Is a robot coming for your job?
The odds are high, according to recent economic analyses. Indeed, fully 47 percent of all U.S. jobs will be automated “in a decade or two,” as the tech-employment scholars Carl Frey and Michael Osborne have predicted. That’s because artificial intelligence and robotics are becoming so good that nearly any routine task could soon be automated. Robots and AI are already whisking products around Amazon’s huge shipping centers, diagnosing lung cancer more accurately than humans and writing sports stories for newspapers.
They’re even replacing cabdrivers. Last year in Pittsburgh, Uber put its first-ever self-driving cars into its fleet: Order an Uber and the one that rolls up might have no human hands on the wheel at all. Meanwhile, Uber’s “Otto” program is installing AI in 16-wheeler trucks—a trend that could eventually replace most or all 1.7 million drivers, an enormous employment category. Those jobless truckers will be joined by millions more telemarketers, insurance underwriters, tax preparers and library technicians—all jobs that Frey and Osborne predicted have a 99 percent chance of vanishing in a decade or two.
What happens then? If this vision is even halfway correct, it’ll be a vertiginous pace of change, upending work as we know it. As the last election amply illustrated, a big chunk of Americans already hotly blame foreigners and immigrants for taking their jobs. How will Americans react to robots and computers taking even more?
One clue might lie in the early 19th century. That’s when the first generation of workers had the experience of being suddenly thrown out of their jobs by automation. But rather than accept it, they fought back—calling themselves the “Luddites,” and staging an audacious attack against the machines.
At the turn of 1800, the textile industry in the United Kingdom was an economic juggernaut that employed the vast majority of workers in the North. Working from home, weavers produced stockings using frames, while cotton-spinners created yarn. “Croppers” would take large sheets of woven wool fabric and trim the rough surface off, making it smooth to the touch.
These workers had great control over when and how they worked—and plenty of leisure. “The year was chequered with holidays, wakes, and fairs; it was not one dull round of labor,” as the stocking-maker William Gardiner noted gaily at the time. Indeed, some “seldom worked more than three days a week.” Not only was the weekend a holiday, but they took Monday off too, celebrating it as a drunken “St. Monday.”
Croppers in particular were a force to be reckoned with. They were well-off—their pay was three times that of stocking-makers—and their work required them to pass heavy cropping tools across the wool, making them muscular, brawny men who were fiercely independent. In the textile world, the croppers were, as one observer noted at the time, “notoriously the least manageable of any persons employed.”
But in the first decade of the 1800s, the textile economy went into a tailspin. A decade of war with Napoleon had halted trade and driven up the cost of food and everyday goods. Fashions changed, too: Men began wearing “trowsers,” so the demand for stockings plummeted. The merchant class—the overlords who paid hosiers and croppers and weavers for the work—began looking for ways to shrink their costs.
That meant reducing wages—and bringing in more technology to improve efficiency. A new form of shearer and “gig mill” let one person crop wool much more quickly. An innovative, “wide” stocking frame allowed weavers to produce stockings six times faster than before: Instead of weaving the entire stocking around, they’d produce a big sheet of hosiery and cut it up into several stockings. “Cut-ups” were shoddy and fell apart quickly, and could be made by untrained workers who hadn’t done apprenticeships, but the merchants didn’t care. They also began to build huge factories where coal-burning engines would propel dozens of automated cotton-weaving machines.
“They were obsessed with keeping their factories going, so they were introducing machines wherever they might help,” says Jenny Uglow, a historian and author of In These Times: Living in Britain Through Napoleon’s Wars, 1793-1815.
The workers were livid. Factory work was miserable, with brutal 14-hour days that left workers—as one doctor noted—“stunted, enfeebled, and depraved.” Stocking-weavers were particularly incensed at the move toward cut-ups. It produced stockings of such low quality that they were “pregnant with the seeds of its own destruction,” as one hosier put it: Pretty soon people wouldn’t buy any stockings if they were this shoddy. Poverty rose as wages plummeted.
The workers tried bargaining. They weren’t opposed to machinery, they said, if the profits from increased productivity were shared. The croppers suggested taxing cloth to make a fund for those unemployed by machines. Others argued that industrialists should introduce machinery more gradually, to allow workers more time to adapt to new trades.
The plight of the unemployed workers even attracted the attention of Charlotte Brontë, who wrote them into her novel Shirley. “The throes of a sort of moral earthquake,” she noted, “were felt heaving under the hills of the northern counties.”
In mid-November 1811, that earthquake began to rumble. That evening, according to a report at the time, half a dozen men—with faces blackened to obscure their identities, and carrying “swords, firelocks, and other offensive weapons”—marched into the house of master-weaver Edward Hollingsworth, in the village of Bulwell. They destroyed six of his frames for making cut-ups. A week later, more men came back and this time they burned Hollingsworth’s house to the ground. Within weeks, attacks spread to other towns. When panicked industrialists tried moving their frames to a new location to hide them, the attackers would find the carts and destroy them en route.
A modus operandi emerged: The machine-breakers would usually disguise their identities and attack the machines with massive metal sledgehammers. The hammers were made by Enoch Taylor, a local blacksmith; since Taylor himself was also famous for making the cropping and weaving machines, the breakers noted the poetic irony with a chant: “Enoch made them, Enoch shall break them!”
Most notably, the attackers gave themselves a name: the Luddites.
Before an attack, they’d send a letter to manufacturers, warning them to stop using their “obnoxious frames” or face destruction. The letters were signed by “General Ludd,” “King Ludd” or perhaps by someone writing “from Ludd Hall”—an acerbic joke, pretending the Luddites had an actual organization.
Despite their violence, “they had a sense of humor” about their own image, notes Steven Jones, author of Against Technology and a professor of English and digital humanities at the University of South Florida. An actual person Ludd did not exist; probably the name was inspired by the mythic tale of “Ned Ludd,” an apprentice who was beaten by his master and retaliated by destroying his frame.
Ludd was, in essence, a useful meme—one the Luddites carefully cultivated, like modern activists posting images to Twitter and Tumblr. They wrote songs about Ludd, styling him as a Robin Hood-like figure: “No General But Ludd / Means the Poor Any Good,” as one rhyme went. In one attack, two men dressed as women, calling themselves “General Ludd’s wives.” “They were engaged in a kind of semiotics,” Jones notes. “They took a lot of time with the costumes, with the songs.”
And “Ludd” itself! “It’s a catchy name,” says Kevin Binfield, author of Writings of the Luddites. “The phonic register, the phonic impact.”
As a form of economic protest, machine-breaking wasn’t new. There were probably 35 examples of it in the previous 100 years, as the author Kirkpatrick Sale found in his seminal history Rebels Against the Future. But the Luddites, well-organized and tactical, brought a ruthless efficiency to the technique: Barely a few days went by without another attack, and they were soon breaking at least 175 machines per month. Within months they had destroyed probably 800, worth £25,000—the equivalent of $1.97 million, today.
“It seemed to many people in the South like the whole of the North was sort of going up in flames,” Uglow notes. “In terms of industrial history, it was a small industrial civil war.”
Factory owners began to fight back. In April 1812, 120 Luddites descended upon Rawfolds Mill just after midnight, smashing down the doors “with a fearful crash” that was “like the felling of great trees.” But the mill owner was prepared: His men threw huge stones off the roof, and shot and killed four Luddites. The government tried to infiltrate Luddite groups to figure out the identities of these mysterious men, but to little avail. Much as in today’s fractured political climate, the poor despised the elites—and favored the Luddites. “Almost every creature of the lower order both in town & country are on their side,” as one local official noted morosely.
1812 handbill
An 1812 handbill sought information about the armed men who destroyed five machines. (The National Archives, UK)
At heart, the fight was not really about technology. The Luddites were happy to use machinery—indeed, weavers had used smaller frames for decades. What galled them was the new logic of industrial capitalism, where the productivity gains from new technology enriched only the machines’ owners and weren’t shared with the workers.
The Luddites were often careful to spare employers who they felt dealt fairly. During one attack, Luddites broke into a house and destroyed four frames—but left two intact after determining that their owner hadn’t lowered wages for his weavers. (Some masters began posting signs on their machines, hoping to avoid destruction: “This Frame Is Making Full Fashioned Work, at the Full Price.”)
For the Luddites, “there was the concept of a ‘fair profit,’” says Adrian Randall, the author of Before the Luddites. In the past, the master would take a fair profit, but now he adds, “the industrial capitalist is someone who is seeking more and more of their share of the profit that they’re making.” Workers thought wages should be protected with minimum-wage laws. Industrialists didn’t: They’d been reading up on laissez-faire economic theory in Adam Smith’s The Wealth of Nations, published a few decades earlier.
“The writings of Dr. Adam Smith have altered the opinion, of the polished part of society,” as the author of a minimum wage proposal at the time noted. Now, the wealthy believed that attempting to regulate wages “would be as absurd as an attempt to regulate the winds.”
Six months after it began, though, Luddism became increasingly violent. In broad daylight, Luddites assassinated William Horsfall, a factory owner, and attempted to assassinate another. They also began to raid the houses of everyday citizens, taking every weapon they could find.
Parliament was now fully awakened, and began a ferocious crackdown. In March 1812, politicians passed a law that handed out the death penalty for anyone “destroying or injuring any Stocking or Lace Frames, or other Machines or Engines used in the Framework knitted Manufactory.” Meanwhile, London flooded the Luddite counties with 14,000 soldiers.
By winter of 1812, the government was winning. Informants and sleuthing finally tracked down the identities of a few dozen Luddites. Over a span of 15 months, 24 Luddites were hanged publicly, often after hasty trials, including a 16-year-old who cried out to his mother on the gallows, “thinking that she had the power to save him.” Another two dozen were sent to prison and 51 were sentenced to be shipped off to Australia.
“They were show trials,” says Katrina Navickas, a history professor at the University of Hertfordshire. “They were put on to show that [the government] took it seriously.” The hangings had the intended effect: Luddite activity more or less died out immediately.
It was a defeat not just of the Luddite movement, but in a grander sense, of the idea of “fair profit”—that the productivity gains from machinery should be shared widely. “By the 1830s, people had largely accepted that the free-market economy was here to stay,” Navickas notes.
A few years later, the once-mighty croppers were broken. Their trade destroyed, most eked out a living by carrying water, scavenging, or selling bits of lace or cakes on the streets.
“This was a sad end,” one observer noted, “to an honourable craft.”
These days, Adrian Randall thinks technology is making cab-driving worse. Cabdrivers in London used to train for years to amass “the Knowledge,” a mental map of the city’s twisty streets. Now GPS has made it so that anyone can drive an Uber—so the job has become deskilled. Worse, he argues, the GPS doesn’t plot out the fiendishly clever routes that drivers used to. “It doesn’t know what the shortcuts are,” he complains. We are living, he says, through a shift in labor that’s precisely like that of the Luddites.
Economists are divided as to how profound the disemployment will be. In his recent book Average Is Over, Tyler Cowen, an economist at George Mason University, argued that automation could produce profound inequality. A majority of people will find their jobs taken by robots and will be forced into low-paying service work; only a minority—those highly skilled, creative and lucky—will have lucrative jobs, which will be wildly better paid than the rest. Adaptation is possible, though, Cowen says, if society creates cheaper ways of living—“denser cities, more trailer parks.”
Erik Brynjolfsson is less pessimistic. An MIT economist who co-authored The Second Machine Age, he thinks automation won’t necessarily be so bad. The Luddites thought machines destroyed jobs, but they were only half right: They can also, eventually, create new ones. “A lot of skilled artisans did lose their jobs,” Brynjolfsson says, but several decades later demand for labor rose as new job categories emerged, like office work. “Average wages have been increasing for the past 200 years,” he notes. “The machines were creating wealth!”
The problem is that transition is rocky. In the short run, automation can destroy jobs more rapidly than it creates them—sure, things might be fine in a few decades, but that’s cold comfort to someone in, say, their 30s. Brynjolfsson thinks politicians should be adopting policies that ease the transition—much as in the past, when public education and progressive taxation and antitrust law helped prevent the 1 percent from hogging all the profits. “There’s a long list of ways we’ve tinkered with the economy to try and ensure shared prosperity,” he notes.
Will there be another Luddite uprising? Few of the historians thought that was likely. Still, they thought one could spy glimpses of Luddite-style analysis—questioning of whether the economy is fair—in the Occupy Wall Street protests, or even in the environmental movement. Others point to online activism, where hackers protest a company by hitting it with “denial of service” attacks by flooding it with so much traffic that it gets knocked off­line.
Perhaps one day, when Uber starts rolling out its robot fleet in earnest, angry out-of-work cabdrivers will go online—and try to jam up Uber’s services in the digital world.
“As work becomes more automated, I think that’s the obvious direction,” as Uglow notes. “In the West, there’s no point in trying to shut down a factory.”


Why did you decide to write about this now?
I’ve been thinking for a year about the Luddites and why they have this dismissed reputation. The word Luddite has come to mean someone who doesn’t like technology. Really it was a political fight over who was going to use the spoils of profits from machinery.
As I started seeing books about automation and the coming waves of jobs that were going to vanish, I started thinking this could have real resonance in helping us understand what’s going on, maybe there’s something to be learned about how the Luddites reacted in the past.
How does the Luddites’ struggle translate to that of workers’ today?
With the Luddites, you had a class of workers who had for long time had an agreement with the people who bought their work, the merchants buying all of the weaving and cropping (wool textiles).
They had an understanding with merchants stretching over decades to centuries that there should be a sense of fair profit. The merchants were buying their stuff, reselling it, and providing some capital, sometimes to buy machines. But the profit should be fairly shared all the way around.
What happens is you get Adam Smith publishing his seminal work on free market capitalism [Wealth of Nations] in late 18th century. At the beginning of the 19th century, the merchants are starting to go, “Wait a minute. There is no such thing as a fair profit. There’s just whatever I can get from the marketplace. There is no moral imperative for us to give up a larger chunk of our profits to these people.”
A scene of English village textile production in the 1700s. (William Hincks, London 1783. British Library)
Adam Smith is arguing if we all behave with high degree of self interest, that will actually improve the economy in the long run. This was the first beginnings of the real embrace of free-market capitalism.
The Luddites were not opposed to the idea of using machines to make things more efficiently or be more productive. They just thought if you’re going to make more money because you’re more productive, you need to kick some of that money back down to the workers. The merchants were really not of that opinion….
[The Luddites] tried to bargain with the factory owners [arguing for minimum prices, a textile tax to support workers pensions, or phased introduction of new machines], but that didn’t go over at all. When the Luddites got to their wits end, they basically started going in and smashing and breaking machines, saying, this is all we got left. We’re going to destroy the means by which you produce this dislocation in our lives.
The Luddite uprising began in the fall of 1811. Pretty soon, they were breaking a couple hundred machines per month. After five to six months the government realized this was not slowing down. This was a real thing and the government fought back ferociously.
Of course, very wealthy factory owners had a lot of sway with the [British] Parliament, who sent in 14,000 soldiers to flood the northern counties where the Luddites were doing this smashing. They passed a new law specifically targeting frame-breaking giving it the death sentence. They worked really, really hard to infiltrate rings with spies.
Screenshot 2017-04-25 17.46.22
1839 newspaper clipping from the Manchester Observer. (Working Class Movement Library)
It took some time. The Luddite uprising lasted for about a year, but the government did eventually break the back of it by putting several dozen Luddites to death; very public trials, very rapidly done. Special gallows would hang several of them at once. They shipped another couple dozen off to Australia. They even hung a 16-year-old boy who had done nothing more than been a lookout. That really put an end to it.
What are the parallels to the Luddites’ situation and that of workers’ today?
One of the things that’s similar is the rapid pace of change. Artificial intelligence has taken this very, very significant leap forward in the last 15 years and it’s probably going take even more significant leaps in the next 10 years.
What’s interesting about the Luddites, is that there was a very similar sudden jump. In the 17th and 18th centuries, the period before the Luddites, it was a fairly stable period for the textile industry in the United Kingdom. What workers were paid was published in broadsheets and newspapers. If you set out to make a living as a weaver, you knew very much what you were going to make. You had a fair amount of control to work from home. You controlled your own schedule. Some led fairly leisurely lives; working only four days a week was enough to make a pretty good living.
And then in 10 or 20 years, all of that was rapidly inverted. There was an economic recession caused by a war with France. Britain suddenly had all sorts of trade barriers. You had a fashion change. Men stopped wearing leggings and started wearing these new-fangled things called trousers, so suddenly there was less demand.
Machine breaking in textile mills in the 1800s. (source unknown, 1812)
History of the cotton manufacture in Great Britain by Sir Edward Baines, History of the cotton manufacture in Great Britain by Sir Edward Baines,
Industrial looms for cotton weaving in Great Britain. (Sir Edward Baines, 1835)
The merchants needed to cut costs, so they decided to take advantage of technologies that had come along. One was more efficient weaving frames. One person could be four to six times more productive. Secondly, they started making factories using stream power to power the looms instead of humans.
Now, the humans’ job is to tend their machines. You need fewer humans and it’s quite dangerous work. The factories are terrible, dreadful places. All sorts of accidents happen because the factories had no safety standards.
You had this sudden shift from workers being paid reasonably well and having a lot of autonomy over what they did to getting paid pretty terribly and needing far less people producing crappier goods.
It sounds like we’re fighting over the same question: who deserves the proceeds from the means of production?
Absolutely. This was of course the great intellectual debate between [Karl] Marx and Smith and those that followed them.
That’s what we see thrown into sharp relief by technologies that can do thinking work [today]. If you go to Silicon Valley, they can very quickly make a technology that can either throw a lot of people out of work or create a whole new category of production, create a lot of wealth and concentrate it in the hands of a few people that run the company, instead of thousands of people.
What’s the biggest difference in terms of the effect of automation on today’s working class?
I think there is one big difference: a lot of the jobs that are going to be outsourced are actually not working class at all. They are healthily middle-to-upper class, and often white collar. Automation has really moved up the income ladder. It’s not just taking away jobs with your hands. It’s taking away jobs from your mind. The difference with the Luddites is that automation is moving up the job chain.
How do you think today’s workers will fare given the [economic and political] system in the US?
Cover of an article on Machine-Breaking from the Westminster Review, Jan 1831. (Working Class Movement Library)
One big dissimilarity is there really is less collective action and solidarity among workers. During the Luddite period, you had workers who knew each other in tight-knit little towns where it was easier to organize and get frame-breaking to happen. It’s a lot harder with a larger country with disparate people all over the place. There has been a very, very active campaign by Republicans over the last 50 years, a successful campaign, to beat back unions. The one force that would have done something Luddite-like just doesn’t have that much power anymore.
There are no factories to monkey wrench anymore. What are you going to do: burn down Facebook or Uber? Their products are software. Any similar software activity would take the form of not smashing a machine but smashing a piece of software. Hacking. It would look more like what Anonymous does.
On the other hand, you have the Internet, modes of communication for disparate people to talk about and spread their ideas in ways that can be powerful. Look at Occupy Wall Street. Many critics said they didn’t have any solid plan. Fair enough, except they had the Internet and the ability to spread their message wide. They did an amazing job with it. They put this discussion of the 1% on the map. That conversation was not happening in the mainstream before they came along. They did that.
That is almost like a Luddite-type of activity. It’s a mass-gathering that tries to draw focus to something. It’s the same thing with Black Lives Matter and the Tea Party in their own way.
Donald Trump’s speech [on Inauguration Day] was filled with promises how he’s going to bring jobs back, but he’s never given even the slightest details about he’s going to do that. … My suspicion is that his supporters are going to be deeply disappointed. So the real conversation for anyone seriously trying to grapple with this is how to share the profits. What industries are growing? What type of manufacturing jobs might be growing?
If you want create jobs you certainly can, but that’s a political conversation the incoming government hasn’t grappled with. Are there ways forward and answers to what’s happening? Yeah, sure, absolutely. But I don’t think anything the Trump administration has proposed is up to that challenge.