gemini://beyondneolithic.life:1965/mueller_archive/breaking_things_at_work/against

What is automation? As we have seen, machines have replicated and augmented human work processes for centuries, and that is often the colloquial use of "automation" in our current moment. But "automation" was not used to describe this process until 1947, when Delmar Harder, vice president of manufacturing at Ford Motor Company, created its Automation Department. The department's engineers redesigned automobile production so that materials were automatically conveyed from one process to another, obviating the need for laborers to load and unload machines.Further, the process was itself increasingly machine-controlled, through a system of timers, switches, and relays — what technology historian David Hounsell calls the "electromechanical brain."

Most of the technologies involved in automation had been developed and implemented in other industries years before their incorporation into Ford's production process. What made automation new was its centrality to Ford's manufacturing strategy, coming at a time of historic unrest among autoworkers, and in particular, on the heels of a costly twenty-four-day strike at Ford's massive River Rouge plant in May of 1949. Not only would the new technologies dramatically reduce an unruly labor force, but they allowed Ford to decentralize its production away from the roiling unrest of Detroit as the company opened new automated factories in Cleveland and Buffalo. Workers immediately perceived the threat, and automation was, from its inception, a deeply politicized issue.

Today, the headlines scream about automation's potential to replace workers, often in language reminiscent of anti-immigrant rhetoric: robots are threatening to "take" or "steal" jobs. You can even go to the website willrobotstakemyjob.com and input specific occupations to get statistics on the likelihood of such theft. Writers have only a 3.8 percent chance — "totally safe" — while machinists face an alarming percent. "Robots are watching," the site cautions. These numbers are drawn from a widely cited 2013 report by economist Carl Benedikt Frey and computer scientist Michael A. Osborne that concluded 47 percent of total US employment would be automated by 2034.

Many writers on the radical left have accepted this framing of automation, and even extended and detourned its implications, making "full automation" central to the transcendence of capitalist exploitation. In "Inventing the Future," Alex Williams and Nick Srnicek argue, "Without full automation, postcapitalist futures must necessarily choose between abundance at the expense of freedom (echoing the work-centricity of Soviet Russia) or freedom at the expense of abundance, represented by primitivist dystopias." Peter Frase's "Four Futures," which plays out an assortment of postcapitalisms, utopian and dystopian, holds "perfect automation" as "the constant in [the] equation." And Aaron Bastani's "Fully Automated Luxury Communism" pushes this idea to its limits, promising a future of boundless leisure for all, supplemented by a profusion of goods and services delivered sans human exploitation: "We will see more of the world than ever before, eat varieties of food we never have heard of, and lead lives equivalent — if we so wish — to those of today's billionaires."

Such a framing is both simple and attractive, especially to those of us trapped in dead-end jobs and eking out precarious existences; if robots, rather than if we and our fellow workers, performed these tasks, and the productivity of technology were widely dispersed, maybe we could live our lives like the rich do. Like those cheesy banner ads that were all over the web in the late 2000s, you could have a fulfilling egalitarian society with "one weird trick." The bourgeoisie would hate this!

The problem for Full Automators, of any political leaning, is that their predictions rely on a faulty understanding of what actually happens when machines are introduced into production processes. In other words, "perfect automation" has little to do with actually existing automation. David Autor, an economist, offers a useful corrective to this mistake in his 2015 article "Why Are There Still So Many Jobs?," its plaintive title a response to John Maynard Keynes's rosy predictions of a future with a reduced work week. As Autor explains, rather than simply replace human jobs with machinic processes, automation affects labor in complex ways:

> Changes in technology do alter the types of jobs available and what those jobs pay. In the last few decades, one noticeable change has been "polarization" of the labor market, in which wage gains went disproportionately to those at the top and at the bottom of the income and skill distribution, not to those in the middle.

Why does automation polarize instead of outright replace jobs? For one, many jobs require labor that is challenging to automate. Computers have to follow instructions laid out by programmers, so in order to substitute a computer for a worker, the worker's tasks must be understood and articulated. However, much of the labor process is encompassed in tacit knowledge that workers are unable to articulate: "There are tasks for which neither computer programmers nor anyone else can enunciate the explicit 'rules' or procedures." Even when tasks are known, automating them is easier said than done. On one end, computers cannot replicate the high levels of abstract thought required for managerial positions. On the other, jobs that require both manual work and flexibility, such as service sector jobs in food preparation and maintenance, are both difficult and cost-prohibitive to automate.

Take an example. In March 2018, Flippy, a burger-flipping robot, was rolled out at the Pasadena location of fast-food chain CaliBurger, to great fanfare and numerous headlines. The implication was clear: Would this spell the end of fast-food jobs, the mascot for low-skilled entry-level occupations? Not exactly. In an event that provoked far less press coverage, Flippy was retired after one day of work. CaliBurger's owners took the honorable path of blaming Flippy's failure on their human employees: workers were simply too slow with tasks such as dressing the burgers, causing Flippy's achievements to pile up. However, a few discerning journalists had previously noted Flippy's numerous errors in the relatively simple task that gave the robot its name. And so, yet another fully automated dream came crashing into messy reality. 10

According to Autor, the introduction of new kinds of information and control technology, such as what is currently hyped as "artificial intelligence," supplements managerial work, and so increases the power and wages of bosses. On the other end, manual laborers (such as Flippy's coworkers) see tasks eaten away and their movements reorganized and tightly controlled to make room for more rigid machines. Wages and working conditions deteriorate. But even then, automation stops short of "full": such systems, as we will see, rely upon a stratum of human labor that is all but ineradicable. This is as true of Flippy as it is of the most powerful AI.

What do tend to be substitutable are not the lowest rungs, but those jobs requiring repetitive physical labor, as well as middle management jobs in operations. For example, Amazon's warehouses use a software-directed system that coordinates human laborers, who select individual goods, with robots, who move large shelves. Algorithms replace middle-income jobs in managing the floor, leading to a polarized workforce of increasingly wealthy and powerful executives and increasingly degraded laborers who are substitutable not by machines, but by other humans; in other words, they are eminently replaceable.

Another way to put this is to use the language of Italian *operaismo* (workerism), which, half a century ago, tracked these movements very carefully as new technologies descended upon the vast and turbulent workforces of the Turin auto factories. Part of the phenomenon Autor describes as "polarization" was articulated in the workerist language of class struggle as "decomposition of the working class." Reorganizing the labor process was a powerful way to disrupt how workers had organized themselves against the bosses. And wherever automation was implemented, it met pitched resistance backed by worker-led militant research, independent of official labor leadership.

In May of 1956, the British Parliament took up a discussion of an industrial dispute at the Standard Motor Company in Coventry. Workers had been on strike for over a week, angered by the company's announcement that 3,000 employees, rendered redundant by the adoption of new automation technologies, would be fired. Some members of Parliament put pointed questions to the Tory minister of labour, Iain Macleod: was he "aware that the introduction of automation in industry is giving rise to serious misgiving among the organised workers"? Macleod attempted to assuage these concerns with little more than platitudes, appealing to automation's contribution to "the prosperity and happiness of the nation" and blithely ignoring its imminent impact on jobs. "It is welcomed by the Government," he replied, "and responsible opinion on both sides of industry as essential to our future efficiency and, therefore, to the continuance of full employment." In response, Labour MP William Owen struck a conciliatory note:

> Is the Minister aware that organised workers are by no means limited to the philosophy of the Luddite movement these days but welcome the development of a new technique in modern industry? However, they are seriously concerned with the probable economic and social effect of the new machines unless there is — as the Minister has indicated — a real possibility of early consultation between both sides of industry, with Government co-operation.

Owen portrayed the Coventry strikers as potentially willing to adopt new technology, provided their jobs were preserved, with the help of government policy. Such measures would not be forthcoming. Less than two months after an uneasy resolution of the Standard Motor Company strike, the British Motor Company announced immediate layoffs of 6,000 workers. Perhaps those workers subscribing to the philosophy of the Luddite movement were right after all.

Across the English Channel in France, Cornelius Castoriadis, a theorist and cofounder of the libertarian socialist group Socialisme ou Barbarie, observed these events closely. Two elements of these struggles stuck out to him. First, automation represented a new phase of class struggle, "a capital offensive against labor, considered as the originating force in production," with the eventual goal of "the elimination of man qua man from the sphere of production" — a goal Castoriadis believed was ultimately impossible, even while it exercised a decisive influence over the course of class struggle. Second, he noted that the strikes emerged from the workers and their chosen representatives, the shop stewards, in contravention of the trade union leadership.

This latter factor — autonomous worker action against both management and union leadership — would mark the struggles against automation in the postwar era. While unions counseled caution and patience, workers paid them little heed, walking off jobsites and vandalizing machines. And on the vanguard of opposing automation were those often marginalized by the official workers' movement — women and African Americans — who produced some of the most enduring critical knowledge of new technologies.

The fight against machines that Castoriadis noticed in Coventry had begun more than a decade earlier, as part of the dramatic restructuring of the industrial economy during World War II. Though the term did not achieve widespread use until after the war's cessation, automation's major victory began with the development of numerically controlled machine tools, which could replace the skilled machinists who had been required for heavy manufacturing. While automation's defenders, such as Macleod, pointed to the gains in "efficiency" brought by the new methods, David Noble's historical research on the process tells a different story. The introduction of automation occurred through the prerogatives of the war economy, what Dwight D. Eisenhower would later dub the "military-industrial complex." This meant that military values, not commercial ones, impacted the new forms of production:

> First was the emphasis placed on performance rather than cost in order to meet the requirements of the military mission … Then there was the insistence upon command, the precise specification, communication, and execution of orders, uncompromised by either intermediary error or judgment. Finally, there was the preoccupation with modern methods, high technology and capital-intensive, to guarantee performance and command objectives and thereby assure the success of the mission: national security against communism.

In other words, the priority for production during the war was consistency and control, not saving time or increasing profits, though wartime demand and wage controls kept corporate coffers full. An alternative form of automation popular among machinists, "record-playback," was never seriously pursued, although it was also efficient. Unlike numeric control, the record-playback method was analog, storing the precise movement of a machinist, and so still required a skilled hand. Rather than pursue efficiency, management sought to wrest control of production away from the machinists.

Military planners and industrialists urgently felt the need for total control during the war. The US workforce reached a pinnacle of unruliness in the 1940s, with the number of strikes per year surpassing the previous high-water mark reached during the Great Depression. On average, Ford plants experienced strike activity every other day. And these strikes were a form of double defiance: of wartime laws forbidding strikes, on one hand, and of pledges of unions to keep a lid on unrest, on the other. Moreover, a clear pattern emerged in which worker rebellion against automation was marked by wildcat strikes.

The need for absolute control over machines permeated the sciences at the time, which themselves were coordinated by the military through the Office of Scientific Research and Development. One emerging branch of research promised to solve this puzzle across a variety of scientific and engineering disciplines by uncovering mechanisms that heralded a future of self-regulated machines: cybernetics. Derived from the Greek word for "steering," cybernetics sought to develop machines that could incorporate "feedback" reflexively into their operation. In other words, they eliminated the need for human control. Norbert Wiener, the mathematician who coined the term "cybernetics" and provided many of its early breakthroughs, began this work with research on the creation of Allied anti-aircraft weapons that were more effective at targeting a zigzagging enemy pilot. In this, as Peter Galison notes, cybernetics as a project centers on a specific problem: how to predict the behavior of a calculating, but opaque, opponent, or a "Manichean devil." On the battlefield, this could mean a tank commander or a fighter pilot. In the context of the factory, these devils could be the workers themselves.

Wiener, whose political sympathies were to the left of many of his military-minded colleagues, realized the terrifying implications of his work. He ultimately abandoned military research after the nuclear attacks on Japan, turning toward social criticism of cybernetics through books and essays written for popular audiences. Beyond the specter of nuclear annihilation, Wiener was concerned with domestic uses of cybernetics: industrial automation. Wiener thought this would be a catastrophe for workers, writing that automation "gives the human race a new and most effective collection of mechanical slaves to perform its labor … any labor that accepts the conditions of competition with slave labor, accepts the conditions of slave labor, and is essentially slave labor."

Wiener did not believe that cybernetics would create a condition where an autonomous technology confronted increasingly superfluous human beings. Rather than a machines-versus-mankind, Terminator -style dystopia so popular in science fiction fantasies of artificial intelligence, he viewed the automatic machine as a potential weapon to be used by powerful people to control others: "Its real danger, however, is the quite different one that such machines, though helpless by themselves, may be used by a human being or a block of human beings to increase their control over the rest of the human race." In other words, automation would be a weapon of class war.

Armed with this insight, Wiener wrote to Walter Reuther, head of the United Automobile Workers, detailing the plans of the industrialists who had attempted to hire Wiener to consult on automating their factories. Although Wiener had refused, he knew other researchers would happily take such positions. He wanted to give Reuther the opportunity to get in front of the "disastrous" levels of unemployment that would result. In his letter, Wiener proposed two alternatives for Reuther to explore. First, he could agitate to "secure the profits in [new machines] to an organization dedicated to the benefit of labor": an automation fund. But Wiener also held out the possibility that the technology itself was too dangerous.

Instead of heeding Wiener's warnings, labor leaders would come to embrace the new machines, cajoled by an aggressive campaign launched by the business elite. Capitalists had been spooked by wartime militancy, which threatened to intensify as millions of soldiers returned to the labor market from overseas. They had flown into action, pushing through coercive initiatives like the 1947 Taft-Hartley Act, but had also held out a few carrots. Firms negotiated with major unions, such as the UAW and the United Mine Workers of America, to tie wage increases to increases in productivity. This meant that labor would accept whatever new machinery businesses introduced, thereby ceding control of the production process to management. So compliant was the union leadership that in 1950, "Fortune" called UMWA head John L. Lewis "the best salesman the machinery industry ever had." Even lightly held reservations about new technology provoked opprobrium among progress- (and business-) minded commentators. As labor economist Ben Seligman complained, "Whenever an exasperated labor leader asserts that automation can be a curse, the head of the U.S. Chamber of Commerce responds that he is a Luddite."

Yet this increased productivity derived, to a large extent, from speeding up the production process; "efficiencies" were borne by the workers' bodies and by the miseries of those thrown into unemployment. Discontent was on the rise. As resistance to automation erupted from below, it would be documented and theorized by Marxist groups, such as Socialisme ou Barbarie, who were estranged from major political parties and trade unions. These groups viewed their task as the analysis of actually existing class struggles at the point of production. Real socialism could follow only from this struggle, not from the endlessly deferred utopias of leisure and abundance imposed through negotiations between bureaucratic unions and capitalists.

In the United States, the Johnson-Forest Tendency, a group of worker-militants who splintered from the American Trotskyist movement, took a keen interest in the postwar transformations at the point of production and the clashes that ensued. With many members embedded in factories, the JFT understood that increased productivity was squeezed out of the small moments of downtime, bits of reprieve from work, to which workers had become accustomed. They produced a pamphlet, "The American Worker," that described this transition in the auto plants: "The worker used to be able to smoke more often. Now he has to spend all day watching, changing and cleaning tools. The interludes are briefer. The end of the day produces a more exhausted worker, mentally and physically."

As conditions deteriorated and the acquiescence of the union bureaucracy continued, workers wildcatted. And one set of wildcat strikes in particular captured the attention of the JFT's "Forest," the pseudonym of Raya Dunayevskaya, a writer and activist who was formerly Trotsky's personal secretary. For months in 1949 and 1950, miners walked off the job and shut down mines with roving pickets in response to the introduction of the continuous miner — or, in the words of the miners, the "man killer." It was the first strike against automation, and in it, Dunayevskaya saw the seed of a new kind of radical politics.

The JFT's "Johnson," the nom de guerre of the great Trinidadian activist and writer C.L.R. James, did not share Dunayevskaya's perspective. In "Facing Reality," cowritten with Grace Lee Boggs and Cornelius Castoriadis after a split with Dunayevskaya, James took a line that was uncharacteristically technologically determinist. The expulsion of workers from the production process indicated "a system committing suicide." For James, automation meant a potentially greater level of organization and control of production by workers:

For Dunayevskaya, automation's reception divided along class lines, determined by one's relationship to the machine. While capitalists, management, and the union leadership praised automation as a progressive force, those who experienced it directly had a completely different view. "If you are the one who operates it," she wrote, "you feel its impact in every bone of your body: you are more sweaty, more tired, more tense and you feel about as useful as a fifth wheel."

Dunayevskaya mercilessly criticized the "labor bureaucracy" as "brainwashed," as they took management's side against their own workers, while their power base was cut out from under their feet. "John L. Lewis disregarded their general strike and announced instead that the union was for 'progress.' The working force in the mines was literally cut in half." When automation took hold in automobile manufacturing, "Reuther told the auto workers to consider 'the future' which would bring them a six-hour day … Meanwhile, there has been no change in the working day since the workers, through their own struggles over decades, won the eight-hour day." This was a classic idealist mistake: "painting the future as it should be instead of speaking of what is." While Reuther promised better living conditions and increased leisure in an automated future, workers testified to the opposite. As one autoworker complained, "All Automation has meant to us is unemployment and overwork. *Both at the same time*." As Dunayevskaya unsparingly put it, "The workers don't go in for abstract argumentation on leisure and plenty at some future, unspecified, time."

Around the time Dunayevskaya was analyzing worker resistance to automation, her collaboration with James was drawing to a close. Seeking a new interlocutor well versed in Hegel and Marx, she struck up a correspondence with Herbert Marcuse, the Frankfurt School philosopher who had since become ensconced in American academia. When Marcuse began his research for his classic analysis of advanced capitalism "One-Dimensional Man," he requested materials on automation from Dunayevskaya, who eagerly supplied him with a bibliography of contemporary texts, as well as a copy of the workers' newsletter she edited, "News and Letters," with an extensive inquiry by Charles Denby on the impact of automation from the perspective of workers.

Denby, a Detroit autoworker and editor at "Notes and Letters," accumulated accounts from workers discussing new automation technologies, including those working on assembly lines and in mines, and even the white-collar workers struggling with new computer technologies. He was as unsparing in his analysis as Dunayevskaya had been in her writing on the miners' strike, detailing the brutal pace, strenuous physical and mental demands, and the strife of technological unemployment brought by the new machines.

> What alienates a production worker is that he is driven to do work that is separated from his thinking … Before Automation, when a major change was made and a new machine was introduced, they had to rely on the workers' knowledge and experience to get it working properly … For a few weeks we felt like human beings working out the problems together and getting things organized and moving smoothly.

> Some years ago, when workers had something to say about how fast they would work and the amount of help they felt they needed if the company wanted more production, the relations among production workers were humanly close. They could help each other with their work. They worked in a way which would make it easy for every one in a group. Today Automation does not allow anyone to help another worker.

> Why do people assume that Automation is the way people will want to work in a new society? Why do they assume that all that matters is that the workers will be in control? Will "being in control" of the machine lighten the work, or make it less boring? Won't work be something completely different? If work will be something different — tied up with life itself — it cannot be the same as Automation that uses men as part of its operations.

Rather than rely on worker testimony, Marcuse depended instead on technical experts and credentialed philosophers. While Marcuse noted some of the problems with automation, and cited Denby, "One-Dimensional Man" provided an optimistic take on new production technology. Like James, he saw salutary effects in how automation organized workers. "The same technological organization which makes for a mechanical community at work also generates a larger interdependence which integrates the worker with the plant," he wrote, imagining a potential for a more social and participatory relationship toward work in the factories. And in a personal translation of a quotation from Marx's "Grundrisse", Marcuse provides one of the earliest appearances in English of the "Fragment on Machines": "One-Dimensional Man" was published in 1964, almost a decade before the English translation of the "Grundrisse." Marcuse's reading of the fragment as presaging a moment of full automation gives his interpretation a decidedly "luxury communism" flavor: "Complete automation in the realm of necessity would open the dimension of free time as the one in which man's private and societal existence would constitute itself. This would be the historical transcendence toward a new civilization."

In her review of "One-Dimensional Man," Dunayevskaya criticized Marcuse for misreading Denby's work, and for folding it into his argument that workers had been incorporated into a totally administered society. Marcuse, she wrote, "leaves out entirely the central point of the pamphlet, the division between the rank and file and the labor leadership in their attitudes toward Automation." Relying on technical analyses of automation led Marcuse to the belief that there was no longer significant opposition at the point of production. Instead, Marcuse should have heeded the words of the workers situated at just this point. Dunayevskaya notes, "It is a question of the voices one hears, the sights one sees, the feelings one experiences depending on which side of the production line you stand." After all, as Denby himself put it, "there is an expression used by miners which is as old as mechanization in the mines. It is simply this: 'A man has no business on a machine who can't break it down any time he wants to.'"

The "News and Letters" group's diagnosis of worker resistance to automation remained somewhat abstract, without a concrete vision of what utopian, or simply more equitable, productive relations might be. However, they highlighted a number of points important to a Marxist theory of social and technological change. Foremost, they held steadfast to the belief that the key to socialist transformation lay not in technological development, but in the struggle of workers, including struggles against new technologies, through which workers would discover new forms of organization and raise deeper political questions. And through their contact with other workers in struggle, they uncovered a desire for autonomous, productive, and sociable working relationships — not simply, as Marcuse would have it, more leisure time.

Perhaps nowhere was that desire for creative and social working relations stronger, or more fully realized, than among the notoriously militant and independent workers at the ports of entry. Stan Weir's memoir of life as a Bay Area dockworker is a moving illustration of the culture of the longshoremen, where hard-fought autonomy encouraged creativity, individuality, and even eccentricity. According to economic historian Marc Levinson, the irregularity of work also meant that dockworkers could, if they wished, take time off for their own activities. The danger, precarity, and difficulty of the work produced a unique culture with powerful bonds of solidarity and an "us against the world" mentality. "Dockworkers saw themselves as tough, independent men doing a very tough job." Long-shoremen dressed as they wished, told jokes, and discussed politics and philosophy. As Weir recounts, "There is presently a nationally recognized small press book club operated by a lone San Pedro longshoreman. There are teachers, artists, realtors, and poets." Author Eric Hoffer worked on the San Francisco docks for twenty years, during which time he wrote acclaimed books such as "The True Believer," a social psychology of mass movements. E.P. Thompson's description of the autodidactic and creative culture of the pre-factory weavers of northern England immediately comes to mind: "Every weaving district had its weaver-poets, biologists, mathematicians, musicians, geologists, botanists."

And like those weavers, longshoremen knew that technology was a death knell for their culture of intimacy and independence. As technology saturated the docks, it transformed the structure of work, and indeed the ports themselves. The shared experiences that bonded longshoremen together in solidarity were balkanized, spread to several different professions and different unions:

> The machinery has caused the boundaries of the longshore industry to become fluid and without exact shape. The maintenance of longshore machinery no longer requires the skills of marine riggers so much as it does that of electricians and truck mechanics. The shoreside cranes are of the type normally run by members of the Operating Engineers Union. The enormous amount of rolling stock that moves containers in the marshalling yards requires the performance of labor that little resembles traditional longshore work. The Teamsters Union has already claimed a portion of this work with some success.

This was the impact of "containerization," the technological standardization of shipping by enclosing all cargo into uniform metal crates. Containers were intermodal: crates could be loaded, by crane, directly onto or off of ships via trains and trucks. This meant the end of the laborious loading and unloading cargo by hand, piece by piece, a process so time-consuming that, in the words of journalist Marco d'Eramo, "Ships therefore spent more time in port than at sea." Containerization meant much less need for workers, something the longshoremen understood immediately. The major dock-worker unions went into negotiations.

The result, on both the West Coast and the East, was the surrender of control of the labor process to the shippers. Long-standing regulations were simply discarded in order to intensify work. According to Weir's account, "On the first day that the mechanization agreement went into effect, hold men found themselves working sling loads of hand-handled cargo that were double or more the weight of those that had been hoisted in and out of hatches the previous day." Output at the docks increased, but at great human cost. "As productivity climbed, so did accident rates. Between 1958 and 1967, U.S. waterfront employers reported a 92.3 percent increase in the number of workers' compensation cases 'despite efforts to engineer problems out of the workforce.'" Higher accident rates stemmed from workers' loss of control of the pace and style of their jobs. The automation contracts negotiated between the International Longshore and Warehouse Union (ILWU) and the Pacific Maritime Association gave up previously negotiated work rules, in particular the 2,100-pound limit on each load of cargo hoisted out of a ship. This strict rule had previously prevented attempts to speed up dock work. While union leadership attempted to channel unrest into wage demands, resistance periodically broke out among the rank and file over working conditions, particularly the physical stresses and the mindlessness of automated work.

Containerization completely upended the dock system. Not only were fewer dockworkers required for the containerized ports, but, by drastically reducing loading and unloading times, and therefore making the shipment of goods dramatically less expensive, fewer ports were needed; many cities, including New York, saw their waterfronts and their associated communities decimated in just a few years. But the revolution didn't stop at the docks. As the transportation of goods was no longer cost-prohibitive, manufacturing could be located wherever labor costs were lowest. It could also be centralized, since there was little need to locate production close to consumption. Containerization was the essential precondition for what has become known as globalization, where production is scattered along far-flung international supply chains.

Like David Noble, Weir notes that containerization was not driven solely by industry needs for efficiency, but also by military order: "The necessary planning for the automation of longshoring and shipping began in 1952 at the initiative of the Pentagon and maritime employers, under the auspices of the National Academy of Sciences." Military contracts to supply the burgeoning invasion force in Vietnam provided commercial shippers with further leverage over intransigent dockworkers during the 1960s. Indeed, without containerization, Levinson argues that "America's ability to prosecute a large-scale war halfway around the world would have been severely limited." In fact, the Vietnam War was instrumental to early processes of globalization: after dropping their materiel in Da Nang, container ships stopped off in Japan to fill up with electronics before returning to the West Coast.

Martin Glaberman, a Johnson-Forest member and auto-worker, keenly observed the effects of automation in industries that had the highest levels of militancy: mining, dock work, and automobile manufacture. "It should be clear," he wrote, "that the problem does not lie in the inability of the unions to find a solution to such problems as automation. They have imposed a solution on the workers." In response to union capitulation, workers took matters into their own hands. As Glaberman wryly noted:

> The workers have no use for the contract and no illusions that contracts can be improved. They have turned to doing their own "negotiating" on the shop floor. Assembly lines have a way of breaking down — and who is to say that the bolt which jammed the line was not dropped accidentally? Who is to know that the warning lights which signal the stoppage of the line were not burned out but merely unscrewed to add a few minutes to the time it takes to repair the line?

This turn to sabotage was, for Glaberman, the germ of new and creative forms of struggle occurring on the shop floor, "a search for new forms of organisation that are adequate for their needs." By many accounts, this search continues on the docks today. In 2011, for instance, ILWU members in Seattle and Tacoma damaged freight cars, dumped grain, and attacked windows with baseball bats during a contract dispute. In Vancouver in 2013, United Grain locked out ILWU dockworkers in response to alleged attacks on its equipment: "The deliberate introduction of a metal pipe approximately two feet long into [the] conveyor system as well as the intentional introduction of a sand and water mixture into the railcar progressor." These techniques draw upon a current of struggle stretching back decades, opened up by the introduction of automation.

The recomposition of the workforce during the Second World War meant tentative gains in the workplace for black workers, as ramped-up wartime production and a massive draft meant that employers were desperate for labor. However, racism and segregation were still the rule. Confined to the worst and lowest-paying jobs and cut off from training programs and apprenticeships that might have provided a greater skill level and more job security, black workers were often first in line when layoffs came. They were therefore disproportionately targeted by the midcentury push for automation, which destroyed "low-skill" jobs, and thus increased black unemployment.

While the civil rights movement is often remembered for its dramatic confrontations over the desegregation of civil society, from lunch counters to buses to public schools, labor politics were also a central concern. Indeed, A. Philip Randolph's plans for a national march on the capital to desegregate the defense industry pushed Roosevelt to do just that in 1941. Two decades later, Martin Luther King Jr. and Bayard Rustin organized the March on Washington for Jobs and Freedom. As the integration of the workplace was a major goal, postwar civil rights leaders viewed automation, and its disruptive impact on labor markets, as a potential problem for their movements. In a 1961 address to the AFL-CIO, Martin Luther King Jr. declared, "Labor today faces a grave crisis. In the next ten to twenty years automation will grind jobs into dust as it grinds out unbelievable volumes of production." King understood that automation was a weapon to be used against organized labor: "This period is made to order for those who would seek to drive labor into impotency by viciously attacking it at every point of weakness." And the unions' only chance to take control of the course of automation was to forge a common cause with the civil rights movement: "The political strength you are going to need to prevent automation from becoming a Moloch, consuming jobs and contract gains, can be multiplied if you tap the vast reservoir of Negro political power." Malcolm X, in contrast, argued that the threat of automation justified a separatist strategy. "At best," he cautioned, "Negroes can expect from the integrationist program a hopeless entry into the lowest levels of a working class already disenfranchised by automation."

The link between automation and the fate of black Americans drew increasing interest over the decade. In 1964, an assortment of figures from the American left and intelligentsia — including Students for a Democratic Society leaders Tom Hayden and Todd Gitlin, democratic socialists Michael Harrington and Irving Howe, anti-nuclear activist Linus Pauling, cyberneticist Alice Mary Hilton, civil rights leader Bayard Rustin, and erstwhile Johnson-Forest Tendency associate James Boggs — formed the "Ad Hoc Committee on the Triple Revolution." The committee issued a statement in the influential New Left magazine "Liberation," and sent a copy to President Lyndon B. Johnson, warning of oncoming destabilization at the hands of three interconnected trends: automation (referred to as "cybernation"), the replacement of conventional weapons with nuclear ones, and the civil rights movement.

This "triple revolution" posed an existential challenge to US institutions — one that would require dramatic policy shifts if it was to be overcome, well beyond the ambitions of Johnson's War on Poverty. Without a radical restructuring of the economy in the face of automation, the civil rights demand for inclusion could not be met (and the reduced need for standing armies meant the military would not sop up the surplus, as it had decades earlier). The Ad Hoc Committee's statement pronounced: "The Negro is trying to enter a social community and a tradition of work-and-income which are in the process of vanishing even for the hitherto privileged white worker. Jobs are disappearing under the impact of highly efficient, progressively less costly machines." If nothing was done, millions would be immiserated: "A permanent impoverished and jobless class is established in the midst of potential abundance."

> We assert that the only way to turn technological change to the benefit of the individual and the service of the general welfare is to accept the process and to utilize it rationally and humanely. The new science of political economy will be built on the encouragement and planned expansion of cybernation. The issues raised by cybernation are particularly amenable to intelligent policy-making: Cybernation itself provides the resources and tools that are needed to ensure minimum hardship during the transition process.

The document went on to enumerate the incredible scale of the measures required: more education spending, a large public works program, affordable housing, investment in mass transit, and a more egalitarian tax system that would redistribute income. While Johnson's advisor Lee C. White assured the committee that the president would establish a commission to study the issue, the manifesto failed to have its intended effect, and the backlash was swift. "Guaranteed Income Asked for All, Employed or Not," scoffed the New York Times, while Daniel Bell, the prominent social theorist and political commentator, quibbled with the economics. Ultimately, the letter, however cogently argued, failed to make an impact on policy makers, who opted to do nothing.

> An important factor in the Negro unemployment problem was industry's substitution of machinery for unskilled labor … Negroes laid off as a result of these developments and young Negroes who found that industry was no longer hiring the unskilled became significant proportions of the hardcore, long-term unemployed.

In "Monopoly Capitalism," Paul M. Sweezy and Paul A. Baran's 1966 analysis of capitalist political economy, the pair of Marxist economists agreed that automation had severely circumscribed the economic prospects of African Americans working in manufacturing: "Since 1950 … with unskilled jobs disappearing at a fantastic rate, Negroes not qualified for other kinds of work found themselves increasingly excluded from employment altogether." By the early 1960s, black unemployment was double that of white unemployment. While Baran and Sweezy were eager to place the blame for black hardship on the capitalist system as a whole, rather than its technology, they admitted that "within the framework of this society technological trends, because of their differential impact on job opportunities, can rightly be considered a cause and undoubtedly the most important cause, of the relative growth of Negro unemployment." Ernest Mandel used this political economic framework to explain growing black radicalization: "The rapid decline in the number of unskilled jobs in American industry is the nexus which binds the growing negro revolt, especially the revolt of negro youth, to the general socio-economic framework of American capitalism."

Radical black intellectuals dealt with automation as a serious social problem for black workers and liberation movements. Robert L. Allen's widely read study of 1960s black radicalism, "Black Awakening in Capitalist America," sounded a dire note. "Not only is the economic situation of the masses of blacks grim," he wrote, "but the prospects are that it will not improve, rather it will deteriorate. This is due partly to the unregulated impact of automation." Allen predicted the continued degradation of the social power and conditions of black people over the next decades, a prediction as stunning for its pessimism (his book was published in 1969, a high-water mark for US radicalism) as its prescience.

For sociologist Sidney Willhelm, writing in 1970, automation threatened to upend the achievements of the civil rights movement: "Though widespread opinion strongly supports the view that integration is likely, many signposts indicate eventual isolation of the Negro people, an isolation made possible by the changing technology of automation." According to Willhelm, for most of US history black Americans occupied a contradictory location — one in which they were subject to racist abuses but simultaneously required as a superexploited population of laborers. However, the replacement of human labor with machines would undermine black workers' economic function. "If machines eventually accomplish what man has in fact developed up to this point," he remarked, "then perhaps people will become surplus baggage? But, then, how long will we tolerate one another as surplus baggage?" Willhelm foresaw a dire situation where, bereft of economic utility, black Americans would be subject to unchecked racism and deepening segregation:

> The Negro is losing out because he is losing out in the technological development of American society; *White America can, for the first time, easily bear the economic costs for implementing its racial values to the point of excluding the Negro race*. More specifically, the developing outcast position of the Negro is in keeping with the technological configuration of White America's economic interests.

In raising the specter of black people as an unneeded "surplus baggage" to the economy, Willhelm did not hesitate to suggest genocidal outcomes: "Where economic competition against whites once contributed to race relations, now the Negro competes against the machine; the first competition resulted in white extortion, the second brings in its wake the Negro's obliteration."

This was an analysis that the Black Panther Party took into the heart of their organizational philosophy, which was geared toward organizing the "lumpenproletariat": the class cut off from wage labor. As Eldridge Cleaver spelled out, the lumpen, which included those "who have been displaced by machines, automation, and cybernation," represented a real contradiction within the proletariat. Indeed, machines were, in part, responsible for this bifurcation. The polarization of skill meant that "every job on the market in the American Economy today demands as high a complexity of skills as did the jobs in the elite trade and craft guilds of Marx's time." This elitist configuration had sapped a portion of the proletariat of its revolutionary zeal, an elán that was now the preserve of the technological outcasts. Huey Newton clarified Cleaver's points in terms of a longer-range strategy:

> In this country the Black Panther Party, taking careful note of the dialectical method, taking careful note of the social trends and the ever-changing nature of things, sees that while the lumpen proletarians are the minority and the proletarians are the majority, technology is developing at such a rapid rate that automation will progress to cybernation, and cybernation probably to technocracy … If the ruling circle remains in power it seems to me that capitalists will continue to develop their technological machinery because they are not interested in the people … Every worker is in jeopardy because of the ruling circle.

Meanwhile, black workers who remained on the factory lines formulated their own analysis of automation. Productivity had increased dramatically over the past decades, but while management credited the machines, these workers pointed to the dangerous speedups they had been subjected to, what they called, in the factories of Detroit, "niggermation." The injuries and deaths — dozens a day, a casualty rate higher than in the Vietnam War — meant that workers understood the factory as itself a warzone. With their concerns over racism and new technology ignored by both management and established unions, black workers formed militant groups like the League of Revolutionary Black Workers and the Dodge Revolutionary Union Movement, successfully shutting down auto factories through wildcat strikes.

In the midst of the tumult of the 1960s, the most significant black radical groups placed a critique of technology at the center of their analysis and their politics. They recognized that the technological recomposition of the workforce would shape the fate of their struggles. In the 1972 update to the Black Panther Party's Ten-Point Program, the last point added "people's community control of modern technology" to the demands for "land, bread, housing, education, clothing, justice, peace."

New technology also came into focus for the women's movement, perhaps the most intellectually vibrant formation of the postwar period. Some women speculated that technology, by reducing the need for strength and skill in jobs, would have a leveling effect on the workplace, allowing women into positions that had been off limits to them. This perspective was questioned by feminist labor process theorists like Cynthia Cockburn, who directly studied the impact of technology on the gendered division of labor. As feminist researchers have continually found, rather than simply remove barriers, new technologies recomposed work in complex ways, often to the detriment of women. "Technology is far from neutral," Cockburn plainly states. "Industrial, commercial, military technologies are masculine in a very historical and material sense. They cannot readily be used in a feminine, nor even a sexless, mode."

Indeed, in many cases, automation eliminated and degraded jobs already occupied by women. A case in point is the fate of telephone operators. A class of worker made up entirely of women, operators struggled for decades against increasingly mechanized working conditions, which concentrated more and more switchboard work among fewer and fewer workers while eliminating downtime. When operators joined unions, their core demands about the degraded quality of their work were cast aside, channeled instead into more traditional trade union demands. Historian Venus Green, a former telephone operator herself, documents how the bureaucratic trade union mentality meant ignoring the concrete demands of operators. "Higher wages, shorter hours, and union grievance procedures," she writes, "did not ameliorate the strictly supervised and relentless, machine-driven work pace from which operators suffered."

Instead of taking operator demands seriously, union leadership praised automation for its promise of leisure. For instance, Communications Workers of America President Joe Beirne stated, "For ourselves, we welcome automation because we see in it a level for higher wages, longer vacations, shorter hours, and ultimately, greater security for ourselves and the American people." If these beneficial effects were to come — and they largely were not — they would be at the expense of operators, who found themselves out of jobs. Green suggests sexism played a part in union indifference, and notes that the male leadership viewed the recomposition of telephone operation as beneficial to male workers, who could take new positions as technical supervisors over automated dialing machines. Those same machines put telephone operators out of their jobs, while removing autonomy over the pace of work from the women who remained. Ultimately the only "long vacations" and "shorter hours" were those that accrued to the unemployed: the thousands of dues-paying women who lost their jobs.

Beyond the workplace, feminists analyzed and politicized the unwaged work of the housewife, the labor of social reproduction that was essential to capitalism but performed without remuneration. Here too was another area of life where technology promised to alleviate burdens and spread the gifts of leisure; and here too those promises went unfulfilled.

Homemaking had long been of interest to scientific management. In the best-selling 1948 memoir "Cheaper by the Dozen," two of the twelve children of industrial engineer Frank Gilbreth, a fervent disciple of Frederick Taylor, recount humorous anecdotes of their father's home life. Gilbreth, the inventor of so-called time and motion studies, possessed an enthusiasm for order and efficiency so all-consuming that he made it the basis of his parenting style. In the Gilbreth home, no moment could be wasted. Baths were supplemented by German language lessons and clearing the table was done according to stopwatch. The book's comedic peak sees Gilbreth pressuring a doctor into performing tonsillectomies of all twelve progeny according to the apparent insights derived from his time and motion studies.

While the book, and its successful 1950 film adaptation, play up Gilbreth's zeal for family-friendly laughs, sociologists Tilla Siegel and Nicholas Levis point out that it can also be read as a prescient depiction of how the rationalization of work quickly bleeds into the rationalization of the domestic sphere. In fact, the most influential capitalists understood this connection well. Henry Ford, troubled by massive resistance to his new assembly line, knew he would need more than a high wage to coax workers into his factories. He would have to mold the worker's subjectivity far beyond the shop floor, what Antonio Gramsci recognized as "the need to elaborate a new type of man suited to the new type of work and productive process." Creating this "new man" was the job of the Ford Sociological Department, who conducted interviews in the factory towns about the organization of workers' home life, collecting details on everything from alcohol consumption to sexual habits. Ford placed a heavy emphasis on the standardization of gender roles: the discovery of an employee's wife working outside the home could be grounds for dismissal.

Gilbreth and Ford's examples demonstrate that the project to rationalize the home has more in common with the needs of patriarchy and capitalism than the needs of women. This is precisely the argument Selma James and Mariarosa Dalla Costa make in "The Power of Women and the Subversion of the Community," their 1975 manifesto for the Wages for Housework movement. In James and Dalla Costa's analysis, the result of Ford's project was a mutually beneficial arrangement between capitalism and patriarchy in the construction of the nuclear family. The nuclear family, they observed, compelled women into providing necessary labor that did not have to be compensated by capital, and the drudgery reproduced women's isolation and subjugation to the male wage earner. Therefore, "the maintenance of the nuclear family is incompatible with the automation of these services. To really automate them, capital would have to destroy the family as we know it." Because domestic labor is a relation of patriarchal power, technology will not liberate women. Instead, their time will be filled with more domestic work: "We all know the saying too well: you can always find work to do in a house." They conclude that it is instead feminism, not technology, that will liberate women from housework.

Ruth Schwartz Cowan's 1983 "More Work for Mother," one of the most acclaimed histories of domestic technology to emerge in the wake of the women's movement, confirms James and Dalla Costa on an empirical level. While official discourse — advertisements, news articles, patents — tell a story of an easier workload for homemakers, "when discussed by the people who actually did housework, or by the people who watched the people who were doing it, it seems not to have become one whit more convenient — or less tiring — during the whole of the century."

Cowan, surveying domestic technology from 1860 to 1960, found that rather than relieve beleaguered housewives, domestic technologies tended to ease home activities traditionally performed by men. Innovations such as the gas cooking stove and the automatic flour mill freed men from chopping wood and grinding grain, allowing them more time to work outside the home. Meanwhile, women assumed a greater proportion of domestic labor: "You [the housewife] bore the whole burden of housework. For your husband and your children, the house became a place of leisure." According to Cowan, one result of the mechanization of housekeeping was a dramatic reduction in the number of paid domestic laborers, with the concomitant extension of housekeeping duties for the woman of the home. The introduction of the washing machine replaced the reliance on professional laundresses — instead, the housewife could do the work for free. The mechanization of the work of social reproduction, in this account, serves to shore up a gendered division of labor, rather than reduce the time spent in toil. "The end result is that housewives, even of the most comfortable classes (in our generally now comfortable society) are doing housework themselves."

Yet in spite of so many feminist scholars' affirmations that technology, both at work and at home, functioned to reinforce the gendered division of labor, the faith remained strong that these new developments could somehow undermine it. One of the most striking technophilic manifestos was Shulamith Firestone's "The Dialectic of Sex," a work whose theoretical sophistication and prosodic brio have kept its influence alive for the half century following its publication.

Firestone begins with the assertion that the oppression and exploitation of women — what she calls "sex class," which she argues is distinct from, and prior to, economic class — is rooted in the supposed biological differences between the sexes, specifically those related to human reproduction. The rigors of childbearing and childrearing fell to women, rendering them vulnerable and dependent on men, and thus leading to a subordinate position that was reproduced through the structure of the family. However, biology, in her view, was not destiny. For Firestone, new technologies of contraception and fertilization held out the potential to diminish the biological basis for sex class, though she was quick to point out that "new technology, especially fertility control, may be used against [women] to reinforce the entrenched system of exploitation." Hence the need for a thoroughgoing feminist revolution, modeled on the Marxist dictatorship of the proletariat, which would entail "seizure of control of human fertility — the new population biology as well as all the social institutions of childbearing and childrearing." Indeed, according to Firestone, the eruption of the feminist movement was predicated on the existence of such technologies. "Feminism," she wrote, "is the inevitable female response to the development of a technology capable of freeing women from the tyranny of their sexual-reproductive roles." In this way, reproductive technologies come into conflict with the existing relations of reproduction — the family — and thus begins the era of feminist revolution.

Firestone borrows from the structure of Marx's most technologically deterministic works, arguing that existing technologies are themselves neutral, becoming exploitative only in their specific uses. While Firestone carefully enumerated potential and actual abuses of contraceptive technology, such as the use of poor black and brown women as human test subjects, she maintained that abuses have to do with who is in power, rather than the technologies themselves. She likened her position to that of ecomodernists, who, rather than conserve nature, wanted to deploy technologies conscientiously to reshape it in egalitarian ways.

> As was demonstrated in the case of the development of atomic energy, radicals, rather than breastbeating about the immorality of scientific research, could be much more effective by concentrating their full energies on demands for control of scientific discoveries by and for the people. For, like atomic energy, fertility control, artificial reproduction, cybernation, in themselves, are liberating — unless they are improperly used.

To be sure, romantic, and reactionary, notions of nature abounded in the counterculture — the "goddesses" that technology theorist Donna Haraway disdains in favor of cyborgs. But the enthusiasm for technology also caused Firestone to make outsized claims. Even under capitalist patriarchy, Firestone believed in automation's progressive effects on gender relations:

The polemical edge of Firestone's work meant that it was fiercely contested among feminists. Many pointed out that, rather than liberate women from childbearing, reproductive technologies would relocate the locus of control over childbirth away from women and further toward the male-dominated fields of science and medicine. In 1985, the Feminist International Network of Resistance to Reproductive and Genetic Engineering, for instance, highlighted the connection between such technologies and imperatives for eugenics, medical experimentation, and population control, demanding, rather Luddishly, "a halt to the research and application of reproductive and genetic engineering in all its forms." Contemporary critics such as Sophie Lewis rightly call attention to the pitfalls of FINRRAGE's stridency, including their deep attachment to gender essentialism, and their Manichean view of technology that ignores legitimate needs of poor women in the global South. As Lewis notes, abolitionist initiatives such as FINRRAGE's "perform opposition to commodification rather than to capitalism." Yet despite these limitations, as feminist technology scholar Judith Wajcman argues, FINRRAGE rightly pointed to how patriarchal politics were embedded in the technologies themselves: for example, in the promotion of methods of in vitro fertilization that uphold the genetic ancestry of a heterosexual couple, rather than alternative forms of parentage that might destabilize the nuclear family. The group's demands for "the recovery by woman of knowledge, skill and power that gives childbirth, fertility and all women's health care back into the hands of women" strikes at the heart of patriarchal domination of technology.

In the postwar period, automation became a major political flashpoint for sundry radical movements, whether they focused on the docks, the factory, or the home. The political divisions were clear enough: workers repeatedly rebelled against these new technologies, while unions worked with capital to discipline unruly workforces to the machines. As radicalism intensified through the 1960s, capital accelerated its pace of technological change as part of a massive global restructuring around the various insurgencies breaking out across the postwar order. At the center of this change was a specific technology that the counterculture greeted with both fear and fascination: the computer.

3 — Against Automation

Automation as Control

The JFT and the Miners

The Docks

Black Workers and Automation

Feminism and Automation