A thousand orange, pumpkin seed–sized ants raced across the earth so quickly they seemed to fly over the stones. They were arranged in a phalanx 4 to 5 meters long and 25 centimeters wide, exhibiting the orderliness of a military parade conducted at a full run. I followed the raid’s almost arrow-straight course up an embankment and onto sandy ground dotted with rock and scraggly patches of sage, huckleberry oak, and powdery-leaved mule’s ears. There, the column began to disintegrate. My companion, Alex Wild, then a graduate student at the University of California at Davis, warned me to watch closely. Ceasing their coordinated advance, the workers spread out over a couple of square meters in an area occupied by the ant Formica argentea and started scouring every hole and crevice for the Formica nest. After five minutes of frenetic searching, they began to emerge from alongside a stone, hefting silky objects—pupae stolen from the Formica.
In their staging of concerted attacks on other ant colonies, the orange ants resemble a species unrelated to it, the army ants, but with a critical difference: this species, Polyergus breviceps, does not eat the plundered ant brood; rather, it keeps the young alive to raise as slaves.
It isn’t necessary to mount expeditions to the remote tropics to see dramatic behavior in ants. Slavery, curiously enough, is known only for species living in the temperate zones. Polyergus breviceps is one of five Polyergus species that jointly range across North America, Europe, Russia, and Japan.1 All of them enslave ants belonging to the related genus Formica. The slavemaking ants of this group are commonly known as Amazon ants, after the mythical female warriors who were said to steal children and make them their own.
Orange-colored Amazons continued to pour in and out of the black fissure next to the stone. A few centimeters away, soil flew as a pugnacious pack of them used their forelegs to dig at another entryway. In the face of the onslaught, the Formica had blockaded this backup passage with soil transplanted from deeper in the nest—an approach to survival I have witnessed also among the recalcitrant victims of army ant sieges.
All Alex and I could see on the surface was the coming and going of Amazon workers and the occasional Formica walking by without putting up a fight. Curious, I overturned the stone, exposing the nest galleries. The only sign of struggle was a Formica who had just attempted to wrestle a pupa from the jaws of an absconding Amazon; she was locked in combat with two assailants who were using their saber teeth to crack her like an egg. There was no massacre going on because the only opposition the Formica argentea are known to raise against the raiding gang is a blockade. If the slavemakers had failed to breach the Formica’s piled sand, they would have returned home empty-handed. But once the Amazons stormed this stopgap fortification, there was hardly a scuffle. The Formica stood back and let the Polyergus take whatever they could lay their greedy little mandibles on.
All this action was happening one late afternoon at Sagehen Creek Field Station, on the eastern slope of the Sierra Nevada north of Lake Tahoe. Starker Leopold, son of the renowned conservationist Aldo Leopold, established Sagehen on U.S. Forest Service land in 1951. The rolling hills below a robin’s-egg blue sky were sprinkled with lodgepole and Jeffrey pines. It was mid-July, the raiding season for the Amazon ants in this region.
If this had been an army ant raid, the column would likely have continued to advance while part of the workforce stayed behind to attack the nest. But instead the entire Amazon regiment had come to a halt, targeting this locale. While the charge was still under way, the Amazons started to ferry hundreds of pupae and a few larvae homeward in ragged formation, and within minutes the looting operation began to close down. Alex and I followed a long, unbroken caravan back to the road where we had first come upon the more tight-knit advancing platoon. From there, the returning column stretched an additional 50 meters until it headed below-ground, into its own nest, where I again saw scattered silver-black Formica workers. Although they looked identical to the ones the Amazon workers had just raided, these individuals were, in fact, Amazon slaves.
What I saw next sent a shiver down my spine. When a Formica slave encountered a returning Amazon, she tugged at the pupa the Polyergus held. The Amazon relinquished her grip, allowing the Formica to take from her the Formica just stolen from its nest. Other Formica slaves, not occupied in hoisting pupae, were actually moving slavemakers: tired out, I imagined, by the conquest or the long journey home, some Polyergus were allowing themselves to be carried the last leg of the trip. Normally this form of adult transport is associated with novice workers being relocated during nest migrations, not with adult warriors fully able to get around on their own. But here the enslaved Formica workers were transporting the victorious slavemakers as if they were royalty curled up comfortably in their palanquins.
For the average Amazon ant, the royal treatment continues after she arrives home. Entering the nest, she lounges around, at most grooming herself or her nestmates, while the Formica slaves tend to her needs. Her daily efforts last for a couple of hours at most. Yet Alex told me only about half the slavemakers are likely to go on a raid; the rest stay behind, doing nothing all day. For a booming society of several thousand ants the tempo of life is abnormally laid-back for Polyergus breviceps. (To our way of thinking, the average ant worker’s life sounds like slavery even when she is in her birth nest, though in that case she is at least toiling for the benefit of her mother, the queen.)
That afternoon’s confrontation at the Formica nest was a soundless blur lasting twenty minutes, more or less the normal length of such a slave raid attack. The timing was typical too: conducting their raids late in the day, Polyergus is forced to get the job done and head home before the sun wanes, since they don’t stay out after dark. It is unclear why the raiders don’t give themselves more time by beginning earlier. Perhaps the delicate pupae would cook if trundled away in the midday sun. Or perhaps as the afternoon cools the Formica ants fetch brood from deep in the nest to warm nursery chambers near the surface, and so the slavemakers don’t have to invade as far underground, thus minimizing their exertions even further.2
A mature Amazon nest houses five thousand ants, comprising both the slavemaker workers and their more numerous slaves. Inside the nest, the slaves tend the stolen pupae until they transform into adults. Like hatchling birds that imprint on their parents, a young ant quickly learns to recognize the individuals around her and thereafter treat them as family. This imprinting is based on the scent of other pismires, an archaic term for an ant that derives from a colony’s pungent odor. Whenever ants meet, they sweep their antennae over each other to confirm the presence of the blend of compounds that identifies their nestmates.3 If the odor matches expectations, they treat each other as sisters-in-arms. If an individual smells wrong, the workers will either run away from each other or fight.
In most ant species, this imprinting is infallible, because the youngsters are surrounded by sisters in the nest of their mother, the queen. But when slaves-to-be mature in an Amazon nest, they imprint on their captors. Assimilated into the wrong society, the ants are duped into a life of servitude, doing all the drudge work their masters won’t: building nests, foraging for prey, harvesting honeydew, slaying free-living Formica that enter their territory, and taking care of the brood. The Amazon slavers’ only job is to go on raids, replenishing the store of Formica pupae as their enslaved workers age and die.
The slavemakers do so little for themselves that when I pulled a sandwich from my backpack and dropped a bit of turkey in front of an Amazon worker, she walked right past it. Incapable even of recognizing a meal, she is unable to feed herself. Eventually one of the slaves found the stash of poultry and retrieved it to the nest.
At once more brawny than a slave and yet as helpless as a baby, the Amazon worker gets her sustenance only after her servants find food and, like birds with their nestlings, regurgitate it to her. She can neither excavate tunnels nor raise the queen’s young. She is a fighting machine, nothing more. The curved daggers she bears as jaws are useless for any chore except assaulting free-living Formica, but they deliver the ultimate in all-purpose tools: a new stash of slaves. Even with their superior armaments, though, the Amazons are so outnumbered they would be massacred if it weren’t for a chemical known as a propaganda substance that they wield as a social weapon, released from a gland associated with their mandibles, which throws the bombarded colony into mayhem and flight.4
Ant slaves harvesting a dead grasshopper as one of their Amazon “masters” walks by in the foreground. The Amazon ants would starve if their slaves didn’t feed them.
The tolerance of Formica argentea to frequent raids is a sign that this species shares a long history with Polyergus breviceps.5 After countless generations of attack and counterattack at Sagehen Creek, the Formica have apparently come to treat their losses as a cost of doing business. “Resistance is futile,” declared the species-enslaving Star Trek creatures known as the Borg, who make decisions collectively, like ants.
LIFE IN A NUTSHELL
A year later, far from California, I found myself studying a second, unrelated species of slavemaker ant—Protomognathus americanus—in the hope of experiencing some of the variety of ant slavemaking behavior. For the second day in a row, I stooped over the little Protomognathus ants, trying not to disturb the action. One of the dark-pigmented slavemakers had been about to find an acorn housing a nest of Temnothorax when two Temnothorax workers managed to sting the slightly larger ant to death. That had been the last of several close calls for the free-living Temnothorax, two of which lay nearby, killed in earlier confrontations. Another hour crept by, as my arms cramped under the weight of the camera, before a slavemaker with better luck found her way into the acorn through a split in its side. Dozens of the Temnothorax within immediately fled, each grabbing a larva or pupa in the stampede, with the queen beside them. Left alone in the empty shell, the slavemaker stood on a heap of abandoned brood. After a few idle moments she picked up a pupa and returned to her own acorn. She was soon back with reinforcements, who helped her abscond with most of the remaining stock of future slaves.
An acorn from Ohio containing a colony of dark brown Protomognathus americanus slavemaker ants and their orange Temnothorax slaves.
Glancing around, I was struck by the unreality of the situation: all this time I had been so caught up in the action that I had forgotten I was in a laboratory, surrounded by Petri dishes and Bunsen burners. To learn about slavery in a nutshell, I had come to the Mecca of acorn-ant research: Ohio State University, base of Joan Herbers. Joan specializes in Protomognathus americanus, which enslaves three species of Temnothorax in the temperate deciduous forests and yards of eastern North America.6 It is difficult to observe raids of these pygmies in nature. Before I flew to Columbus, Joan had been kind enough to sort out some colonies for me, collected by her students and encased in Ziploc bags. Each bag contained either a slavemaker colony or a colony of free-living Temnothorax, housed in an acorn. All I had to do was put a mix of these acorns in a plastic arena, settle down in front of it, and wait.
Fifteen hours later, I had finished documenting my first Protomognathus slave raid. Over the course of an hour, the slavemakers had taken part of the Temnothorax brood to their old acorn, while expanding their colony into the new one as well. Having multiple nests like this is called polydomy, and it is common among ants that live in acorns and other small, convenient places. Meanwhile, the Temnothorax adults were still scattered over the ground, having lost both progeny and home.
To bring troops to a Temnothorax colony, these slavemakers employ a variation on something called tandem running, a follow-the-leader approach to recruitment in which an ant tracks the successful scout to a site by touching her repeatedly or, if they lose contact, by orienting to a short-range pheromone released by the first ant. Because the leader is responsive to the follower, stopping at intervals to wait for her touch, the relation has been likened to that between teacher and student.7 With Protomognathus, the “teacher” brings along a whole class, for a conga line of several nestmates follows the successful scout.8
The Amazon ants, Polyergus, belong to the Formicinae, a group of ants that includes the carpenter ants and their relations. The Formicinae evolved slavemaking several times independently in different species in different locations. Protomognathus belongs to the Myrmicinae, a second large group in which slavemaking is common.9 The Temnothorax species it enslaves commonly reside in fallen acorns that have been opened up by one of two acorn specialists, the acorn moth or the acorn weevil. The adult females of both these insects lay eggs on or in the nuts; the larva then eats part of the meat before chewing a hole in the nutshell, from which it emerges. The exit hole becomes the entryway for a succession of motley residents, often culminating in Temnothorax.
I described this array of relationships for National Geographic magazine while I was a graduate student looking for cool projects in my neighborhood.10 In researching that article, I spent a lot of time gathering acorns and dropping them in water. Those with residents float because of the eaten-out cavities. Cracking them open—and exercising patience—I eventually uncovered a whole society: several dozen Temnothorax with a queen and pale brood, occupying hollows carved in the nut. After hundreds more acorns, I came upon a mixed colony with two forms of worker—much scarcer. One kind (with a bigger head, stronger jaws, and a groove along each side of the head into which she withdraws her antennae for battle) was the Protomognathus; the other was a Temnothorax worker—in this circumstance, a slave.
Slavemakers like the Amazon and the acorn ants are known as social parasites. They acquire nutrients not by tapping into an organism’s tissues, in the manner of a tapeworm, but by exploiting the selfless, cooperative behavior of a host animal or, in the ants’ case, host society, as one superorganism exploiting another.11 Social parasites escape the burden of foraging by letting their captives collect food. Slavery is just one means to this end. In some ant species that share a nest, the two groups of ants benefit equally from the housing arrangement; as we have seen, the acrobat ant Crematogaster levior and the carpenter ant Camponotus femoratus raise plants together in canopy gardens. At another extreme are colonies that occupy adjoining chambers in a nest, with one kind of ant soliciting food from the other or surviving on the other’s rubbish. The minute and stealthy thief ant nests in the walls of the chambers of larger ants, infiltrating to steal food and brood. The British banker and naturalist Sir John Lubbock found this social parasite appalling. In 1883 he wrote, “It is as if we had small dwarfs about eighteen inches to two feet long, harbouring in the walls of our houses, and every now and then carrying off some of our children into their horrid dens.”12
In Ohio, while I was watching ants come and go from acorns, Joan Herbers and I talked about ants and people. Naturalists have referred to abducted ants as “slaves” ever since Swiss entomologist Pierre Huber first used the term to describe the behavior in 1810.13 Darwin devoted several pages of his Origin of Species chapter on instinct to a discussion of what he characterized as the “remarkable” “slavemaking” activities of certain species of ants.14 Although the analogy is not perfect, it has become established in the literature.
Ant slavery has notable differences from human slavery. Ants, lacking commerce between societies, don’t buy and sell or trade slaves from colony to colony. Ant species such as the Amazon ant are more dependent on their slaves than humans have been, aside from a few “slave societies” such as the Roman Empire at the time of Augustus.15 Ant slaves can’t breed (but then, they fare no better in their birth society: in ant colonies, usually only the queen has the privilege of reproducing). Ant slaves also seem remarkably acquiescent about their subjugation. Only two species exhibit signs of mutiny: some acorn-dweller slaves will, if not outright revolt, at least undermine the colony by killing their masters’ pupae.16 (Though this could simply be their normal response to finding something strange in the brood pile—some brood just doesn’t smell right.) And slaves of the Amazon ant in Europe will make a getaway at times, with some running off to form satellite nests, even adopting a nest-founding queen of their own species should one pass by. Independence is usually short-lived, however, because the Amazons retaliate with periodic raids to retrieve the escapees, which engage in restrained fighting that quickly leads to acquiescence.17 Other than in these situations, ant slaves seldom try to thwart their captors or attempt to escape them; they die in the defense of their masters exactly as they would have done in their birth nest.18
Kidnapped before they have formulated an identity, the victims imprint unconditionally on their captors through ignorance, not brainwashing. They are similar to the human working class described by Karl Marx—a whole population whose efforts are misdirected to benefit an oppressor.19 As slaves, the ants have lost not freedom (which they never had) but the biological imperative to raise the offspring from their own genetic family.
The indolence of ant slavemakers relies on their captives being programmed to slog through the day without objection, which they do. And just as ant slaves seem oblivious to the fact that they are in a slavemaker colony, slavemakers might not be able to distinguish slaves from their own kin.
For some ants, this is hardly surprising. In the red imported fire ant and some other species, one colony will raid a smaller colony of their own kind and rear their brood as slaves.20 But Amazon ants and certain other social parasites enslave not their own kind but related species that have similar diets, nests, and communication systems. In these cases, the activities that come instinctively to a slave are of continuing use, allowing the adjustment to slave life to occur rapidly.21 No domestication through breeding is needed, as occurred with humans and dogs, nor is training or coercion necessary, as it is, for instance, with human prisoners in a chain gang.
Instead of taking slaves outright, human civilizations bent on expansion have often usurped villages and exacted tribute and labor from them while expanding their dominion to encompass the vanquished people’s land. The losers are often allowed to remain with their families and communities; unlike with slaves, their former identities are not completely lost. With time and luck, they may even be incorporated into the victorious society as full citizens. This middle ground of empire building, which requires a large population of victors to quell rebellions, is unknown for ants, for whom surrender followed by a midlife switch in social allegiance is not possible. As part of the spoils of war, ants either take slaves or kill the losers (in which case cannibalism is frequent, as it was in the early stages of human warfare).22 Though the victors commonly reduce the defeated colony in size, they seldom destroy it—as we saw also for army ants, which raid a nest to the point of diminishing returns, then leave the remnants. Unless a colony is weak or its queen is killed, it will likely see another day.
Are ants and humans the only animals that have slavery? Female primates may capture or enforce the adoption of an infant. In Old World monkeys with hierarchies of female dominance, such as the Lowe’s guenon of West Africa and the Bonnet macaque of India, a female may take a baby from a low-ranking mother, possibly to interfere with a competitor or simply because she is attracted to the infant. Female langur monkeys share their young, but an inexperienced juvenile impatient to get her hands on a baby may abduct one from another troop. In no primate, though, is the abducted individual a source of forced labor.23
Much more analogous to the ant model is the activity of a large Australian bird, the white-winged chough. During the four years it needs to reach maturity, a chough stays with its parents and helps them raise its siblings. Without enough assistants, the parents will be unable to adequately build their elaborate mud nest, incubate the eggs, and feed the chicks. In some situations, the parents or their immature helpers will bully a neighboring family until one or more of its fledglings can be shepherded to their own nest, where the new youngsters are raised until they can serve as helpers. Just like ant slaves, fledglings don’t recognize that they’ve been abducted.24
Given our differences from animals, is it reasonable to apply the word slavery to ant practices? Most words usefully encompass a variety of phenomena, and slavery, like many of our names for things, was used first and foremost to define human relationships, before being applied by analogy to the natural world. But just as slavery as practiced in Augustan Rome doesn’t correspond exactly to the behavior of an ant, neither is it the same as slavery in other human societies (consider today’s largely secret slave trade); similarly, the behavior of one ant species does not match that of any other. The attributes of slavemakers vary even from place to place in a single species of ant.25
Joan offers piracy as a more neutral word for this activity in ants.26 Yet piracy, though it involves robbing or plundering, inaccurately describes the ants’ behavior. After all, if pirates forced a person into a life of servitude, that person would be described as a slave. Retired Hunter College professor and Amazon ant expert Howard Topoff quips that adoption may be a better word, but this term isn’t used for situations so insidious as to entail theft followed by a life of travail for another’s benefit.27
I don’t have any problem with using the term slavery for ant behavior unless people commit the naturalistic fallacy—assuming that if a behavior exists in nature, it is somehow “natural” and therefore morally acceptable in human society. The earliest humans were hunter-gatherers organized around equality, not dominance, and had little use for slaves and limited capacity to keep them, which suggests that slavery is not a part of our own heritage.28 Thus if, as Edward O. Wilson and Bert Hölldobler propose, Marxism is better suited to ants than to humans, then, by orders of magnitude, slavery is even less well suited to the human social order.29 While apes and some other vertebrates have been known to express empathy and to act in accordance with rudimentary moral standards, ants do not.30 Regardless of the power of Aesop’s fables, among animals other than ourselves, actions are neither right nor wrong. They just are.
PROPAGATING A SLAVE COLONY
Given the slavemaker penchant for avoiding work, how does an Amazon queen go about founding a new colony?31 I caught a glimpse of the first step during the raid at Sagehen Creek. Among the onrushing workers ran ten queens, their cellophane wings glittering. En route they left the mass to scramble up tufts of grass. At such raised locations unmated queens attract males, using pheromone secretions from their mandibular glands.
After mating, the queen has a choice: she can continue with the raiding party and establish her colony in what remains of the Formica nest after the slavemaker workers have plundered it; or she can strike out on her own to locate a different Formica nest and found her colony there. The first choice has its risks. If she moves into the conquered nest, ants from her original colony might come back later to raid the site again. Retaining no memory that she is a relative, they would kill her burgeoning family.
In either case, the Amazon queen rushes the nest with savage fury, shoving aside any Formica workers that get in her way. She is protected from their attacks by both a tough exoskeleton and repellent secretions.32 Leaping on the Formica queen, the Amazon punctures her counterpart repeatedly with her dagger jaws and then licks the fluids draining from the dying queen’s body. The transformation of the colony is almost instantaneous; mere moments after their queen’s death, her workers undergo what Howard Topoff has called brainwashing: “The Formica workers behave as if sedated. They calmly approach the Polyergus queen and start grooming her—just as they did their own queen. The Polyergus queen, in turn, assembles the scattered Formica pupae into a neat pile and stands triumphantly on top of it. At this point, colony takeover is a done deal.”33
Because the ants in a colony imprint on each other’s scent, licking the dying queen is the slavemaker queen’s macabre way of applying the home-grown perfume. Once she has the colony’s identifying odor, the invader is accepted as one of its own. The opposite is true as well: if the mother queen is removed from a Formica colony prior to the arrival of an Amazon queen, the impostor has no way to appropriate the local scent, and the workers will bite her until she dies.34
The coup d’état does not end with the first queen’s murder. Formica argentea colonies have multiple egg-laying queens, which the slavemaker queen roots out from their safe havens in the nest and executes one after another. Why this is done is not known. Having procured the correct odor by killing one queen, she could commute the death sentences of the others and leave them to produce more workers, which is the resource she will need from the colony. The Amazon queen would thrive without her own workers, and her species would eventually evolve to lose the worker caste entirely, enabling her to concentrate on laying eggs that will grow into future parasitic queens.
Some ants employ that very strategy. Teleutomyrmex schneideri is a European species that produces no workers. The queen infiltrates a colony of a Tetramorium pavement ant, and her concave undersurface enables her to attach herself to the back of the resident queen, where she clings for the rest of her life.35
Polyergus queens, however, have retained their workers. I assume there must be a downside to keeping a breeding stock of slave queens around. Perhaps Amazon nests that retain a slave queen are subject to insurrection, in which case it wouldn’t pay for the slavemaker to share reproductive control, the ultimate source of power in an ant society.