas with few inhabitants, such as those of Alaska’s North Slope, management could be quite light, whereas in those existing closer to dense human settlements, such as California’s Diablo Range, it would perforce be more intensive.

A pragmatic vision of rewilding not only embraces the fact that there is no singular Nature to recreate or revive. It is for this reason that bottom-up methods are better suited to accommodate the diverse needs and desires of multiple constituencies. In the developing world in particular, people living in areas proposed for natural restoration must be encouraged to engage in the process and advocate policies that would allow them to retain their homes and enhance their livelihoods. Otherwise, they can easily become the victims of nature preservation.

Despite the continuing and often vociferous debates between old-school conservationists and ecomodernists, we don’t have to choose between the “wilderness” of the traditional green imagination and the “domesticated garden” that is the supposed desideratum of the new school. A pragmatic vision of rewilding understands how both would firmly have their place in a good Anthropocene. /


1. Bill McKibben, “Epilogue,” in Wilderness Comes Home: Rewilding the Northeast, ed. Christopher McGrory Klyza (Hanover: Middlebury College Press, 2001) 275−278.

2. Douglas Smith and Gary Ferguson, Decade of the Wolf: Returning The Wild To Yellowstone (Guilford, CT: Lyons Press, 2012).

3. Michelle A. LaRue, et al., “Cougars are recolonizing the Midwest: Analysis of cougar confirmations during 1990-2008,” Journal of Wildlife Management 76, no. 7 (2012).

4. Michael E. Soulé and John Terborgh (eds.) Continental Conservation: Scientific Foundations of Regional Reserve Networks (Washington, DC: Island Press, 1999).

5. For a range of views on the subject, see Foreman 2004; Fraser 2009; Klyza 2001; MacKinnon 2013; Manning 2009; Martin 2005; Monbiot 2004; and Soulé and Terborgh 1999.

6. See usage in George Wuerthner, et al, (eds.) Keeping the Wild: Against the Domestication of Earth (Washington, DC: Island Press, 2014), 7, 34−35, 40, 164, 184, 186, and 199.

7. For example, David Johns, “With Friends Like These, Wilderness and Biodiversity Do Not Need Enemies,” in Keeping the Wild: Against the Domestication of Earth, eds. George Wuerthner, et al. (Washington, DC: Island Press, 2014), 31−44; Ned Hettinger, “Valuing Naturalness In the ‘Anthropocene,’” in Keeping the Wild: Against the Domestication of Earth, eds. George Wuerthner, et al. (Washington, DC: Island Press, 2014), 174−182; Reed Noss, et al., “Core Areas: Where Nature Reigns,” in Continental Conservation: Scientific Foundations of Regional Reserve Networks, eds. Michael Soulé and John Terborgh (Washington DC: Island Press, 1999), 99−128.  

8. Deanne Stillman, Mustang: The Saga of the Wild Horse in the American West (New York: Mariner Books, 2008).

9. The Wildlife Society, “Feral Horses: Get the Facts,”

10. Paul S. Martin and Richard G. Klein, eds. Quaternary Extinctions: A Prehistoric Revolution (Tucson: University of Arizona Press, 1989).

11. Jay Kirkpatrick and Patricia Fazio, “The Surprising History of America’s Wild Horses.” Live Science, July 24, 2008,

12. Paul S. Martin, Twilight of the Mammoths: Ice Age Extinctions and the Rewilding of North America (Berkeley: University of California Press, 2005).

13. Connie Barlow, Connie, The Ghosts of Evolution: Nonsensical Fruit, Missing Partners, and Other Ecological Anachronisms (New York: Basic Books, 2000). 

14. Martin, 2005.

15. Woolly mammoths and Columbian mammoths were so closely related that they were able to hybridize; see

16. Kirkpatrick Sale, The Conquest of Paradise: Christopher Columbus and the Columbian Legacy (New York: Knopf, 1990), 316.

17. One study of Smilodon teeth preserved in the La Brea Tar Pits of Los Angeles, however, suggests that camels and bison were the cat’s main prey. 

18. Johan T. du Toit, Kevin H. Rogers, and Harry C. Biggs (eds.), The Kruger Experience: Ecology And Management Of Savanna Heterogeneity (Washington, DC: Island Press, 2003), 332−248. 

19. Rhidian Harrington, et al., “Establishing the Causes of the Roan Antelope Decline in the Kruger National Park, South Africa,” Biological Conservation 90, no. 1 (1999) 69−78. 

20. Ian J. Whyte, Rudi J. van Aarde, and Stuart L. Pimm, “Kruger’s Elephant Population: Its Size and Consequence for Ecosystem Heterogeneity,” in The Kruger Experience: Ecology And Management Of Savanna Heterogeneity, eds. Johan du Toit, Kevin H. Rogers, and Harry C. Biggs (Washington, DC: Island Press, 2003), 332−248. 

21. Jenny A. Cousins, Jon P. Sadler, and James Evans, “Exploring the Role of Private Wildlife Ranching as a Conservation Tool in South Africa: Stakeholder Perspectives,” Ecology and Society 13, no. 2 (2008), 43.

22. Some private game reserves have reintroduced wild dogs, but populations are difficult to maintain; see

23. Daniel S. Licht, Brian C. Kenner, and Daniel E. Roddy, “A Comparison of the South African and United States Models of Natural Areas Management,” ISRN Biodiversity vol. 2014 (2014) 1−7.

24. See Julie Phillips, et al., “Safe Passage for Coyote Valley,” DeAnza College, Environmental Studies Department, 2012,

25. See Julie Cart, “California: The Next Grizzly Habitat? Some Want to See It Happen,” LA Times, Aug 2, 2014,; Carlos Carroll, et al., “The Return of the Wolf, Grizzly Bear and Wolverine to Oregon and California: Biologically Feasible?” in Large Mammal Restoration: Ecological and Sociological Challenges in the 21st Century, eds. David Maehr, Reed F. Noss, and Jeffery L. Larkin (Washington, DC: Island Press, 2001), 25−46.



Martin Lewis is senior lecturer in the department of history at Stanford University. He is the coauthor of The Myth of Continents (University of California Press, 1997).




by Jesse Ausubel



by Mark Sagoff