As we turn toward the future, we can’t overlook the blank expanses on the map, where an old time cartographer might have inked in dragons or other symbols of the unknown. Today, pretty much every corner of Earth has been explored, surveyed, and imaged from space, so those blanks might seem more boring than mysterious. They’re the places where few or no people live. I’ve argued that cultural and technical innovations in cities have fueled the accelerating pace of social change; this, too, can tempt us to focus exclusively on urban life, where we’re presumably getting a glimpse of things to come. Yet as we’ll see, the changing countryside is its own kind of science fiction.

Let’s examine the term “countryside” more carefully. I’ve used the word in an indiscriminate way to refer to something like 95% of our planet’s landmass—everywhere that isn’t brightly lit on the NASA Black Marble view of Earth’s night side. In rough order of decreasing human occupancy, those darker places can be broken down like this:

• Villages and towns. These are the small human settlements that haven’t, and in general won’t, tip into the kind of exponential growth and “ignition” that results in cities.

• Industrial zones, including ports, pipelines and railways, warehouses, power plants, sewage treatment plants, dams, reservoirs, oil wells, mines, and factories. These make up the globe-spanning technical life-support infrastructure cities require.

• Farms, orchards, and plantations. These are the expansive cultivated areas that supply cities with most of their food.

• Wilderness. Although no part of Earth’s surface is free of humanity’s effects, there are large areas of uncultivated land where nobody lives or works full-time. These include mountains and deserts, forests, jungles, national parks, prairies and scrublands, swamps and backwoods. The oceans, also, mostly fit into this category.

City dwellers like me tend to have mental pictures of such places, sometimes dating back to childhood. I grew up in Mexico City in the 1970s and early ’80s, and didn’t see much of the countryside as a kid, but I do remember being deeply impressed by a collection of stereoscopic View-Master slides mounted on cardboard reels, a kind of early all-analog VR developed by the US’s largest producer of scenic postcards. Holding the bright red View-Master up to a window for backlight and peering into its plastic lenses, I gawked at commanding views of places with iconic faraway names, like Bryce Canyon and the Grand Tetons; also, the majestic Hoover Dam, a cattle ranch, and a beautifully turned out Main Street in small town America.

The View-Master reels were highly curated, of course, and might even have had a whiff of Cold War propaganda about them. Still, those were real photos of real places. If you don’t live there, it’s easy to imagine that such antique images of life outside the big city remain accurate today. After all, we’ve seen that people living in rural places tend to be older and more conservative—I’ve argued that they’re in this sense “living in the past”—and this seems consistent with the postcard pictures: old-timey and changeless.

In reality, life outside the city is utterly different now, and many rural residents are deeply uncomfortable with a lot of the changes they’re seeing; hence the desire for those changes to stop, and even for the clock to turn back. It’s not just about an opioid crisis, an economic downturn, or other “weather phenomena” of the moment. The changes are immense and systemic, climate-like (indeed, some are climate-driven). Renowned architect and urban theorist Rem Koolhaas has put it like this: “The countryside is now the frontline of transformation […] more volatile than the most accelerated city.” ¹

Let’s turn over these postcards one by one and consider real life there today, beginning with villages and towns. As we’ve seen, the countryside is emptying out as young people move to the city (or suburbs near the city) for education, opportunity, and community, leaving behind an aging, shrinking, and increasingly homogeneous rural population. While some small towns remain vibrant (often by thriving on tourism, preserving and capitalizing on the “postcard look”), many more are shrinking and ultimately turning into ghost towns and ruins. Moss covers roofs, animals nest in attics, mold peels back wallpaper, tree roots twist their way through floorboards, blackberry brambles pull down walls. Life goes on, but without people.

This process is well underway in most countries with advanced economies, where urbanization and declining birth rates combine to erase smaller settlements. As of 2021, 6,000 villages lay abandoned in Italy. Another 15,000 Italian villages had lost more than 95% of their residents. By the time you read this, the numbers will be higher. An entire medieval hamlet, Pratariccia, was sold on eBay for $3.1 million in 2012, but more recently, asking prices for abandoned villages have been far more modest. It’s a buyer’s market—and even in the picturesque Italian countryside, few people are interested in buying. ²

In fact, the price has probably gone negative. In 2017, the desperate mayor of Bormida, a small town in Liguria, posted to Facebook apparently offering $2,100 to anyone willing to move there, though he beat a hasty retreat when 17,000 inquiries flooded into his inbox; lots of people were strapped for cash, but so, apparently, was the regional government. ³ So, the town’s demographic collapse has continued unabated. As of 2022, Bormida is home to 12 people over 90 years old, but only 10 young people between the ages of 10 and 19. ⁴

Japan, similarly, has an epidemic of akiya (空き家), or abandoned homes, and has set up special banks to try to sell them off, but supply is far outpacing demand. The phenomenon began with the abandonment of old-fashioned country houses of the Spirited Away variety, ⁵ but given Japan’s exceptionally low birth rate, aging population, low immigration, and already maxed-out urbanization, abandonment is spreading into higher-density places too. Whole islands have been abandoned, some of which, like Hashima, off the coast of Nagasaki, have become tourist destinations—though such sights are increasingly commonplace. Even in Tokyo, more than 10% of the housing stock stands empty today. By 2033, the Nomura Research Institute projects that about one third of Japanese houses will be uninhabited, and by 2040, the country’s unclaimed land will reach 28,000 square miles—roughly the area of Ireland. ⁶ The process will speed up over time.

There’s an undeniable post-apocalyptic romance in this abandonment and decay. We’re not the first generation to notice it. The French romantic painter Hubert Robert (1733–1808) loved this motif so much that his friends nicknamed him Robert des ruines. His paintings featured peasants grazing cattle among imaginary Roman ruins, as well as the imagined future ruins of real places, like the grand gallery of the Louvre. Just as the artistic contemplation of a skull gives us the spooky, pleasurable shiver of memento mori (“remember Death”), images like Robert’s remind us that all things must pass, and there’s a special grandeur in the passing of great things—monumental galleries, cities, entire lost civilizations. ⁷

Now, let’s turn over the farm postcard. Many smaller family farms have succumbed to similar forces, and are now rewilding: reverting to forest, scrubland, or prairie. Where the land was agriculturally marginal to begin with, or the soil degraded, the return to nature isn’t always quick, and the species mix may not look much like it did before farming, but it’s a return to nature nonetheless.

Economies of scale mean that the total area devoted to hog farming is much smaller than it used to be, though. The “factory farming” practices on many large farms are both cruel and unhealthy, but even if they all went free-range, the area saved relative to millions of family operations would still be vast—as are the savings in other resources. In effect, pigs have urbanized, just as humans have.

A similar pattern of consolidation of larger farms and decreasing total agricultural area over the past 20 years holds not only in the US, but also in the industrialized economies of most rich countries. Yet all this consolidation has happened at the same time as an overall increase in population size. Cities today need more food than ever, but the need is being met with a decreasing amount of land under cultivation. Since 1982, the land used for farming in the US has decreased by an area equivalent to Washington State. ⁹ In fact, we appear to have reached the worldwide peak in our total demand for agricultural land; even in Africa, one of the last remaining regions where the area under cultivation is still rising, its growth has slowed, and looks likely to start falling within the next few years, well before the continent’s population peaks.

This is a remarkable turnaround. When children’s book author Bill Peet wrote Farewell to Shady Glade to build ecological awareness among young readers in 1966, cities and factories were pouring untreated effluent into streams, and earth movers were plowing places like Shady Glade into the ground, though by that point, usually to build new suburbs—the area under cultivation in the US had been declining since 1950.

“Robots taking jobs” still seems like a Jetsonian fantasy in the city. I certainly don’t see much evidence of a robot invasion in Seattle. It’s true that slowly, over decades, ATMs have taken over certain tasks that bank tellers used to do, and the self-checkout scanners at Safeway seem to be processing quite a few customers with “15 items or less.” On the other hand, Amazon has decided to close its 68 experimental fully automated shops, where you used to be able to grab items and just walk out, while unblinking robotic eyes added charges to your Prime account. ¹⁰ Undoubtedly these technologies will continue to advance and eventually become commonplace (indeed, the same technology is now showing up in Amazon Fresh stores), but the changes seem modest and gradual.

As Koolhaas notes, though, the frontline of the transformation is the countryside, not the city. The reason is simple: so far, robots are still not very good at interacting with people, and cities are all about interactions with people. That’s why, even at the self-checkout line in the supermarket, the biometric security checkpoint at the airport, the computerized conveyor belt sushi restaurant, or the automated check-in kiosk at a trendy “robotic hotel,” ¹¹ there are employees standing by to help when—inevitably—the human-machine interaction goes awry. By contrast, the large-scale infrastructure in the countryside needed to support humanity today is increasingly robotic and hands-free—to the degree that its machinic nature makes much of the land humans directly depend on, paradoxically, a “human exclusion zone.” That’s one reason most of us are unaware of the dramatic changes. We’re simply not around to see them. ¹²

While most factories are still early in their transition to these kinds of practices, the container ship ports that global commerce relies on so heavily are well on their way to becoming fully robotic, surrounded by chain link fencing that excludes everyone save a handful of technicians and security guards. ¹⁴ The geometric, Lego-like landscapes of automated ports are a far cry from the bustling anthills of human activity characterizing traditional ports, from antiquity through most of the 20th century. Indeed, in centuries past, such ports, along with the businesses that sprang up to cater to their sailors, longshoremen, traders, and merchants, seeded many of today’s coastal cities. Now, the connection between trade and cities has become abstract.

Increasingly intensive automation extends to the vessels carrying the cargo, too. The first container ship, the Clifford J. Rogers, completed in 1955, was crewed by 15 people and could carry the equivalent of roughly 65–70 modern shipping containers or “TEUs” (twenty-foot equivalent units). The Ever Alot, a 2022 container ship, carries 24,004 TEUs, and is crewed by 25 people—a more than two hundred fold decrease in the need for human labor. ¹⁵ Much the same is true of wind farms, oil pipelines, mines… all of these basic industries are undergoing a rapid transition to full automation, or something close to it.

Even information processing is becoming increasingly automated. Giant datacenters of the kind run by Google, Microsoft, and Amazon (they used to be called “server farms”) have pretty much been human exclusion zones for years, with a lone technician going in now and then to swap out a broken part. In 2011, Apple, for instance, built a billion dollar datacenter in Maiden, North Carolina, once a thriving manufacturing town. The community, initially overjoyed at the prospect of a big new employer, soon learned that only 50 full time employees would be hired. Residents “[couldn’t] comprehend how expensive facilities stretching across hundreds of acres [could] create so few jobs,” as the Washington Post put it; “[I]n the newer digital economy, capital investments that a generation ago would have created thousands of new positions often equal only a handful today, with computers and software processing the heavy lifting […].” ¹⁶ Since 2011, datacenter automation has advanced considerably.

Amazon’s warehouses are increasingly robotic too, which doesn’t create great working conditions for the humans remaining there doing as-yet-unautomated jobs. Multidisciplinary researcher-activists Kate Crawford and Vladan Joler have written poetically about the uncanny valley characteristic of such workplaces:

At Amazon distribution centers, vast collections of products are arrayed in a computational order across millions of shelves. The position of every item in this space is precisely determined by complex mathematical functions that process information about orders and create relationships between products. The aim is to optimize the movements of the robots and humans that collaborate in these warehouses. With the help from an electronic bracelet, the human worker is directed through warehouses the size of airplane hangars, filled with objects arranged in an opaque algorithmic order.

Hidden among the thousands of other publicly available patents owned by Amazon, U.S. patent number 9,280,157 represents an extraordinary illustration of worker alienation, a stark moment in the relationship between humans and machines. It depicts a metal cage intended for the worker, equipped with different cybernetic add-ons, that can be moved through a warehouse by the same motorized system that shifts shelves filled with merchandise. Here, the worker becomes a part of a machinic ballet, held upright in a cage which dictates and constrains their movement. ¹⁷

Of course, the inventors would doubtless argue that these metal cages aren’t designed to imprison people, but to keep them safe from heavy machinery in an environment well on its way to becoming a human exclusion zone. Then again, one could argue that since cities are wildlife exclusion zones, the only safe way to keep a wild animal in the city is in a zoo cage. Outside human enclaves, will we be tomorrow’s wildlife, caged for our own safety amid our robotic infrastructure?

Most farmland in the industrialized world is a human exclusion zone now, too. Great tracts of the Midwest that used to be divided up into family farms are now tilled, planted, and harvested by hulking agricultural robots whose smaller ancestors we used to call, quaintly, “tractors.” Needless to say, none of these robots looks anything like the droids on the Skywalker family farm.

In fact no working robots today are anthropoid, like the Star Wars “protocol droid” C-3P0, because they don’t really do human protocol; why would they speak our languages, or have arms and legs, or person-like dimensions at all, if their environment is entirely non-human? Why would they sense the world as we do, or operate on our timescales? We tend not to recognize robots as such because we bring in so many anthropomorphic assumptions. Even the beepily articulate R2-D2 turns out to be a human-enough shaped chap (as evidenced by the fact that he could be operated from the inside by three-foot-eight-inch English actor Kenny Baker). Droids were buddies—not only because they could communicate with you, ¹⁸ but because they could share spaces with you, go where you go. With a bit of Jedi persuasion to overcome human prejudice, they could even accompany you into an interplanetary dive bar.

You don’t want to try walking into a bar with a bright yellow, twelve foot tall, eleven tonne FANUC M-2000iA. Oops, you can’t—it’ll be bolted down. Little pick-and-place machines that solder circuit boards at high speed might be more compact, but they’re generally fully integrated into an assembly line—it might not even be obvious where one robot ends and another begins. Perhaps a bright green autonomous John Deere 8R tractor, then? ¹⁹ It may be highly mobile, but it, too, wants you to keep your distance. Its beeps aren’t the friendly dialtones of R2-D2, but loud, monotonous, garbage-truck-backing-up sounds telling any nearby human to stay well clear. If you, or anything else out of the ordinary, are in the advancing path of this behemoth, it will stop in its tracks and call for supervisor assistance, just like a grocery checkout robot when you forget to put your scanned item on the scale. As Doug Nimz, a fourth generation corn and soybean farmer from Minnesota, puts it in a promotional video,

I can pull up the app, I can monitor the tractor, see how much of the field it’s gotten tilled. I can check the fuel level. I can check the app to see how much of the field is left. If there was something in the field that it wasn’t sure about the tractor will stop and alert me […]. Autonomy will help because we will be able to put a tractor out in the field and let it run for 24 hours a day because it’s not manned. But it also helps us with the weather because we can run so hard when soil conditions are fit. ²⁰

Superficially, autonomous tractors resemble driverless cars—and autonomous driving is one domain in which city dwellers have expressed real anxiety about job loss to automation. There are millions of truck, taxi, and Uber drivers in the US. It’s interesting, though, that despite many years of incremental progress, real adoption of driverless car technology always seems like it’s still a few years off. In the meantime, driverless tractors, ports, container ships, and mines are already here, largely unnoticed. They differ in that, once more, these applications can take place in human exclusion zones!

In a hypothetical New York where the streets were given over entirely to robotic cars, passengers could be whisked anywhere at great speed. There would be no need for traffic lights, stop signs, or indeed any traffic signs, as vehicles could coordinate the timing of their flows and turns through each intersection with the precision of a zipper’s interlocking teeth. We might finally have the personal freedom of movement and speed that car manufacturers have been promising in their advertising for a hundred years. As economist Judy Wajcman snarks,

The incongruity of the automobile’s promise of freedom of movement with the actuality of a largely sedentary existence in a landscape dominated by traffic-overloaded motorways is […] pronounced today. However, it is the speed of information flows rather than of motorcars that is at the forefront of our imagination. ²¹

Information flow would in fact be the key to fulfilling the promise of real autonomous mobility. With communication at near light speed across the street grid and nose-to-tail “trains” of cars forming and splitting at need, jams and delays would be a rarity. Lanes could be much narrower and street parking would not be needed, letting café tables and gardens spill out onto wide sidewalks. Residential and shopping streets could fully pedestrianize, as many of the cuter towns in Europe have done. No idling, no stop-and-go traffic, and far higher utilization would mean fewer cars and a lower ecological footprint. Electric charging or battery swapping could be fully automated, decreasing the need for resource-intensive long-range batteries. Public transit would have to up its game to remain competitive—or perhaps ride sharing and public transit could merge into a single system.

It’s an attractive picture. None of it is likely to happen at scale anytime soon, though. Driving in a real city during rush hour is a mess, whether for a human or for a robot. That’s why the London Underground will remain popular for a long time to come. ²² Indeed, today, “riding a Victorian technology in central London—the bicycle—during peak hours is faster than traveling by car.” ²³ This is the issue: driving involves constant interaction and social negotiation with other drivers, motorcyclists, bicyclists, pedestrians, and sometimes other animals—all of which have their own agendas, places to be, good or bad attitudes, and sometimes, horrifyingly, little screens they’re looking at instead of the road. A competent driver is aware of everyone else, conscious of what they’re paying attention to, and modeling what they might do next. There’s constant give and take, occasional risk taking and boundary pushing.

As density rises, the social component of driving becomes increasingly challenging. Sometimes honking needs to happen. That’s why, even as robotic Ubers glide through the eerily depopulated suburbs of Phoenix, we’re still nowhere close to having effective driverless vehicles in Mumbai. New York and London might fall somewhere in the middle… but probably more on the Mumbai side.

A few years ago, many of us who worked in the various sub-disciplines of AI believed that the fundamental challenge for autonomous driving was perceptual: machines’ inability to make sense of a complicated environment reliably, the way people can. However, the problem turns out not to be an inability to program machines to perceive cows, people, and bikes. While no sensor system is perfect, modern AI can perform all of these tasks, and with better reliability than distractible human drivers with our measly two eyes pointing in just one direction out of our heads. The problem is negotiating social interactions in a mixed human-machine environment, without the machine coming to a standstill. Safe and effective urban driving simply can’t be reduced to a set of preprogrammed rules, even given complete and accurate information about everything on the road.

At very high density, human social modeling of robots becomes a major challenge too. If people don’t consider a robotic driver a social entity deserving of respect—and perhaps even posing a credible threat to one’s own safety if suddenly cut off, or angered—then cautious, well-behaved autonomous vehicles simply won’t be able to make any headway during rush hour. They’ll be frozen in place as bolder, pushier pedestrians and mopeds swerve and swarm around them, confident in the machine’s unlimited patience and (presumably) near-zero tolerance for risk to human life.

Paradoxically, then, introducing autonomous cars can make traffic worse. It’s possible that driverless vehicles simply can’t be made to work in dense human environments unless they become convincingly social “persons”—or manual driving is banned altogether, and the street becomes a human exclusion zone. ²⁴

This contrasts dramatically with conditions in the Australian outback, where massive and increasingly autonomous ²⁵ “road trains” of trucks have been doing long-distance ore hauling for many years. It’s cheaper, faster, more fuel-efficient, and doubtless has saved the lives of truckers who might otherwise have nodded off somewhere between Tanami and Yuendumu—although many of those truckers might have preferred dosing up on Red Bull and taking their chances rather than being out of a job.

The outback and Mumbai differ, of course, in that the interior of Australia is one of the least populated places on the planet, while India is one of the densest.

Somebody is still directing those “autonomous” vehicles in the outback, of course. Originally, that person was driving the lead truck. Now, that somebody might be directing fleet operations remotely from 750 miles away. Autonomy and human exclusion zones mean that the human-machine interface becomes higher-level, more abstract and removed from the action. Ironically, it means that most of us—even when we’re “operators”—have less direct contact with technology than we used to. For example, a single farmer can operate (or, increasingly, just monitor) their agricultural robots on a tablet from the farmhouse—or from a Starbucks in town. Robotic tractor paths get overlaid on robotic drone-mapped fields showing crop growth and health. It’s like real-life Farmville.

For Rem Koolhaas, this abstraction takes on an almost metaphysical quality:

You could even say that the landscape and iPad have become identical. That the iPad is now the earth and the farmer works with it. And that on the iPad the ground is now defined. And every single action from planting to weeding is specified for the smallest pixel to create the largest possible yields. ²⁶

Cattle management and dairy milking, pigpens and chicken farms are similarly being run increasingly robotically nowadays. ²⁷ For the moment, fruit picking is delicate work that requires real humans with depth perception, hand-eye coordination, and nimble fingers, but lots of startups are working on that problem. Greenhouses of truly astonishing scale are now run in a way that looks a lot like lights-out manufacturing, with air mixtures and light spectra designed for maximum plant output—and decidedly not for human occupancy. Within greenhouses, we carry out a kind of localized terraforming, engineering the landscapes of alternative planets where tomatoes or lettuces are at home, but people must wear environment suits.

The Netherlands, with its high population density, low birth rate, and limited arable land area, has been a big innovator here. As a result, they’ve become—despite their tiny size—the world’s second biggest food exporter by volume! ²⁸

Since so much of the farmer’s work is now information work, it’s hard to imagine that the farmer’s physical presence will be needed at all in the future. Perhaps soon the farm will be managed entirely from afar, with just the occasional tech crew or veterinarian going out to debug a machine or animal. Eventually even those crews may be robotic. Seasonal migrant laborers, who today are needed to harvest certain crops, may no longer be needed as automation progresses. Food farms will start to look a lot like server farms.

If you’re a city dweller, this may all sound pretty weird—discovering that the View-Master is just a historical VR fantasy, and you’re now living in a habitable bubble on an otherwise alien planet.

Well, that’s not entirely true. The traditional “farm postcard” still does exist in places, just as some picturesque villages still survive; though like villages, artisanal farms aren’t “load bearing” today. They don’t, and couldn’t, supply much of humanity’s caloric needs. However, as technology for intensifying agriculture continues to improve and population starts to decline, that may change. Small-scale farming can also serve meaningful social purposes, and shouldn’t be dismissed merely as supplying “organic” luxuries for the rich. Food isn’t just about sustenance, but also about identity, dignity, and community.

Small-scale farming remains inherently communal; even in the Malthusian days of subsistence agriculture, it knit together the social fabric. ²⁹ Now that subsistence is less at issue but lack of social connection has emerged as a major problem, supplanting large-scale commercial food production with small-scale farming or gardening makes sense. Modern small-scale farming and gardening don’t focus so much on yield as on participation, connectedness to our food, and human interaction. People will pay extra for these things, or, given the opportunity, will invest time and love in them even when they don’t have that economic “extra.” That’s why small farms tend not to be in the deep countryside, but nearer to towns and cities, or even inside them. We see such practices today in many parts of Europe, at farmer’s markets, and for that matter on urban micro-farms in Detroit and in community gardens in Baltimore. The future of small-scale farming is already here—it’s just, like the other futures described in this book, not evenly distributed. ³⁰

There’s also more to eating well than meeting a minimum calorie count. Fresh herbs, beautiful tomatoes, lettuces, and many other non-staple crops can be grown in small plots or on rooftops. Still, big farming operations are here to stay, and it’s clear that we’re not going back to a world where those great tracts of land (or urban farms, or greenhouses) are worked by a peasant class. Automation and efficiencies of scale will remain essential.

Over time, sophisticated robotics and AI should allow even large-scale, highly automated farming to become more varied and organic. Swarms of smaller drones and robots with better sensors and brains can replace today’s steamroller-like iron giants. Instead of endless fields formatted like circuit boards with rows of identical plants, more nimble, intelligent cultivation of multi-species ecosystems will be able to take better care of the soil and produce more food per acre. Hopefully, individualized medicine for animals will do away with indiscriminately mixing large doses of antibiotics and antifungal agents into feed. Similarly, “precision agriculture” is increasingly allowing fertilizers and pesticides to be applied to individual plants, adaptively microdosing them rather than blanketing the earth with chemicals mostly destined to run off into groundwater. Especially in combination with greenhouse technologies, methods like these will likely result in further dramatic decreases in the land area needed to feed the world, as well as a much lighter ecological footprint. ³¹

We desperately need conservation and rewilding of the land freed up to arrest and begin to reverse our planet’s ecological free-fall. Taking a hands-on approach (both human and machine “hands”) to rewilding former farmlands may become necessary, as we may not be able to afford to wait the decades or centuries it’ll take for some of these degraded landscapes to recover on their own.

In one of his last books, the legendary biologist E.O. Wilson (1929–2021) advocated for half of the Earth’s surface to be designated a human-free reserve, to restore the planet’s biodiversity. ³² While the New York Times called Wilson’s plan a “grand retreat” ³³ —and it is, relative to land use today—some environmental scientists who have studied this topic closely believe that, to quote the title of one paper from 2019, “To Conserve Nature in the Anthropocene, Half Earth Is Not Nearly Enough.” ³⁴

Automation and human exclusion zones, then, aren’t dystopian, but necessary: just the kinds of rabbits we must pull out of hats to avoid ecological collapse. Fortunately, key large-scale trends are moving us in the right direction already. We are retreating from large areas of the planet as we urbanize. Technical systems, including intensive automation, are giving us the means to do it.

There’s something missing from Wilson’s picture, though. The sense that we have been ruining nature, so must now wholly withdraw from it for the Earth to heal, rests on the idea that humanity is separate from nature, even in opposition with it—the exact belief that has caused so much of our self-inflicted harm. When economists, even conservation-minded ones, talk about “natural resources,” “ecosystem services,” or pollution as an “externality,” they reinforce this worldview. So do the romantics who see, in abandoned villages and family farms gone to seed, a “return to nature.”

In reality, nature is all around us, and includes us. It has been here all along. Human activity hasn’t only extinguished species, but has also been giving rise to them at a furious pace, both through agricultural engineering and through the unwitting creation of novel ecosystem niches. New kinds of mosquitoes have made homes in subway tunnels; rodents of all kinds have specialized to take advantage of human infrastructure; certain insect populations respond differently to light now, and have changed their appearance to blend in against new surfaces; urban birds have evolved to eat novel foods and sing different songs in cities, which are loud yet full of tasty snacks. ³⁵

Of course we should do everything in our power to protect the remaining Himalayan snow leopards and spotted owls, but it’s a mistake to consider them “natural” and the new species (including ourselves) “unnatural,” or to turn from despoiling the environment without a second thought to setting it apart from us on a pedestal—a pattern curiously akin to what Freud called the “Madonna-whore complex,” in which a man insists on seeing women as either saintly or debased. The genderedness of the metaphor is hardly accidental, for we’ve once again encountered the Baconian vision of scientific patriarchy: “man” raping a feminized “nature,” albeit in these more enlightened times, withdrawing after the deed in shame and remorse rather than triumph.

Can one respect nature without treating it as a saintly, untouchable “other”? Absolutely: Native Americans, Aboriginal people, and many other indigenous populations were active, constructive participants in their ecologies for thousands of years, until their ways of life were disrupted by colonialism and globalization. For instance, many cultures traditionally use controlled fire to manage forests and scrubland, enhancing biodiversity and decreasing the likelihood of catastrophic uncontrolled burns; this is far more sophisticated than Smokey the Bear’s doctrine, which presumes that the ideal interaction between people and the forest is to “leave no trace.”

Similarly, in her 2013 book Braiding Sweetgrass, ³⁶ Robin Wall Kimmerer, an environmental biologist and member of the Potawatomi tribe, cites many examples of the “honorable harvest,” in which the interdependence between humans and other species goes far beyond “sustainability.” Sweetgrass, black ash, and a number of other plant species have been shown to thrive better in environments where they continue to be harvested traditionally than where the harvesting has stopped.

Far from being an oddity, such mutualism between species, including humans, is normal. Fur trappers who practice their trade thoughtfully increase the local populations of the animals they hunt. ³⁷ So, too, do modern, gun-toting game hunters. Ducks Unlimited, founded by duck hunters in 1937, has conserved more than 15 million acres in North America ³⁸ —roughly the area of West Virginia—and the organization’s efforts seem to be working. While North American bird populations in general have declined alarmingly since 1970, waterfowl have increased by over 50%. ³⁹ Trout Unlimited, Salmon Unlimited, Pheasants Forever, and other similar groups have mobilized to protect many other species people are invested in preserving. ⁴⁰

There should be no surprise here, for this is how everything else in nature works: coevolution, co-adaptation, and interdependence, even between predators and prey. The whole ecosystem flourishes as each species finds new tricks to eke more out of its own niche, in the process providing increased inputs or services to neighboring species, and helping them to thrive in turn. In nature, relationships are paramount.

The animist worldviews of many indigenous people reflect a practical understanding of this principle. They often attribute personhood to other species; Kimmerer describes her elders referring to beavers as “Beaver people,” bears as “Bear people,” and even trees as “standing people.” ⁴¹ If you’re committed to the modern Western idea that beavers, bears, and trees are not people, you might think of it as a mind hack, analogous to what anthropologists call “fictive kinship” (see Chapter 4). Fictive kinship can socially extend relationships like brother-sister to entire clans or moieties, both strengthening social bonds and discouraging inbreeding by generalizing the incest taboo. “Fictive personhood” can similarly extend our instinctual empathy and reciprocity to the nonhuman, mobilizing both our emotions and our ingenuity on behalf of companion species.

Clearly, fictive kinship and fictive personhood are useful concepts, but do they rest on lies, or reflect deeper truths? Is that even the right question? We once more grapple with the quandaries of authoritative definition that have come up throughout this book—with handedness and ambidexterity in Part I, sex and gender in Part II, and cities, suburbs, and countryside earlier in Part III. If Europeans and Aboriginal people define kinship differently, what makes the European definition “real,” and the Aboriginal version “fictive”? If the Potawatomi consider the trees people, who has the standing to tell them they’re mistaken, and why? They clearly have better ideas about forestry than the colonists did.

This is not a matter of pitting science against animism, but rather, a matter of how we understand categories and identity, and the consequences of that understanding. Insofar as identity affects real world outcomes, we might all benefit by learning something about kinship and personhood from Native elders. In time, we may look back on today’s Western ideas about personhood and find them as provincial as the 19th century belief that living things are governed by a different chemistry from inanimate ones, or that humans have no common ancestry with the other great apes.

Similarly provincial thinking has given rise to the idea of so-called “invasive species”—a term coined at the height of the Cold War by English zoologist Charles Sutherland Elton. ⁴² The notion that nature comes packaged into a predefined taxonomy, with every species native to a certain place and anything discovered out of its place an “enemy alien,” is reminiscent of the old “Great Chain of Being” cosmology wherein God, as the universe’s CEO, assigned every living thing a spot on His org chart (see Planetary Consciousness, between Chapters 4 and 5).

In fact, species evolve and differentiate continuously. Moreover, they migrate all the time, both seasonally and as a result of shifting ecological conditions. The checkerspot butterfly, which flies only a few paces from its cocoon before laying eggs, is about as sedentary as a flying animal can get, yet its range has been creeping steadily northward and uphill as the climate warms. Trees, which seemingly don’t move at all, have been shifting their territorial ranges too. Between bird droppings spreading seeds from one continent to another and rafts of vegetation ferrying creatures across oceans, it turns out that life on Earth has never respected geographical boundaries; it’s always on the move. ⁴³

Yes, cruise ships, muddy shoes, and undeclared veggies sneaking through customs have sped up and assisted the migrations of many species. Migrations have at times been disruptive, and occasionally catastrophic, especially on remote islands, where highly specialized, long-undisturbed bird and mammal populations have been decimated by cats, rats, and pathogens introduced by human colonists. The global war on “invasive species” has been waged on the strength of a few such ecological horror stories.

However, as French ecologist Audrey Muratet has noted, on the entire European continent, not one documented extinction can be attributed to an “invasive species”; in fact, in most cases, their introduction increases biodiversity—as one might expect, given that every introduced species creates new relationships, niches, and opportunities for others.

We’ve gotten the story wrong, Muratet argues, because the “construction of knowledge about [invasive] species is marred by ideological values that bias scientific reasoning and experiments.” Moreover:

It is difficult not to draw a parallel between exotic species and immigration. […] The long history of racism and xenophobia, whether in the United States or in Europe, has contributed to the denigration of what appears to be “foreign.” In ecology, the restoration of native landscapes through the control or elimination of alien species is predicated on a discourse […] similar to xenophobic nativist sentiments aimed at human populations. […] [Such] arguments, like those against immigration in the human sphere […], are devoid of scientific rigor but not of nauseating ideologies. ⁴⁴

Muratet’s arguments should not be taken as a call to stop worrying about biodiversity. Many wild plant and animal populations have gone into dangerous decline as human populations have exploded. The main drivers have not been “invasive species” as usually defined, but rather people, domesticated animals, crops, and technologies terraforming the landscape at high speed—shrinking or destroying some niches, creating others, and, of course, altering the climate everywhere. ⁴⁵ We cannot expect the rest of nature to remain static amid such dynamic upheaval. Migration has always been the way living organisms escape worsening conditions and seek out better ones, and during times of great change, great migrations are bound to occur. ⁴⁶

We humans are no different. A 2020 ProPublica and New York Times report entitled “Where will everyone go?” describes conditions at the leading edge of an unfolding crisis, in regions of Central America where climate change and poverty intersect:

Even as hundreds of thousands of Guatemalans fled north toward the United States in recent years, in […] a state called Alta Verapaz, where precipitous mountains covered in coffee plantations and dense, dry forest give way to broader gentle valleys—the residents have largely stayed. Now, though, under a relentless confluence of drought, flood, bankruptcy and starvation, they, too, have begun to leave. Almost everyone here experiences some degree of uncertainty about where their next meal will come from. Half the children are chronically hungry, and many are short for their age, with weak bones and bloated bellies. […] The odd weather phenomenon that many blame for the suffering here—the drought and sudden storm pattern known as El Niño—is expected to become more frequent as the planet warms. Many semiarid parts of Guatemala will soon be more like a desert. Rainfall is expected to decrease by 60% in some parts of the country, and the amount of water replenishing streams and keeping soil moist will drop by as much as 83%. Researchers project that by 2070, yields of some staple crops […] will decline by nearly a third. ⁴⁷

As less habitable or even uninhabitable zones open up around the equator and habitable zones shift toward the poles, life, both human and nonhuman, will do what it has always done: migrate. Over a billion people are likely to be displaced by 2050. ⁴⁸ Humans today, though, face a new barrier to migration both unique to our species and, despite our tendency to unthinkingly accept it, only a couple of centuries old: the legally regulated form of personal identity known as “national citizenship.”

As a formalized proxy for language, culture, and values, citizenship is very imperfect. Nearly all countries have become multilingual and multicultural to one degree or another, precisely because of accelerated human migrations (voluntary and not) from the colonial period onward. Globalized media mixes cultures up even further. Ensuing diasporas have created widely dispersed populations with shared cultures and languages.

Pushing against such emerging complexity in the 20th century, colonial administrations created many countries out of whole cloth, especially in Africa and the Middle East, with arbitrary borders insensitive to cultural realities on the ground. Yet arbitrary lines drawn on a map matter when people’s movement across them is prevented by force. Then too, our human tendency to embrace identities and undergo divergent cultural evolution often paves a road to nationalism. The partition of India and Pakistan, or of North and South Korea, shows how quickly this process can operate. At any rate, in many economically less-developed parts of the world under climate stress, from Myanmar to Afghanistan to Guatemala, the old nation-state model seems broken.

Using a computer simulation, ProPublica ran two potential scenarios for Central American regions like Alta Verapaz over the coming decades. In the grimmer of the two, northern countries like the United States seal their borders, resulting in the immiseration and early death of tens of millions of climate refugees. Under such stress, local governments may topple, preventing investment in infrastructure and embroiling the region in resource conflict and civil war. The numbers of the dead and starving will balloon, due to the powerful correlation between poverty and birth rate. From the viewpoint of the rich and secure, having more children under these conditions seems irrational, irresponsible, or even inhumane, but as we’ve seen, this view is out of touch with reality. When child mortality is high and survival depends on subsistence agriculture, we have more children. Or in Malthus’s terms, negative checks only come in response to a decline in positive checks.

In the more optimistic ProPublica scenario, the United States allows the flow of climate migrants across the border. Although a porous border would be a major political challenge, it would be the better choice by far. Firstly, it would save many people, avoiding the worst of a humanitarian catastrophe. But also, immigrants are needed north of the border, where the highly urbanized United States is undergoing a demographic crash. Without an influx of young people, soon there won’t be a workforce to carry out many important jobs, both skilled and unskilled. The nursing and hospice needs of aging Americans alone are a cause for concern in the coming years, especially since such “caring labor” is less automatable than information work. In his book The New Nomads, author and activist-entrepreneur Felix Marquardt makes the case that countries with advanced economies like the US would do well to distribute refugees throughout their territories, including into their less-populated countrysides, as these are the places with the most urgent need for an influx of a younger workforce in essential services. Additional benefits would include avoiding the urban ghettoization and isolation of immigrant communities, and allowing organic cross-cultural contact to gradually soften fears of foreign invasion. ⁴⁹

While such measures can delay the inevitable, demographic collapse will ultimately cause the growth-based engine of capitalism to sputter. We’ve seen a preview in the countryside, where the emptying-out is already well advanced. Sooner or later, we will need to rethink the economy in terms that optimize for collective wellbeing rather than for unending growth; but in the richer countries, that day of reckoning will come sooner without immigration, as an aging (and increasingly post-reproductive) population shrinks. Delaying such rethinking has great human costs, as it appears that the growth economy has tipped from one that raised many boats during the baby boom to one that is more zero-sum today, both within countries and between them.