But machines are ugly too, and we sense in them a threat to things we hold dear. Machines may be a conduit of human power, but that power has usually been wielded by the industrialists and financiers who own the contraptions, not the people paid to operate them. Machines are cold and mindless, and in their obedience to scripted routines we see an image of society’s darker possibilities. If machines bring something human to the alien cosmos, they also bring something alien to the human world. The mathematician and philosopher Bertrand Russell put it succinctly in a 1924 essay: “Machines are worshipped because they are beautiful and valued because they confer power; they are hated because they are hideous and loathed because they impose slavery.” 6
As Russell’s comment suggests, the tension in Macmillan’s view of automated machines—they’d either destroy us or redeem us, liberate us or enslave us—has a long history. The same tension has run through popular reactions to factory machinery since the start of the Industrial Revolution more than two centuries ago. While many of our forebears celebrated the arrival of mechanized production, seeing it as a symbol of progress and a guarantor of prosperity, others worried that machines would steal their jobs and even their souls. Ever since, the story of technology has been one of rapid, often disorienting change. Thanks to the ingenuity of our inventors and entrepreneurs, hardly a decade has passed without the arrival of new, more elaborate, and more capable machinery. Yet our ambivalence toward these fabulous creations, creations of our own hands and minds, has remained a constant. It’s almost as if in looking at a machine we see, if only dimly, something about ourselves that we don’t quite trust.
In his 1776 masterwork The Wealth of Nations, the foundational text of free enterprise, Adam Smith praised the great variety of “very pretty machines” that manufacturers were installing to “facilitate and abridge labour.” By enabling “one man to do the work of many,” he predicted, mechanization would provide a great boost to industrial productivity.7 Factory owners would earn more profits, which they would then invest in expanding their operations—building more plants, buying more machines, hiring more employees. Each individual machine’s abridgment of labor, far from being bad for workers, would actually stimulate demand for labor in the long run.
Other thinkers embraced and extended Smith’s assessment. Thanks to the higher productivity made possible by labor-saving equipment, they predicted, jobs would multiply, wages would go up, and prices of goods would come down. Workers would have some extra cash in their pockets, which they would use to purchase products from the manufacturers that employed them. That would provide yet more capital for industrial expansion. In this way, mechanization would help set in motion a virtuous cycle, accelerating a society’s economic growth, expanding and spreading its wealth, and bringing to its people what Smith had termed “convenience and luxury.”8 This view of technology as an economic elixir seemed, happily, to be borne out by the early history of industrialization, and it became a fixture of economic theory. The idea wasn’t compelling only to early capitalists and their scholarly brethren. Many social reformers applauded mechanization, viewing it as the best hope for raising the urban masses out of poverty and servitude.
Economists, capitalists, and reformers could afford to take the long view. With the workers themselves, that wasn’t the case. Even a temporary abridgment of labor could pose a real and immediate threat to their livelihoods. The installation of new factory machines put plenty of people out of jobs, and it forced others to exchange interesting, skilled work for the tedium of pulling levers and pressing foot-pedals. In many parts of Britain during the eighteenth and the early nineteenth century, skilled workers took to sabotaging the new machinery as a way to defend their jobs, their trades, and their communities. “Machine-breaking,” as the movement came to be called, was not simply an attack on technological progress. It was a concerted attempt by tradesmen to protect their ways of life, which were very much bound up in the crafts they practiced, and to secure their economic and civic autonomy. “If the workmen disliked certain machines,” writes the historian Malcolm Thomis, drawing on contemporary accounts of the uprisings, “it was because of the use to which they were being put, not because they were machines or because they were new.”9
Machine-breaking culminated in the Luddite rebellion that raged through the industrial counties of the English Midlands from 1811 to 1816. Weavers and knitters, fearing the destruction of their small-scale, locally organized cottage industry, formed guerrilla bands with the intent of stopping big textile mills and factories from installing mechanized looms and stocking frames. The Luddites—the rebels took their now-notorious name from a legendary Leicestershire machine-breaker known as Ned Ludlam—launched nighttime raids against the plants, often wrecking the new equipment. Thousands of British troops had to be called in to battle the rebels, and the soldiers put down the revolt with brutal force, killing many and incarcerating others.
Although the Luddites and other machine-breakers had some scattered success in slowing the pace of mechanization, they certainly didn’t stop it. Machines were soon so commonplace in factories, so essential to industrial production and competition, that resisting their use came to be seen as an exercise in futility. Workers acquiesced to the new technological regime, though their distrust of machinery persisted.
IT WAS Marx who, a few decades after the Luddites lost their fight, gave the deep divide in society’s view of mechanization its most powerful and influential expression. Frequently in his writings, Marx invests factory machinery with a demonic, parasitic will, portraying it as “dead labour” that “dominates, and pumps dry, living labour power.” The workman becomes a “mere living appendage” of the “lifeless mechanism.”10 In a darkly prophetic remark during an 1856 speech, he said, “All our invention and progress seem to result in endowing material forces with intellectual life, and stultifying human life into a material force.”11 But Marx didn’t just talk about the “infernal effects” of machines. As the media scholar Nick Dyer-Witheford has explained, he also saw and lauded “their emancipatory promise.”12 Modern machinery, Marx observed in that same speech, has “the wonderful power of shortening and fructifying human labour.”13 By freeing workers from the narrow specializations of their trades, machines might allow them to fulfill their potential as “totally developed” individuals, able to shift between “different modes of activity” and hence “different social functions.”14 In the right hands—those of the workers rather than the capitalists—technology would no longer be the yoke of oppression. It would become the uplifting block and tackle of self-fulfillment.
The idea of machines as emancipators took stronger hold in Western culture as the twentieth century approached. In an 1897 article praising the mechanization of American industry, the French economist Émile Levasseur ticked off the benefits that new technology had brought to “the laboring classes.” It had raised workers’ wages and pushed down the prices they paid for goods, providing them with greater material comfort. It had spurred a redesign of factories, making workplaces cleaner, better lit, and generally more hospitable than the dark satanic mills that characterized the early years of the Industrial Revolution. Most important of all, it had elevated the kind of work that factory hands performed. “Their task has become less onerous, the machine doing everything which requires great strength; the workman, instead of bringing his muscles into play, has become an inspector, using his intelligence.” Levasseur acknowledged that laborers still grumbled about having to operate machinery. “They reproach [the machine] with demanding such continued attention that it enervates,” he wrote, and they accuse it of “degrading man by transforming him into a machine, which knows how to make but one movement, and that always the same.” Yet he dismissed such complaints as blinkered. The workers simply didn’t understand how good they had it.15
Some artists and intellectuals, believing the imaginative work of the mind to be inherently superior to the productive labor of the body, saw a technological utopia in
the making. Oscar Wilde, in an essay published at about the same time as Levasseur’s, though aimed at a very different audience, foresaw a day when machines would not just alleviate toil but eliminate it. “All unintellectual labour, all monotonous, dull labour, all labour that deals with dreadful things, and involves unpleasant conditions, must be done by machinery,” he wrote. “On mechanical slavery, on the slavery of the machine, the future of the world depends.” That machines would assume the role of slaves seemed to Wilde a foregone conclusion: “There is no doubt at all that this is the future of machinery, and just as trees grow while the country gentleman is asleep, so while Humanity will be amusing itself, or enjoying cultivated leisure—which, and not labour, is the aim of man—or making beautiful things, or reading beautiful things, or simply contemplating the world with admiration and delight, machinery will be doing all the necessary and unpleasant work.”16
The Great Depression of the 1930s curbed such enthusiasm. The economic collapse prompted a bitter outcry against what had, in the Roaring Twenties, come to be known and celebrated as the Machine Age. Labor unions and religious groups, crusading editorial writers and despairing citizens—all railed against the job-destroying machines and the greedy businessmen who owned them. “Machinery did not inaugurate the phenomenon of unemployment,” wrote the author of a best-selling book called Men and Machines, “but promoted it from a minor irritation to one of the chief plagues of mankind.” It appeared, he went on, that “from now on, the better able we are to produce, the worse we shall be off.”17 The mayor of Palo Alto, California, wrote a letter to President Herbert Hoover imploring him to take action against the “Frankenstein monster” of industrial technology, a scourge that was “devouring our civilization.”18 At times the government itself inflamed the public’s fears. One report issued by a federal agency called the factory machine “as dangerous as a wild animal.” The uncontrolled acceleration of progress, its author wrote, had left society chronically unprepared to deal with the consequences.19
But the Depression did not entirely extinguish the Wildean dream of a machine paradise. In some ways, it rendered the utopian vision of progress more vivid, more necessary. The more we saw machines as our foes, the more we yearned for them to be our friends. “We are being afflicted,” wrote the great British economist John Maynard Keynes in 1930, “with a new disease of which some readers may not yet have heard the name, but of which they will hear a great deal in the years to come—namely, technological unemployment.” The ability of machines to take over jobs had outpaced the economy’s ability to create valuable new work for people to do. But the problem, Keynes assured his readers, was merely a symptom of “a temporary phase of maladjustment.” Growth and prosperity would return. Per-capita income would rise. And soon, thanks to the ingenuity and efficiency of our mechanical slaves, we wouldn’t have to worry about jobs at all. Keynes thought it entirely possible that in a hundred years, by the year 2030, technological progress would have freed humankind from “the struggle for subsistence” and propelled us to “our destination of economic bliss.” Machines would be doing even more of our work for us, but that would no longer be cause for worry or despair. By then, we would have figured out how to spread material wealth to everyone. Our only problem would be to figure out how to put our endless hours of leisure to good use—to teach ourselves “to enjoy” rather than “to strive.”20
We’re still striving, and it seems a safe bet that economic bliss will not have descended upon the planet by 2030. But if Keynes let his hopes get the best of him in the dark days of 1930, he was fundamentally right about the economy’s prospects. The Depression did prove temporary. Growth returned, jobs came back, incomes shot up, and companies continued buying more and better machines. Economic equilibrium, imperfect and fragile as always, reestablished itself. Adam Smith’s virtuous cycle kept turning.
By 1962, President John F. Kennedy could proclaim, during a speech in West Virginia, “We believe that if men have the talent to invent new machines that put men out of work, they have the talent to put those men back to work.”21 From the opening “we believe,” the sentence is ringingly Kennedyesque. The simple words become resonant as they’re repeated: men, talent, men, work, talent, men, work. The drum-like rhythm marches forward, giving the stirring conclusion—“back to work”—an air of inevitability. To those listening, Kennedy’s words must have sounded like the end of the story. But they weren’t. They were the end of one chapter, and a new chapter had already begun.
WORRIES ABOUT technological unemployment have been on the rise again, particularly in the United States. The recession of the early 1990s, which saw exalted U.S. companies such as General Motors, IBM, and Boeing fire tens of thousands of workers in massive “restructurings,” prompted fears that new technologies, particularly cheap computers and clever software, were about to wipe out middle-class jobs. In 1994, the sociologists Stanley Aronowitz and William DiFazio published The Jobless Future, a book that implicated “labor-displacing technological change” in “the trend toward more low-paid, temporary, benefit-free blue- and white-collar jobs and fewer decent permanent factory and office jobs.”22 The following year, Jeremy Rifkin’s unsettling The End of Work appeared. The rise of computer automation had inaugurated a “Third Industrial Revolution,” declared Rifkin. “In the years ahead, new, more sophisticated software technologies are going to bring civilization ever closer to a near workerless world.” Society had reached a turning point, he wrote. Computers could “result in massive unemployment and a potential global depression,” but they could also “free us for a life of increasing leisure” if we were willing to rewrite the tenets of contemporary capitalism.23 The two books, and others like them, caused a stir, but once again fears about technology-induced joblessness passed quickly. The resurgence of economic growth through the middle and late 1990s, culminating in the giddy dot-com boom, turned people’s attention away from apocalyptic predictions of mass unemployment.
A decade later, in the wake of the Great Recession of 2008, the anxieties returned, stronger than ever. In mid-2009, the American economy, recovering fitfully from the economic collapse, began to expand again. Corporate profits rebounded. Businesses ratcheted their capital investments up to pre-recession levels. The stock market soared. But hiring refused to bounce back. While it’s not unusual for companies to wait until a recovery is well established before recruiting new workers, this time the hiring lag seemed interminable. Job growth remained unusually tepid, the unemployment rate stubbornly high. Seeking an explanation, and a culprit, people looked to the usual suspect: labor-saving technology.
Late in 2011, two respected MIT researchers, Erik Brynjolfsson and Andrew McAfee, published a short electronic book, Race against the Machine, in which they gently chided economists and policy makers for dismissing the possibility that workplace technology was substantially reducing companies’ need for new employees. The “empirical fact” that machines had bolstered employment for centuries “conceals a dirty secret,” they wrote. “There is no economic law that says that everyone, or even most people, automatically benefit from technological progress.” Although Brynjolfsson and McAfee were anything but technophobes—they remained “hugely optimistic” about the ability of computers and robots to boost productivity and improve people’s lives over the long run—they made a strong case that technological unemployment was real, that it had become pervasive, and that it would likely get much worse. Human beings, they warned, were losing the race against the machine.24
Their ebook was like a match thrown onto a dry field. It sparked a vigorous and sometimes caustic debate among economists, a debate that soon drew the attention of journalists. The phrase “technological unemployment,” which had faded from use after the Great Depression, took a new grip on the public mind. At the start of 2013, the TV news program 60 Minutes ran a segment, called “March of the Machines,” that examined how businesses were using new technologies in place of workers at warehouses, hospitals, law firms, and
manufacturing plants. Correspondent Steve Kroft lamented “a massive high-tech industry that’s contributed enormous productivity and wealth to the American economy but surprisingly little in the way of employment.”25 Shortly after the program aired, a team of Associated Press writers published a three-part investigative report on the persistence of high unemployment. Their grim conclusion: jobs are “being obliterated by technology.” Noting that science-fiction writers have long “warned of a future when we would be architects of our own obsolescence, replaced by our machines,” the AP reporters declared that “the future has arrived.”26 They quoted one analyst who predicted that the unemployment rate would reach 75 percent by the century’s end.27
Such forecasts are easy to dismiss. Their alarmist tone echoes the refrain heard time and again since the eighteenth century. Out of every economic downturn rises the specter of a job-munching Frankenstein monster. And then, when the economic cycle emerges from its trough and jobs return, the monster goes back in its cage and the worries subside. This time, though, the economy isn’t behaving as it normally does. Mounting evidence suggests that a troubling new dynamic may be at work. Joining Brynjolfsson and McAfee, several prominent economists have begun questioning their profession’s cherished assumption that technology-fueled productivity gains will bring job and wage growth. They point out that over the last decade U.S. productivity rose at a faster pace than we saw in the preceding thirty years, that corporate profits have hit levels we haven’t seen in half a century, and that business investments in new equipment have been rising sharply. That combination should bring robust employment growth. And yet the total number of jobs in the country has barely budged. Growth and employment are “diverging in advanced countries,” says economist Michael Spence, a Nobel laureate, and technology is the main reason why: “The replacement of routine manual jobs by machines and robots is a powerful, continuing, and perhaps accelerating trend in manufacturing and logistics, while networks of computers are replacing routine white-collar jobs in information processing.”28
The Glass Cage: Automation and Us Page 3