Overview

Food is our number one concern. For millions of years—and despite walking upright, making tools, and (maybe) talking to each other—we fed ourselves much the same way that most other animals did: we roamed around to gather our food. Then, less than a dozen millennia ago, a few of us started to farm. That then triggered a wave of new tools, which led to many changes in our lives until nearly all of us farmed. That remained our life for millennia. Then, just a couple of centuries ago, another wave of new tools triggered yet more changes in our lives until today less than half of us farm. But we didn’t plan all those changes. Nor did they happen entirely randomly. Generally speaking, we and our tools together form a self-organizing network, a swarm, and it can fall into certain configurations simply because that’s what certain kinds of complex networks do. For example, as the last ice age ended, our swarm entered a feedback cycle called ‘autocatalysis.’ As that spread, our swarm then rearranged its structure in a process called ‘phase change.’ Those two kinds of ‘swarm physics’ behaviors have helped shape us for millennia, and might well continue to help shape us into our near future.

Autocatalytic Runaway

It’s 11,600 years ago and the last ice age is ending. As the planet warms, global climate is shifting, and so are many of the herd migration patterns. Our food is getting harder to find. In the wind-swept hills of today’s northern Iraq, one of our bands is both bickering and starving. Perhaps we haven’t sighted a herd of gazelle or wild ass in many days. Or maybe we’d herded a few goats but disease killed them. What to do? Keep walking and starve to death? Or make camp and starve to death? That’s why we’re bickering. But as we near a golden field of long-stemmed grass—a wild wheat—we’re all at least decided on one thing: We’re so hungry that we’re ready to eat grass.

As our band of 25 or so enters the grass field, we wonder whether the seeds clustered on the waving stalks are good to eat. Squatting to pull some up, we find that they come off in our hands, as if designed for easy harvesting. We then gather several lapfuls of wheat seeds, pound them into a pulp, mix with water, and lather the paste on a flat rock near the fire. That gives us unleavened bread, much like today’s tortillas or chapatis.

That night, painted by the warm firelight, our little band forms a tableau, a freeze-frame of our species before the fast-forward changes to come. Perhaps months (years? centuries?) later, we figure out how to ferment the same wheat seeds. Now we have beer. We don’t, however, give up hunting and gathering to become bread-eating, beer-swilling hicks. Foraging is what we know. It’s kept our lineage alive for millions of years. So we keep roaming, following our food.

Of course, we’ve been hungry enough to eat grass before this. For example, around 23,000 years ago we’d ground grass seeds into paste in a brushwood camp that’s now submerged on the southwest shore of the Sea of Galilee in today’s Israel. But for some unknown reason only now do seeds become vital to us, for descendants of our little Iraqi band keep returning to the field of wild wheat. Then, one day, again for some unknown reason, instead of building our usual brushwood huts then moving on, we invest more effort and use mud and stone to build a more sturdy camp. Even so, we still don’t settle down. But we do keep returning to that camp seasonally as game dries up elsewhere.

Now it’s about 300 years later. We still haven’t tamed the wild wheat plants near our camp, nor do we plant them, but we’re starting to store their seeds. For instance, in Dhra’, southeast of the Dead Sea in today’s Jordan, one of our bands builds something more than just a seasonal camp—we build a settlement. As before, we use mud and stone, but this time we build our huts around a central one with a raised floor to protect against mice and moisture. In it, we store our new treasure—seeds.

Now it’s a further 700 years or so later. At Dhra’, our numbers have swelled to around 90. We no longer have a central granary. Each of our huts now comes with the latest marvel—a built-in granary. We still hunt and gather, but our old rover ways are fading away. We’re stepping into a trap, but it’s one that we can’t see.

In all, it’s now been about a thousand years since our first mud-and-stone huts and for several of our bands in the area, the trap swings shut. For instance, at about this time in northern Iraq at both Jarmo and near Shanidar Cave, about 150 of us live together. That number may be the final turning point for those of us there because it’s around six times more than foraging alone can feed. What had been a small band of 25 or so wanderers has grown into a small village. We’re now too many to live by hunting and gathering alone.

By as early as 9,000 years ago thousands of us might live together in small cities. For instance, at Çatal Höyük in today’s central Turkey we’re all crammed together near a marshy plain surrounded by some live volcanoes. With hundreds of our mud-brick huts jammed together higgledy-piggledy, we have no streets and no doors—instead we climb into our homes via holes in the roof. We still hunt aurochs, horse, and deer, just as we had millennia before, but we also now herd sheep and goats. We also still don’t farm, but we now routinely harvest wheat, barley, peas, and lentils. Plus we gather almonds, pistachios, and also fruit, and tubers growing in the marsh. Further, we bury the dead under our floors. At this point, those of us there have settled.

There’s still much that we today don’t know about the challenges and opportunities we must have faced over those 2,600 or so years, but one that probably mattered a great deal was genetic changes in wheat. Look again at those wild wheat plants 11,600 years ago. First, their seeds are a little bigger than other grass seeds, so it makes more sense to gather them than other seeds. Also, although most of their stalks shatter as they ripen—so that their seeds fall to the ground, ready to sprout—the stalks of a few mutant plants fail to shatter. Normally that strain would be rare. It can’t make new plants, so from the plant’s species point of view, it’s a reproductive dead end. But as the ice age ended, those mutants may have saved some of our bands from starvation.

Probably we ignored the wheat stalks that had done the right thing and shattered. Picking up their scattered seeds would have taken more energy than eating them would give—something we only do when we’re truly starving. But the few mutant plants would still have had their ripe seeds on the stalk, which would have left them in the perfect position for us to harvest cheaply. Then, after a thousand years or so, we started planting some of the mutant seeds that we didn’t eat. That gave those mutants an edge over their normal cousins, so they spread. As we kept selecting among them, they grew taller, which made them easier for us to harvest, and their seeds grew bigger, which made them even more worthwhile to harvest. With that, we hadn’t just settled, we had also begun to farm. Over the next few millennia we tamed other plants in several other places around the planet: rice, beans, squash, and yams in southern China, northern India, north-central Africa, and central America.

Today we still live with the genetic changes that we unwittingly started back then. For example, we first tamed maize in Mexico’s central highlands around 9,000 years ago. Early on, those wild corn cobs were less than half an inch long, but over the millennia we kept selecting bigger and bigger cobs. By 1492, when Columbus dropped in on the Americas, some cobs were already six inches long. Today, some are 18 inches long. Without our help, such mutant maize plants would die out—their kernels couldn’t escape the cob to sprout.

It’s the same for many of today’s egg-laying hens. Over many generations, we’ve selected them to produce lots of eggs—but also to not sit on them. If we died out, so would they. So, too, would today’s cows. If we weren’t here, they’d stop being fat and placid—and pregnant every year. Their numbers would plummet, and any survivors would both lose their fat and go feral. Over time, their descendants would transform back to something like their ancestors—big, swift, wiry aurochs. Watermelons, too, would shrink back into small, bitter berries. Poodle numbers would also crash, and any survivors would grow back into something like their ancestors—gray wolves. So, too, would bulldogs, because without us around to do their cesarean sections, the heads of eight out of ten of their pups would be too big for them to be born. Today’s sheep and apples and rice, and all our other tamed animals and plants, are all products of the same unplanned, millennia-long genetic experiment. They’re all made things.

But while we bent all those things to our will, doing so also changed us. For example, once we settled and started farming, slavery became more likely. When we were all mobile we likely weren’t meek—skeletons from that time show that we could maim and kill just as well then as now—but we likely didn’t take slaves. There’d have been no point. To a bunch of rovers, each new mouth might have meant having to add about 80 acres to our band’s yearly food-gathering range. So capturing someone to darn our socks would have made zero sense. (Not to mention that we had no socks.) However, once we started turning into farmers, new hands became worth more than their new mouths ate. Plus, as farmers, we also had more work to do. Slavers could now both feed slaves and force them to make more food than they ate. Also, once we were stuck in place, we could invest in more permanent things—like slave pens. Even herders can’t stop slaves from running away as cheaply as farmers can. Then once we had longer-distance trade via the horse, the camel, the ship, and such, slavery would have made even more sense. Before then, as slaves we could just run away. Our ex-captors could only follow on foot.

Farming’s biggest change, though, was what it did to our numbers. First, as today’s female athletes and dancers know, lowering body fat lowers female fertility. Ovulation slows, or even stops. Thus, hunter-gatherer women are lean and wiry, and bear few children. But as we settled and our food supply steadied, female body fat grew more uniform throughout the year. Ovulation regularized. Birth rate rose. It rose for another reason, too. Hunter-gatherer mothers suckle infants longer—perhaps because of having to carry them. And breastfeeding releases hormones that reduce fertility. But once we settled, we no longer had to carry our babies all the time. That in turn reduced suckling, which then raised our birth rate. Settlement raised our birth rate for yet another reason. Once we no longer had to carry our young all the time, women could manage more than one toddler at a time. Plus, farming made an extra pair of hands, enslaved or not, worth more than the extra mouth that goes with them. Thus, as we settled and then started farming, the cost of our kids relative to their future labor value fell. Women, instead of reproducing every three or four years, turned into yearly baby machines. We multiplied like rats in a grain silo.

That kind of self-feeding cycle isn’t unique to our swarm. Chemists might call it an autocatalytic (‘self-stimulating’) reaction. A catalyst is anything that stimulates a chemical reaction while remaining unchanged. An autocatalytic chemical reaction thus helps itself continue—it makes its own catalyst. The more the cycle goes on, the more catalyst it makes, which helps the cycle go on, which then makes more catalyst. It’s a beast whose hunger rises the more it eats.

Something like that apparently began happening to a few of our groups around nine millennia ago. The more mutant plants we got to eat, the more of us there would be to ensure yet more mutant plants to eat. We’re thus as much a part of our mutant plants’ reproductive system as they are of ours. All our tamings over the last few millennia have changed many things around us, but that’s also changed how we must live. We’ve changed our food, but our food has also changed us.

Changing Phase

Autocatalysis may help explain why farming began to spread, but not why we stuck with it for so many millennia after that. Why did we? It can’t be because farming is built into our genes. That’s true for the few other species that farm. For example, some termite lineages have farmed fungus for perhaps 50 million years. For them, farming is inborn and they can no longer back out of it. But none of our ancestral species, going back billions of years, were farmers. Further, our gene variations usually take around 2,000 to 4,000 generations to spread widely among us. For us, farming is only a few hundred generations old. So from our genes’ point of view, it’s a mere fad, like the hula hoop or the pet rock. We picked it up just a little while ago and half of us have already given it up. The bulk of our genes mostly haven’t even noticed it yet—and given how fast we’re fleeing the farm today, likely never will.

We also probably didn’t get into farming because we somehow suddenly got smarter. When we today think of our past hunter-gatherer life, it’s common to imagine that we were dressed in rough-cut hides wandering through desolate, virgin, landscapes. That’s the picture that movies often paint for us. It may be reasonable if 11 millennia ago we were naked apes with bad haircuts and heavy jawlines hefting stone tools while we grunted at each other about how nasty, brutish, and short our lives were. But genetic change is slow, so we likely weren’t any stupider then than we are now. Plus we had lots of spare time. Plus we had millions of years to fine-tune our clothes and tools. So it seems more reasonable that back then we wore well-tailored clothes and intricate tattoos and body paint—literally dressed to kill. We may also have carved totems of our passing into many rock faces, hillsides, riverbanks, and trees that we camped nearby, like dogs marking our terrain. Over the millennia, such unsheltered signs would have weathered away, leaving only the few remains of cave art today. Adorning ourselves or adorning our territory—both may also have helped us keep the peace. Finally, for millennia we were on foot, so weight was the enemy, so, likely, we probably mostly made grass shoes, net and leather bags, string or strap baby slings, light-weight weapons, and lots of ornaments—not lots of stone axes. Chipped rock from that time may well be the main relic today only because it outlives bone, wood, grass, and leather. So calling that era the ‘Stone Age’ may be misleading.

Likely, too, we probably didn’t get into farming because it was necessarily any better than foraging. Of course, as rovers, we sometimes starved. Whenever we stripped all the berries from the bushes, dug up all the tubers in the ground, and netted all the rabbits we could, we had to move until those things grew back. Judging by today’s few remaining rovers, each of us then had to schlep about 20 pounds of food, weapons, tools, and babies. Maybe any adult who couldn’t do that—anyone weak, or sick, or maybe even just lame—died. However, as rovers we also had few diseases. Not enough of us lived together for them to persist. Also, while we didn’t have much, we also didn’t need much. Plus, our dogs, who we may have tamed as far back as 43,000 years ago, may have helped with the hunt. We were also hardy from unceasing exercise. Then, too—likely—none of us were slaves, and—likely—all of us had a say in what we did next. All of us in our band of 25 or so were related, and, likely, all of us, including children, were armed. Finally, our skeletons from that time show that we were tall and slim and fit and healthy.

But once we started farming our lives changed a lot. For example, skeletons from about 10,400 years ago in Abu Hureyra, on the banks of the Euphrates in today’s northern Syria, speak of females in pain. They had enlarged and often injured toe joints, curved and buttressed femurs, enlarged knees, and damaged, or even crushed, spinal disks. They must have had to kneel when grinding nuts and grain. And with only stone grinders and sweat, it must have taken them hours to grind enough flour for just one family meal. Adolescent skeletons also speak of regular and excessive strain. They must have had to carry heavy loads on their heads daily. And everyone—girls and boys, men and women—also had fractured teeth. Likely, that came from eating partially ground grain and bits of stone flaked off during the grinding. Millennia later, some of us in today’s Israel would mourn that “In the sweat of thy face shalt thou eat bread, till thou return unto the ground.”

But if farming was so stressful, once we tried it why didn’t we simply give it up and go back to foraging? Perhaps some of us did, but most of us couldn’t. We were already too many. For instance, in Abu Hureyra we’d originally sited our village on a gazelle migration path. Fresh meat would thus have delivered itself to our doors regularly. For a while, the pickings must have been good. We must have grown fat. With autocatalysis, our birth rate must then spiked. But as the planet kept warming, the herds declined. Oops. Nurturing mutant grain would then have become more and more important to us. The forager shtick that had worked for our lineage for millions of years quit working. To go back to it would mean that most of us would have to die.

Instead, in Abu Hureyra we started making new tools. For instance, by about 9,000 years ago, weaving there had already become a speciality among a few women. Fewer fractured teeth overall suggest that we then figured out how to weave sieves fine enough to sift our flour. Then, as the Syrian climate dried further, the gazelles vanished entirely. But we couldn’t uproot ourselves to follow them—most of us would starve to death. Instead, we tamed sheep and goats, and we stepped up our planting. Then we figured out how to hand-throw clay pots (the potter’s wheel was still 2,500 years into our future), and pots made for a good way to store grain against rats. We may then have made more pots to soak our grain and cook it into porridge because our fractured teeth then disappeared. Those of us with few or no teeth could now survive. Before that, we probably simply died (or perhaps our children or grandchildren chewed our food for us...). However, with our new soft foods, tooth decay, once rare, grew. Also, with no more teeth-breaking, our numbers grew. But our ever larger numbers also helped diseases persist. They could now circulate among us for decades without dying out. They could also move back and forth between us and our newly tamed food animals. So we got sicker. Infant deaths then climbed. In short, each solution to a problem led to new problems, which led to yet more tools.

We made all those new tools, but we couldn’t have foreseen their network effects. It’s thus unlikely that we became farmers because we looked into the future and loved what we saw. What hunter-gatherer would have interrupted the campfire songs to say: “Let’s farm! Sure, we’ll lose about five or six inches in height. We’ll also lose a lot of teeth. We’ll be sickly, too. We’ll get less protein, we’ll eat too much starch, and we’ll have less variety in our diet. Plus we’ll have to work much harder—and all day long. Did I mention that we’ll also invent slavery? Also, more women will die making more babies, more of whom will die young. We’ll also be in great pain for thousands of years. But we’ll just have to bear it so that our remote descendants can have video games and radio telescopes and mobile phones.”

But if farming was so hard on us, why did it spread? And once it spread, why did it keep its grip on us for millennia? Again, it probably wasn’t because we wanted to change. Instead, it may simply have been a question of numbers. About 1,000 farmers can live on the same land that 25 foragers need. So if the two groups ever fought, those legions of sickly, gap-toothed farmers would wipe out the few healthy nomads. Even when the few tall rovers won, after a while the hordes of runty farmers that they would then rule would swallow them, as a pond swallows a flung pebble, so it would be much the same as if they’d lost. We’d farm mostly the same foods. We’d make about as many babies—and about as many of them would die each year. We’d be able to feed about as many priests and warriors—although they might serve different gods and have different weapons. Life would go on much as it had before, except that we’d have changed some of our leaders, names, languages, faiths, customs, gene lines, and such. So whether the rovers won or lost may have made no real difference to farming’s spread.

That seems to be what happened to those of us who were hunter-gatherers in central Europe about 7,500 years ago. That’s when those of us who were farmers in today’s Turkey swept through from the south. In Europe, the forager way of life then vanished—as did most of Europe’s male foragers—since male farmers then fathered much of today’s Europe. That’s also what seems to have happened to those of us who were Amorites west of Sumer in today’s Syria. Over 4,000 years ago we lived up in the hills, tending our flocks. We ate raw meat, didn’t build houses, and didn’t plant grain. But we were good warriors. So when we swooped down on the plains, there was much wailing and gnashing of teeth. But not for us. For us, life must have stayed the same—for a while. Over time, though, we turned into farmers, just like the folks we then ruled. Same story for the Hyksos who swept in on Egypt around 3,700 years ago. Same for Khoisan hunter-gatherers in Africa about 3,000 years ago, when Bantu farmers and herders started expanding south and east. Same for the Vikings in Europe 1,200 years ago. The Turks in Persia, the Mughals in India, the Mongols in China—the farm swallowed them all.

Any tool that ramps up our ability to gather food reliably can increase our birthrate. It needn’t be a sickle for grain plants. It could be woven traps for shellfish, or fishing nets for salmon runs, or using torches to stampede bison off a cliff. If it also induces us to settle, then the more our numbers rise, the more we must depend on our new food source. The more dependent we are, the more precarious our lives become. We can find only so many calories given our current toolbase. The advantages of settlement and intensive food production rise. Autocatalysis then drives our birthrate up so much that over time we cultivate even the most marginal foods. So when the next climate change, or plant blight, or other food catastrophe hits, we’re always caught with our pants dropping and our numbers rising.

In brief, become farmers and we must stay farmers. And any non-farmers who warred with farmers must also become farmers. Thus, as rovers, to stay nomadic we either had to kill all the nearby farmers, or run away. Our only other choice was to become herders and trade with them—or run away. And every generation would bring yet more farmers crowding our land. Farming was thus the roach motel of human life: we could check in, but we couldn’t check out.

As with autocatalysis, that network process isn’t unique to our swarm. Physicists might call it a phase change. Water, for example, has several phases. In its liquid phase it won’t expand to fill a room, but as a gas it will; as a liquid it will dissolve sugar, but as a solid it won’t, and so forth. Its properties abruptly change as it boils into steam or freezes into ice. The advent of farming shows that our swarm can phase change, too. It’s stable if we’re all in one phase or the other—either rovers or settlers, but not both. When we’re a mix, rovers feel network pressure to settle. It doesn’t matter whether we like it. It’s irrelevant whether we planned it. It’s unimportant if we even notice it. In time, most of us will phase change into farming—and stay there.

Thus, we didn’t choose farming. It chose us. When all of us lived in small roving bands, our food didn’t stay put, so neither could we. Since we could walk only so fast, and since only so many new berries and tubers and rabbits could pop up per acre per year, only about 25 of us could forage together. With settlement and farming, our numbers exploded, but as farmers we weren’t therefore taller, healthier, longer-lived, more relaxed, less pained, or even better fed. Of course, as farmers we could have things that we couldn’t have as foragers—slaves, for example. But as foragers we also could have things that most farmers couldn’t have—free time, for instance. Farming yields more food per acre of land, but foraging yields more food per hour of labor. However, we didn’t have a free choice between those two options. Settlement, then farming, probably first occurred because it was the only choice for a few of our bands during a stressful time. Autocatalysis then took over. But once that happened to enough of our bands in enough places, the spread of farming amplified until we phase changed into grass-eaters all over the planet. We didn’t decide to do that. In some sense, our swarm did.

Eat Your Heart Out

By about five millennia ago many of us around the planet had phase changed into farming. We then began a way of life that’s still the norm for almost half of us today—and even though billions of us now no longer farm, nearly all of us still live on farmed food, much of which, even today, is still either grass or based on grass. In Europe, for example, our staples were, and still are, cereal grasses: wheat, barley, oats, rye. We baked them into bread and boiled them into porridge or pottage. We also fermented them, or various fruits, to make ale or cider. To those we added peas and beans, turnips and cabbages, leeks and onions, and—when we could get it—honey.

For all those millennia, our peasant diet varied with the region and the season, but in essence it was, and remains, everywhere the same: mostly vegetable soup. We ate vegetables and drank ale because we had to live low off the land. Here, for instance, are some prices from around 1300 in England: Two dozen eggs cost a penny; two hens, thruppence; two geese, five pence. A sheep cost one shilling and tuppence (that is, 14 pence). A hog cost three shillings and four pence. Two gallons of ale cost a penny, and a gallon of wine cost four pence. Also, those prices varied with the weather. In good times, two bushels of wheat cost five pence, but after a big storm, that much wheat could cost four shillings. In a famine, it could cost five shillings, or more. To today’s ears such prices may sound low, but back then even an experienced carpenter only earned tuppence a day. A laborer took two days to earn thruppence. A maid took two weeks to earn tuppence. Plus, at least a third of us were held in bondage. A slave and his family sold for 13 shillings and four pence, or about as much as a cow—or a warhorse.

Thus, for most of us in Europe (and elsewhere), meat was a rare treat. Most of our animal protein came from eggs, lard, or bacon drippings. Oxen and horses were more valuable for dung and plowing. Cows were for milk, butter, and cheese. The forest deer were for our manor lord. Get caught killing one, and the lord’s foresters might castrate us, if male, blind us, if female, or simply hang us—that is, if the wolves or bears or wild boars didn’t get us first. Instead, as peasants we raised pigs. They can live on kitchen slops, forest beechnuts, and acorns. We built fishponds in the nearest river, and we raised poultry—though not for the meat, but for the occasional egg. Men with swords or bibles took the rest. Much of our extra animal protein went to our lords temporal as tax and our lords spiritual as tithe.

A very few of us—monks in wealthy monasteries, knights in shiny armor, ladies in funny hats—ate more richly, but the vast majority of us were poor. And rich and poor alike, we all built our meals around the staff of life, bread. However, wheaten bread was mostly for our few rich. Most of us more often ate rough bread made from beans, oats, and other cheap grains. Also, since mold was always a problem, soft breads wouldn’t keep, so we often baked them into bricks. We didn’t eat them—we gnawed them. The toothless ate porridge.

Further, even without famine, most of us in Europe went hungry twice a year—in the spring, after winter stocks were gone, then in July, the month before autumn harvest. If that harvest failed, it didn’t always mean famine, but it always at least meant widespread hunger. First we foraged for beechnuts, berries, and nettles. Then we ate our older farm animals. Then we ate our future by eating all the rest. When those were gone, we phase-changed back into foragers—roaming the land, hunting for food. We then risked death to poach eels from the lord’s millponds, and squirrels, rabbits, and birds from the lord’s forest.

So we salted what meat we could, but salt was so dear that we sometimes used it as money. (Pepper and sugar were even dearer.) We also had fresh fruit and vegetables, but only in summer. The same goes for dairy produce, except for a little hard cheese and perhaps some salted butter. Plus, not all our animals could survive winter because we couldn’t feed all of them for all those months when no grass grew. And we, too, needed food to survive winter. So at year’s end, we killed most of our livestock and smoked or salted the meat.

Thus each year we had to solve hard math problems. A cow, for instance, was hugely important. First, it was a milk source. It was also a lot of meat on the hoof. Its flops were also a cheap fuel source—and a building material. In a pinch, it could also pull the plow if the oxen died. Plus, it was also a source of dung for the fallow fields—and thus future hay, and so future cows. So we had to balance our yearly cow-killing against the amount of fodder that we could gather and store before the autumn rains spoiled the hay. Kill too many, and we might starve next year. But kill too few, and again we might starve next year.

As with our cows, our children also presented hard math problems. After five or six winters, we could put them to work in the fields. However, they needed food for all that time. Each year we had to figure out how many kids we could afford five or six years later. Thus, each year we walked a tightrope, and sometimes we fell off, as happened to Europe in 1314.

That year, Europe fell into a Great Famine lasting seven years. Climate change triggered a mini ice age, and weather in Europe, after four centuries of relative mildness, turned cold and wet. That summer, crops, pelted by rain, rotted in the fields. Harvests were poor, and food prices rose. Spring brought months of storms and heavy rains. Dikes collapsed in England. Fields washed away in France. Rising rivers drowned villages in Germany. Crops failed from Ireland to Poland, from Scandinavia to Italy. In England, the price of wheat rose eightfold. Both fodder and grain, covered in fungus, rotted in the damp hay-barns and storage sheds. In heavy-clay regions, waterlogged fields rotted the standing crops even when the sun managed to shine. Continual rain leached nitrates from the soil, leaving it poor even when better weather returned. Disease then killed the cold, wet, and hungry livestock. Fewer oxen meant less plowed land. They also meant less manure for the fields. And they meant less transport—so even if one village did well its grain couldn’t reach nearby villages.

Another bitter winter followed. The Baltic iced over so much that ships froze in place. That spring, torrential rains returned, turning roads to mires and waterways to rivers, thus further blocking food transport. In Hungary, the Danube river—Europe’s principal transport corridor—flooded, washing away fields and villages. A sporadic war between France and Flanders, ongoing since 1297, worsened the crisis as armies stole draft animals and added banditry to their list of job skills. Northern Europe began to fall apart. As peasants, those of us there ate diseased cattle, pets, rats, insects. We ate edible weeds, like mugwort, hyssop, mallows, bittercress. We ate the leaves off trees. We ate grass. Then we stumbled across the countryside, begging for food. In the towns, unable to forage, we suffered even more—or so say the chroniclers, but then most of them lived in the few towns.

As among our cattle, disease then spread among us, killing rich and poor alike. Robbery and murder rose. So did food riots. In war-torn regions like Ireland and the Scottish border, we mined graveyards for fresh corpses. We cut the hanged down from their gibbets and ate them. Jailed convicts ceased to be fed and allegedly ate new prisoners alive. Families were said to eat their dead. Many chronicles even speak of parents killing their children for food. A poem from 1321 says it all: “A mannes herte mihte blede for to here the crie / Off pore men that gradden [wail], ‘Allas, for hungger I die.”

But while such mass death was extreme, it was hardly surprising. Widespread famines had come to Europe just before (in 1257-9), and would come again just after (in 1346-7). In the 50 years before 1314, England alone had suffered famine roughly every 11 years. That pattern had held for at least the previous thousand years. Nor was our famine cycle special to Europe. In Africa, India, and China we suffered just as much and just as often. Famine, pestilence, and war was a familiar cycle—everywhere.

Nor were those our only problems. In Europe at least, damp and cold killed us just as casually as hunger, conflict, or disease did. The poorest of us lived in small, dark, smoky huts. We built them with poles daubed with clay, cow dung, and brush. We roofed them with thatch and covered their earthen floors with straw. We had no chimneys and no windows. At around five feet tall, we were short and bent. Our skin, like a cured ham, was leathery from our household smoke. By 30 we were nearly toothless, and many of us didn’t live to see 35. Dirty and rank, we lived with our livestock and knew everything about lice, fleas, and dung—and nothing about microbes. One in four of us died before we were a year old. And all of us were always working. We tended the fires, the livestock, the fields. We made food, thread, cloth. We repaired clothes, bedding, cottages. In any scraps of time left over, we sewed or carved something to trade. Our skeletons from that time show extensive osteoarthritis, spinal deformations, bony growths, and joint enlargements, which together speak of decades of toil. When we were foragers, to stop walking was to die. Once we were farmers, to stop working was to die.

Of course, our lives weren’t always that hard. Most of us lived very low off the land—with meat and cheese and even wheaten bread rare—but we were used to our daily porridge. So, most of the time, we managed to stave off outright hunger. When we made just the right number of kids, and we slaughtered just the right number of food animals, and the weather behaved, we even feasted. We ate such little meat that slaughtering just one ox on a feast day might feed a whole village. Also, not all of us in a village were equally poor. The very wealthiest family might own a couple of oxen, a bullock, two horses, some cows and calves, a pig and sow, a hundred or so sheep, some geese and chickens, and maybe even a cart. Plus, it might have as many as five brass pots and pans, a jug and basin, a trestle-table, and maybe even a chair. But for all the rest of us, the fear of hunger was always there. Over nine-tenths of us were rural, and roughly a third to two-fifths of us not only had no food surplus, we didn’t even have access to enough land to give us all the grain we needed to survive. We had to earn the rest with non-farm labor. And if we didn’t, we starved.

For millennia, and all over the globe, we lived that way. Then, starting only a couple centuries ago, a fifth of us left that hungry world. Soon, another fifth will likely do so. Instead of constantly asking ‘When next can we eat?,’ more and more of us started asking today’s big three imponderables: Why are we here? Where are we headed? Do these clothes make me look fat?

That was a phase change just as big as the one that had dragged us into farming in the first place. How did it happen? Well, Balzac put one popular story about it this way: “The secret of great wealth is a forgotten crime.” It’s a good story. Our rich can indeed steal. Our poor have indeed been stolen from. For instance, in the late nineteenth century, a few of us in Belgium caused the deaths of at least eight million of us in the Congo, while stealing literally tons of ivory and rubber. Our rich have really good housebreaking tools—inside information, numbered Swiss bank accounts, tanks. But that can’t be why so many of us are so rich now. No amount of theft among ourselves can explain the difference in our lives between 1300 and today. All theft can do is rearrange who has what stuff. It can’t make more stuff. Most of what we have today we stole from the cosmos itself.

We did that with technology, which needn’t only mean something shiny that we buy in a store. Cows are technology, too. So is corn, butter, dirt roads, shoes, and chopsticks. So is a toilet, a sewer pipe, thread, mail-order catalogs, and chewing gum. Those are all tools. Much of how we organize ourselves are also tools. An installment plan is a tool. So is a bond market. So is a city. So, too, is a stock exchange. Even in the wilderness we make tools. Stacking stones into a cairn to cover a dead body is a tool. So is dropping a tree over a river to cross it. So is kindling a fire. We make all those things to serve our needs, and they wouldn’t exist without us. We eat technology. We sleep on technology. We live in technology. Technology is what we do.

By inventing new tools and trading what those tools help us make we can make more stuff. So the secret of most great wealth isn’t a forgotten crime; more often it’s a forgotten tool or a forgotten deal. There’s thus no fixed limit to our food. Even something as seemingly simple as our invention of pots changed our food supply. With pots, we could store more food than we could have before, so our overall food store grew. Whenever we make an important new tool—the pot, the plow, the tin can, the train, the shipping container—we can use it to make or store or trade more food. That also works for many of our institutional tools: banks, credit systems, pension plans. Many of our new tools push down our food-to-tool exchange rate.

Further, the periods during which that exchange rate stayed fixed—once millennia, then centuries, then decades, and perhaps someday, years or even less—are shortening. That’s happening because beyond a certain point we fell into a new autocatalytic cycle—an industrial one. Once we had a critical mass of industrial tools, the more we had, the more new food we could make (or store, or trade). The more food we had, the more of us there could be. The more of us there were, the faster our industrial toolbase could grow. In some parts of the globe today, our tool supply is now growing faster than our numbers can. Hence, some of us now worry about being too fat rather than too thin. We started fleeing the farm’s roach motel life.

In short, for millennia our swarm seems to have busied itself building a worldwide food machine. In that view, all our calamities have been mere blips on a long networking project. In many ways, we’re just like termites building a nest, except that ours now covers a whole planet.

But if that’s so, how come most of us don’t seem to see ourselves that way? Well, until a couple centuries ago, big increases in our food supply used to be rare. Most of the time, it really was mostly fixed. Thus, the only way for me to get more food was to kill you—or steal from you. Hence the Belgian sack of the Congo, among many others. That then led to the belief that we have a carrying capacity, as gazelles do. More of us—often a code phrase for more of us who are poor—just make for a bigger drain on our resources. Sooner or later we starve back to our carrying capacity.

Many of us who are rich today apparently still believe that, just as those of us in Europe in 1300 believed it. But back then, it made more sense. At the time, nearly all of us lived in a hungry world. Our most important labor-saving devices were oxen, children, and slaves—and they all needed food. Further, our transport tools were so slow and costly that trade was low, so our manors had to be largely self-sufficient. So when hard times came we had nothing to do but starve—or try to make someone else starve. Today, many of us, especially in our rich nations, have apparently forgotten that past, but we somehow still preach personal, familial, corporate, or national self-reliance, just as Europe’s manor lords once did. It’s a wonderfully bracing fiction—the frontier life, rugged individuals, everyone for themselves—but in practice, and with rare Robinson Crusoe exceptions, we’ve never been self-reliant. We’ve always banded together. It’s our awareness of, empathy for, and network linkage to each other that distinguishes us from the beasts that perish. We live by each other’s grace.

Seeds as Factory Embryos

As we’ve grown more industrial over the last few centuries many of us have changed our food tools a lot. But they might well keep changing, perhaps even in our near future, because in many ways we today are still farming much as we did long ago. In particular, we still waste a lot of energy to get our food, just as we did millennia ago.

That energy waste begins with our crops. Consider a carrot plant at the equator. It needs water, minerals, and energy. But it also needs carbon dioxide, for without that it can’t make carbohydrates, which means it can’t grow. However, there’s only a trace amount of carbon dioxide in the air. So even though the plant is bathed in solar energy, it only uses a little of that. And of that, it uses about half just to transpire water from its leaves. That’s how it keeps its sap flowing upward, against gravity, which is how it extracts water and minerals from the soil. It also uses a little energy to transpire water and thus keep itself cool. So its problem isn’t how to gain more energy, but how to lose more energy. It thus throws away over 99 percent of the energy that the sun gives it.

When we cultivate that carrot plant we lose even more energy. Even with pesticides, we lose at least 30 to 40 percent of all our crops to weeds, pests, and disease. For instance, we lose between a quarter and a third of all our soybean, wheat, and cotton before harvest. We lose even more of our maize, rice, and potatoes. Even after harvest, on average we throw away around 93 percent of our food plants’ bodies as inedible. That’s partly because we don’t eat most parts of our plants—roots, stems, branches, leaves. And of what’s left, we still lose a lot. For example, we lose around 40 percent of all our cereals after harvest because of spoilage. Then, of the remaining seven percent, we eat only about 13 percent. And we even waste a lot of that. For instance, in 1995 in the United States we threw away 27 percent (96 billion pounds) of our edible food. In 2007 in Britain we threw away almost 15 billion pounds of our edible food. We use much of the remaining energy to seed, nourish, protect, harvest, transport, prepare, package, and sell those edibles. In all, between our plants’ problems and our problems, when we eat a plant we’re losing at least 99.99 percent of the energy that the sun gives us.

Nor is that the end of the waste. Meat is energy-rich, and most of us like it, but making it costs far more energy than we get out of it. For example, in the United States in 2005 we each ate about 200 pounds of meat. To grow that meat, we fed our food animals over half of all the grain we grew. But if we feed a carrot to a rabbit, then eat some rabbit stew, we’re wasting a lot of energy since the rabbit must burn a lot of the energy in that carrot just to exist. Similarly, when a chicken pecks grain, then we eat a chicken sandwich, we’re eating something whose grain equivalent could have fed perhaps ten of us. When we eat seafood, often we’re wasting even more energy. For instance, tuna eat herring, which eat zooplankton, which eat phytoplankton (tiny sea plants). Every stomach in the chain means a roughly tenfold loss of energy. So when we eat some tuna salad, we’re forfeiting lots of energy. In short, when we eat any animal, we give up at least 99.999 percent of all the energy that our planet gets from the sun.

To compensate for all that wasted solar energy, we add yet more energy by digging it out of the ground. For instance, on the Canadian prairies an acre of wheat needs about 80 pounds of nitrogen fertilizer a year. Making that fertilizer costs energy—which usually comes from natural gas. Thus we inject into the Canadian soil about 1.6 million kilocalories of energy per acre per year. And that’s not counting the energy cost of phosphorus, sulfur, and potassium fertilizers. Nor does it count the diesel fuel we burn for tillage and transport. Processing our food burns yet more energy. For instance, it costs three times as much energy to put a can of applesauce on a grocery shelf as it does to put the same amount of apples in the produce department. It takes 15 times as much energy to make a bag of potato chips as it does to make an equal amount of potatoes.

We could do more with less, but we’d first have to change how we think about plants. We get about 80 percent of all our nutrition directly from them, but they aren’t good at feeding us. What they’re good at it is making more of themselves. That is, a plant is a self-building factory. It sucks in raw materials and sunlight to then extrude parts. It builds a structural shell (its stem), and inside that it builds pipes, filters, and hydraulic pumps (its vascular bundles). It also builds storage bins (its roots), and solar cells and gas exchangers (its leaves). Plus it builds factory embryos (its seeds, tubers, or spores) to make more self-building factories.

Those factory embryos are rich in the building materials that future factories will need to start building themselves. A few of them—grains, legumes, nuts, tubers, and such—are relatively big, easy to harvest, and storable. That makes them worth our while to eat—just as we eat hen eggs, but don’t bother with rooster sperm. Sometimes, though, we instead eat the factory’s solar cells, as in spinach and lettuce. Sometimes, as in celery and asparagus, we eat its shell and hydraulic system. We might also eat its storage system, as in carrots and radishes, or its casing (sassafras and cinnamon), its hydraulic fluid (maple), its capsule casing (dandelion, nasturtium, pansy), or the whole thing (sprouts and mushrooms). (Although, a mushroom is a fungus, not a plant; but it, too, is a factory.) A salad is thus really just a mess of factory parts.

Seeing plants as factories helps us see that farming is really a branch of engineering. We already engineer our little green factories in many ways. We aerate their soil. We raise their nitrogen, phosphorus, and potassium supplies. We raise and control their water supply. And we separate them in the soil so that they don’t have to compete for such supplies. So what we’re really doing when farming is manually injecting our cleverness into a plant’s life cycle to get it to do what we want. One day, though, we might pack many of those smarts into the plant’s seed itself.

For example, plants need nitrogen to make their genes and proteins. (We animals need it for the same reason.) Nitrogen makes up 78 percent of the atmosphere, so it might seem like plants wouldn’t have a problem getting it. However, most plants are such saps that they never figured out how to capture it from the air. For them, nitrogen needs to be in the form of ammonia or nitrates in the soil. But those get made only rarely (fire, lightning, rock weathering). Plus, once made, they dissolve easily. So it’s lucky for plants that we animals concentrate nitrogen compounds in our bodies. Thus, to a plant, an animal is just a walking bag of nitrogen (and carbon dioxide). That’s why corpses, urine, and dung can make such good fertilizer. So when we fertilize a plant, we’re mainly giving it more nitrogen.

But we may not need to do that forever. Legumes—like peas, soybean, and clover—long ago solved their nitrogen problem. They don’t suck it in from the air; their roots make a home for nitrogen-fixing microbes. That symbiosis lets them make lots of protein. And that’s where we get most of our protein today. All around the planet we make meals by pairing proteins with carbohydrates. If it’s not soy and rice, it’s bean and maize. If it’s not lentil and potato, it’s bangers and mash—or burger and fries. So one day we might remake our cereals to extract nitrogen as legumes do. If so, they’d no longer need us to get their daily nitrogen fix.

The multicellular photosynthetic autotrophs that we call plants didn’t spend their last 425 million years on land just sunbathing. For example, since they can’t run away, they try to defend themselves from being eaten by lacing their bodies with all sorts of poisons: antibiotics, fungicides, insecticides, herbicides—from which we made nearly all our earliest poisons, perfumes, medicines, and spices. Plants are chemical warfare combat veterans, armed with bioweapons in far larger amounts than the little we add when we spray them with pesticides. For instance, when we eat a chili pepper, what burns our mouth is capsaicin. It’s a neurotoxin. Plants developed it in their weapons labs to deter insect diners. They also adapted to varying soil and rainfall conditions; to variations in cloud cover, temperature, and humidity; to the amount of carbon dioxide in the air; and to many pests. Our monoculture farming has weeded out much of that variation, but it hasn’t gone away. It’s still there in our plants’ wild cousins. As we learn more about their genetic heritage, we may make hybrids with all those survival smarts built in. Such smart crops may be hardy over wider variations in soil and climate, so we may not need to do as much as we have to do today to help them along. We might even be able to cover deserts with such crops.

The ground is already shifting underfoot. As of 2005, new genetic editions of soybean, corn, cotton, canola, squash, and papaya, plus several others, already covered 141 million acres worldwide. Such smart seeds would likely change us, just as mutant grass seeds began changing us 11 millennia ago. For one thing, even fewer of us might need to be farmers as those seeds become more efficient at feeding us. Our use of fertilizers and pesticides and so on may also fall. But as farming becomes more like computer programming, such new seeds may also cost more as they take more effort to debug in a wholly new sense. Seed piracy may also grow since the biggest cost might then become that of designing the seeds in the first place. Real estate prices may also change as smart seeds change the meaning of the word ‘arable.’ We may also, no doubt, make many mistakes—we might even accidentally cover the planet in a new superweed, like today’s kudzu. Engineered plants are just as shameless as any other plant. They’re all hussies and will blithely mate with their wilder cousins out in the bush, creating all sorts of hybrids. However, if all that doesn’t destroy us, more of the planet’s biomass might then turn itself into our food at lower energy cost to us.

Food Machines

Several smart seeds are already here, but no matter how smart our seeds get, our farms are still highly inefficient food factories. For one thing, they are, by far, our biggest water drinkers—and our biggest water wasters. Of all the fresh water we use, about ten percent goes to our homes and around 20 percent to our industries. Our farms gulp 70 percent. Plus, our homes and industries both return up to 90 percent of water used. Our farms return only 30 percent. Most of the rest evaporates, then finds it way to the oceans. That won’t change as long as we irrigate our crops by simply spilling water on the ground.

Our farms drink all that fresh water because our foods are full of water. For instance, an apple is about 84 percent water. A mushroom is about 90 percent water. Lettuce is about 96 percent water. Our processed foods, too, are full of water. An apple pie, a hamburger, and a pizza are each about half water. Much of our food averages about 65 to 75 percent water by weight. So when we grow food, what we’re mostly doing is making tasty water.

Not only are our foods sopping with water, the plants composing them need even more water to grow. Thus, while each of us drinks about a gallon of water a day, the food we eat that same day would have needed even more water. For example, a pound of wheat needs 120 gallons of water to grow—and a pound of rice, 144. Growing the ingredients for a single loaf of bread may need more than two tons of water. Plus, eating meat means consuming even more water. For instance, raising a lamb for six months costs over 22,000 tons of water. So when water-rich New Zealand exports a lamb cutlet to water-poor Saudi Arabia, what it’s mostly exporting isn’t meat, but water.

We, like other animals, are mostly just mobile bags of water—walking globules of the oceans as they existed about half a billion years ago. So we, and our foods, need lots of water. As our numbers rise, so does our need for food, and thus our need for water. But not only are we getting more numerous, we’re also getting richer, and the richer we get the more meat we eat, which means even more water. For instance, during the twentieth century our water consumption rose sixfold, which was almost twice as fast as our numbers grew. Today, in India and China, irrigation alone claims over 80 percent of all water used. In parts of India, China, and the United States, we’re now consuming groundwater faster than it can replenish itself. And Pakistan, Australia, and Spain are reaching the same limits, too. Thus, often when we buy a farm today, mostly we aren’t buying land. We’re buying water rights with some land attached.

But water is only part of our food problem, for our farms today are also, by far, the single biggest modifier of the planet. As of 2000, we gave at least 3.7 billion acres to our crops and at least 8.4 billion acres to our livestock. In total, that was about a third of all available land on earth. We were also losing about 15 million acres of primary forest a year. And we were losing about 4.7 tons of topsoil per acre per year. Our food gathering, on both land and sea, also destroys millions of species. For instance, nine-tenths of all big predatory fish—including tuna, cod, and halibut—have vanished from the continental shelves just since 1950. The earth is having a fire sale, where, apparently, everything must go. Except us. Maybe.

Simply making our seeds smarter likely won’t change much of that. Such a huge change, if one comes, would have to wait until we better understand the proteins that make our food.

All living things on earth are mostly made of the same stuff as a carrot plant. A sperm whale, a bowl of petunias (minus the bowl), Diana Ross and the Supremes, we’re all bags of just a few elements. The big four are: oxygen, carbon, hydrogen, and nitrogen. We’re also made of a little phosphorus, potassium, sulfur, and such. (That’s why those seven are the chief components of fertilizers.) But to feed ourselves we can’t simply down a bucket of carbon and inhale a cubic foot of hydrogen. We need the stuff in particular arrangements. For example, glucose is made of carbon, hydrogen, and oxygen. Swallow some, and our body might combine it with other foods and so convert it into other structures of exactly the same elements plus a few others. Arrange those one way and we get albumin—the chief protein in egg whites. Arrange them another way and we get urea—the chief component of urine. From this point of view, all of us animals are just machines that can convert eggs into urine.

That conversion happens inside our cells. There, thousands of proteins react together to build everything that a living thing needs, and about half our body’s dry weight is protein. (That’s why nitrogen, which makes up a big part of all proteins, is so important.) In essence, those proteins are cellular machines, and, if we knew what we were doing, we might one day reprogram the genes that specify them to specify new proteins. We might, for instance, tell a microbe how to eat plastics, thus making our plastics self-disposing. We might also make a microbe that cleans up oil spills. Or one that clears clogged arteries. Or one that makes steaks.

Making a steak isn’t hard—if you’re a cow. But we haven’t yet figured out how to do it without a cow. It’s hard to tease apart the effects of thousands of proteins working together inside the cell. That’s a bit like crashing a big party and identifying circles of friends by seeing who jabbers at whom, then watching each circle to see who gossips most, and what the topics are. (Well, it’s not really that simple. For example, topics can change depending on nearby topics. But that’s the general idea.) And we now have a tool that simplifies all that cross-checking—the computer. The next problem is that of making a life-support system for synthetic genes to copy themselves in. If we solve that, we might start making foods in factories.

We’ve already tiptoed in that direction by redesigning several microbes. For example, we’ve remade some to build wholly synthetic proteins. We’ve built viruses from scratch. We’ve changed cells to build themselves simple digital circuits. We’ve made cells that eat sugar and squirt diesel oil. We’ve even made synthetic cells that carry corporate logos in their genomes. Someday we’ll likely be able to fool cow cells into making a hunk of cow—without a cow. In essence, a cow is a walking fermentation vat that turns grass and air into muscle and milk. Do we really need the rest of the cow?

“Don’t look now, Elsie, but have you noticed anything
odd about Bessie since she got back from the vet?”

Today we may think that we’ve industrialized our farms, but all we’ve really done is mechanized them. We don’t make our food. We watch it being made by small factories inside a large, automated factory that we call the earth. We fiddle with its knobs and dials, but mostly it does the job for us. Only now do we know enough about genes to begin to make our own food. But that can’t happen overnight. Plus, if we ever do make food machines, our first ones might well be too costly, awkward, or unsafe for most homes. So at first they’ll likely only shift a little food production from farms to factories. But over time we might slowly shift from growing potatoes to extruding potatoes. Of course, just as plastics at first faced strong pressure to look just like the things they replaced, such potato products will likely be under strong pressure to look just like lumpy potatoes that came out of the earth. One day, though, we might well get used to potatoes that come from a factory. New potato styles—new colors, textures, shapes, sizes—won’t scare us anymore. The same sorts of things might happen to meat and milk and eggs and all sorts of other food factories. And perhaps once we get there we won’t need to expend quite so much energy, water, and land—and kill so many species—as we do now.

Future Tense

Given how our swarm has helped shape our food supply in the past, what might our near future food supply be? As of 2006, about 17,000 of our children under five starved to death each day, and over 920 million of us were hungry, living lives not much different from those that nearly all of us lived in 1300. Today, even in our rich countries, where there’s more than enough for all of us to eat, if we’re really poor we can still sometimes go hungry without enough money—or a gun. Were we to one day have food machines, they might lead to huge changes both in our food supply and in our effect on the planet, but how likely are we to develop them anytime soon?

From the point of view of physics, food is just the fuel that our body uses to convert solar energy into chemical energy. And our body’s energy needs aren’t all that great. Today, worldwide, the average adult human male consumes only about 2,700 kilocalories of food energy a day. (An adult female consumes about 2,000.) That’s the same amount of energy that we’d get by burning around a pound of coal. Today, that much coal costs roughly five cents U.S., and just about all of us could afford that Were physics all that mattered, feeding ourselves would be easy and cheap and none of us would starve. Since that’s not so, or not yet so anyway, the laws of physics must not be the only things that matter. So what else does?

First off, were such devices to come to exist, they’d bring wrenching change, and many of us won’t like that. For example, our present food producers might well fight such a possible future, just as foragers fought farmers in the long ago. Many of us in businesses devoted to food serving, handling, processing, and transport may also fight the change, perhaps taking to the streets to warn the rest of us of the new horror of frankenfoods. To nobody will it matter that we’ve been eating frankenfoods for thousands of years already; we’re so used to them that we don’t see them that way. Also, food consumers, which is all of us, would probably be rigid with fear. Many of us would probably oppose any change in something as vital as food. So politics, not physics, might well decide whether food machines become widespread.

Also, in countries with good transport networks, economies of scale may well work against home food machines, just as they do today for things like water and power. Few of us have our own wells or generators; it’s usually cheaper to centralize. So economics, not engineering, might well decide whether food machines become more like power stations or more like fridges.

Further, legal and perhaps even military needs may also matter. For example, were food machines to come to exist, food snobs would likely continue to pay large sums for hand-grown food. Elites would likely continue to eat salted fish eggs laid only in special rivers. And connoisseurs would likely continue to quaff hand-made fermented grape juice grown only in specially blessed earth. If something is machine-made but expensive those of us who are rich might well want it, but once it gets cheap, we likely won’t. None of us will want our costly foods tainted with anything cheap any more than cheap color copiers made currency free. The whole point of being rich is to get things that others can’t. Also, food machines would almost surely soon lead to new foods, and thus new fears. That’s already led (since 2001) to bans in Europe on engineered seeds, which has led to constraints on countries in Africa that want to export food to Europe. For example, in 2002 in Malawi, Mozambique, Zambia, and Zimbabwe we were facing famine, but we still refused food aid from the United States because some of that food was engineered. We feared that Europe would then ban our own food crops. So strictures on the use of food machines both locally and globally might well follow their introduction.

Even further, were food machines to come to exist, as partly digital devices, they could crash. None of us will want a famine caused by that kind of bug. Plus, even though they might live in factories and kitchens, not fields and gardens, as food sources they’d still have pests. Rodents and roaches would still be hungry. Also, as computerized devices, someone, somewhere, would likely hack one. It might then be used to make homemade cocaine, mescaline, or heroin—or gunpowder, gelignite, or nerve gas. A food machine needn’t only make food; it would be a disguised organic matter converter. Even were we to use it only to make food, for a select few that food might be—who knows?—human flesh. Presumably, clientele for long pork would be quite exclusive. Were any of that to happen, it would scare lots of us, and we would respond in our usual ways.

While all such problems might smooth out over time, they all depend on the existence of food machines in the first place. But what might drive us to create one? Today, food is a huge cost for our poor countries, but a small one for our rich countries. For example, as of 2003 in the United States, average food cost was around ten percent of income. In Eritrea, it was as much as 71 percent. If eating costs us ten cents on the dollar, and food prices double, we grumble. But if eating costs us 71 cents on the dollar, and food prices double, we die. For instance, when global food prices rose in 2007-2008, in Britain we had to pay, on average, about ten percent more—around £2 a day—for our food. Many of us complained, but nothing more. In India, though, the same change meant that 40 million more of us fell into poverty. Food riots in West Bengal followed. Around the planet, that single food price hike meant that 75 million more of us went hungry. In short, those of us who most need food machines can’t afford to make them, and those of us who might afford them, don’t care.

But wait, we all care about starving babies, don’t we? Well, we sure say we do. But based on how we spend our money, in our rich countries we care far more about being too fat. For instance, as of 2007 in the United States, around two-thirds of us were too fat. In Britain, about six in every ten of us were too fat. In Canada and Germany, at least half of us were too fat. All our rich countries are bloating. Meanwhile, around the world, of every ten of us who die each year, six die of hunger-related causes.

In our rich nations we’re too fat because we have cheap food, and we have cheap food because we have large and reliable food supplies. We’ve built a huge toolbase to ensure that. It consists of costly machines, trained farmers, and extensive credit systems. Those support huge equipment purchase and training, and vast storage facilities and transport networks. Add to that many schools and research institutes for farming. Also add fast, long-range data exchange, and a wealthy and stable—and literate—home market. Don’t forget large and powerful standing armies in case we need to go fetch cheap resources from other countries. Meanwhile, in our poor countries we have little of that toolbase. We also have far less incentive to increase our food supply. What’s the point? We couldn’t sell more of it abroad anyway—rich countries won’t let us. Trade barriers shut out much foreign food.

For example, In 2006, our rich countries gave $372 billion U.S. in subsidies to our farmers there. That’s over a billion dollars a day. As a result, the United States exports cotton far below cost. The European Union exports sugar far below cost. Japan marks up foreign rice so much that it costs between four and ten times more than local rice. Meanwhile, tens of millions of us in Benin, Burkina Faso, Chad, Mali, and Togo are starving. Why? Given our climate, and our toolbases, we can only grow cotton, sugar, and rice. We work for some of the lowest wages on the planet, yet still we can’t compete. For instance, in 2003, every cow in Europe got about $2 U.S. a day in subsidy—and that was twice as much as half of us in Africa got that same day. In effect, in rich countries we pay domestic farmers to starve foreign farmers.

So millions of our babies starve to death, yet we aren’t trying to invent food machines to give them milk. Instead, we’re spending billions forcing them to continue to starve to death. It thus seems likely that if we never get cheap food machines it won’t be because the laws of physics prevent us from building them. More likely it will be because our rich countries don’t really want them. We thus seem likely to build machines that make mountains of expensive, empty food—fat-free, calorie-free, taste-free—long before we build any that make cheap, nutritious food.

However, while all those forces may well work against food machines, we might still produce them, if only by accident. After all, exactly that has happened many times before. For example, the industrial phase change that went on to give us tractors had nothing to do with farming. As much as anything else, it started with an ironmonger in Cornwall trying to pump water out of tin mines. Or again, our first fridge had nothing to do with preserving food. It started with a doctor in Florida trying to prevent malaria. Similarly, our first tin can wasn’t made of tin—it was a champagne bottle. We got it when a cook in Paris found a new way to preserve fruits for his candy business. Our first synthetic fertilizers, too, were also accidental. They fell out by mistake when a chemist in France tried to make cheap diamonds, an inventor in Canada tried to make cheap aluminum, and two dyers in Germany tried to extract gold more cheaply. None of them had planned to make fertilizer.

Further, were food machines to come to exist, many of us needn’t even notice them. We might continue to buy our food in grocery stores and markets, except that they might get some of their food from factories rather than farms. That’s already happened with Quorn, a fake meat made from vats of fungus that’s been selling since 1985. Today a few of us are planning to do the same with vat-grown pork from pig stem cells. But we don’t yet know how to do it cheaply, so were such meat-sheets to go on sale today they’d cost over $1,000 U.S. a pound. One day, though, that price might drop to $1 U.S. a pound. If so, such food builders might then hop from factories to stores. Limited use in rich homes might then be just a question of time. The word ‘homemade’ might then gain a whole new meaning. However, a meat product that one day might cost $1 a pound, but that today costs $1,000 a pound, has little chance to come to exist anytime soon given that in our rich countries today beef can cost less than $3 a pound.

In short, the laws of physics likely don’t control whether or not we’ll get food machines anytime soon. The laws of ‘swarm physics’ seem likely to matter more. Yet we know far more about the laws of physics than we do about the laws of swarm physics.

Learning more about that kind of ‘physics’ seems especially urgent today because our swarm is now in the midst of another phase change based on food. That change started a few centuries ago as we began to develop new industrial tools, but it’s far from over. More change is likely ahead because we’re now leaving our agrarian age and entering our urban age. For the first time in history, half of us now live in cities, and well over a billion of us now live without food fear. By 2015, another half-billion of us will likely live that way. By 2030, we may be almost two-thirds urban. By 2050, we’ll likely peak at roughly nine billion, and probably most of us will be urban. Our numbers are rising and so are our incomes, and thus so are our food demands.

Thus, autocatalysis seems once again to be driving us to change, maybe even phase change. As our food demands—and thus energy, water, and land demands—rise, we’re going to be under more and more pressure to change. Large, long-term trends seem to be converging on a point somewhere not too far into our future. We seem to be nearing a crisis point.

But maybe we’ll keep inventing ourselves out of the way of onrushing calamity, for, despite our many catastrophes, self-inflicted or otherwise, we’ve already changed our food supply a lot. For example, in 1998 Eritrea was one of our poorest countries, with three in four of us there undernourished—the highest figure in the world at the time. Yet in Eritrea in 1998 we got more kilocalories per person than we did in France in 1705. Similarly, India in 1998 ate better than Britain did in 1850. As of 2008, about one in seven of us were starving; but in 1970 about one in four of us were. Just since then, our proportion of poor has dropped, our average income has doubled, our infant deaths have halved. In our poor countries, over four in five of us now have at least adequate diets. In all, in our poor countries from 1961 to 1992, our numbers doubled, yet our per-person daily kilocalorie intake still jumped by a third.

Once upon a time we invented pots to protect grain from rats. Then we found that we could also use them to make porridge. (Or perhaps it happened the other way around.) Once we did, our teeth-breaking ended and our food stores grew, so our numbers multiplied—and farming spread. That’s nothing to do with anything we planned. Likely, the same sort of network forces will be at work on us in future. Today we’re beginning to think of extruding potatoes instead of growing potatoes. One day we might have fridges that make milk and eggs, rather than merely keeping them cold. Such food machines might even go into our spaceships. Over time, dirt farmers may become increasingly weird throwbacks to some past unthinkable time. Perhaps they’ll be left to grow only the bulk plant-products we need for fibers, fabrics, timber, paper, rubber, spices, and livestock feed. If that’s our future, much of our farming industry might well go away. Perhaps the poorest half of us would then finally be free to do something other than farm all day or starve to death. But while we seem to be approaching the technological capability to do that, that needn’t mean that we’ll actually do it.

In sum, we seem to be subject to large, long-term network forces that act on us, sometimes (often?) contrary to our will, or even our imagining. From the moment that one of us thought that certain grass seeds weren’t half bad when pounded and baked, we entered into a covenant with the soil. It hasn’t ended. That covenant cuts both ways, though. Since the first ragged flint sickle tore through a sheaf of wild grass millennia ago we’ve shaped our plants and they’ve shaped us. Through all that time we knew neither what we were doing, nor where we were going. We don’t seem to be any different today. We’re always stumbling into the future, and mostly we’re looking at our sore feet, not the distant horizon. Ignorant of our swarm’s physics, that has been our way for millennia. Learning more about that physics might or might not change that, but if we continue in ignorance of it, it seems likely that at least 17,000 more of our babies will continue to die of hunger-related causes today—and tomorrow—and the next day—and in the minute it takes to read this paragraph, 12 more of our babies will be dead.