The Congress of Industrial Organizations, better known as the CIO, was the main labor federation representing America’s mass production workers. So think auto, steel, rubber, electrical, packinghouse. It existed for some twenty years, formally from 1935 to 1955. It started as a committee within the American Federation of Labor [AFL] dedicated to industrial organization. But in 1938, it split and formed its own separate and rival international labor federation to the AFL. Then in 1955, it went back into the AFL and formed the AFL-CIO.

It’s the organization we associate with finally figuring out how to unionize America’s large industrial corporations. It represents the triumph of one form of union representation: industrial unionism — the form most appropriate to mass industry. Union organization moved forward outside of mass production, surging with the organization of miners at the turn of the century and peaking at close to 20 percent at the end of World War I. But it wasn’t until the 1930s that unions organized mass production, despite all the attempts before that.

Founded in the late nineteenth century, the AFL was in existence for fifty years or more before a wing of it split off to form the CIO. The largest labor federation from the 1880s to the 1950s, it was a diverse and decentralized organization. It had many different kinds of union affiliates — machinists, garment, mining, transportation, hospitality (a sector I’ve written a lot about) — and its leaders were also diverse, particularly at the state and local level. A lot of AFL leaders belonged to the Socialist Party or, if not card-carrying socialists, were sympathetic to socialism and to progressive third-party political formations. In addition, there was always a large group inside the AFL pushing for expanding union organizing and for experimenting with new forms of unionism, and that included industrial-style organizing.

The standard view of the AFL is that its commitment to craft unionism held back the organizing of mass production workers. I think there’s truth to that. Certainly, there were AFL leaders who believed there was only one right way to organize workers and that was by trade or craft. There were also AFL leaders who feared an influx of mass production workers into their organization. They might lose power or be voted out of office. There were some who simply did not care about organizing new sectors.

My point has been that to reduce the AFL to that wing misses its complexity and heterogeneity. Certain sectors of the AFL were always pushing for industrial organization, and there were massive organizing campaigns in mass production before the ’30s. Those experiments were important, and in many ways they helped build the unions that ended up founding the CIO. We know that some of them failed, the big efforts in 1919 to organize in packinghouse or in steel, but there were other attempts that did succeed.

Here I’m thinking about the organizing of the garment and the mining industry. The ILGWU [International Ladies Garment Workers Union], which was an AFL union founded in 1900, grew by some two hundred thousand members in the two years before the CIO became a committee. John L. Lewis and others helped reorganize the miners too. We can think about the Teamsters. They were rebuilt.

The final thing I’d say about the AFL and how, in some ways, it paved the way for the CIO, is that as limited as its legislative agenda was, the AFL’s almost obsessive focus on what I would call “labor civil rights” paved the way for successful organizing in the 1930s, both in mass production and in other sectors. The AFL pushed for decades for basic labor and civil rights: the right to organize, the right to free speech, the right to picket, boycott, and assemble. It changed the conversation around these issues and was instrumental in passing, in 1932, the Norris-LaGuardia Act.

That legislative breakthrough should be thought of right up there with the Wagner Act, because it limited some of the horrendous judicial restrictions on labor’s direct-action techniques and tactics. From 1932 to 1947, from the passage of Norris-LaGuardia to the enactment of Taft-Hartley, unions had more freedom to strike and to engage in secondary boycotts, recognitional picketing, mass picketing, and all kinds of pressure tactics that they didn’t have before or after. I think that’s an important context for understanding the rise of the labor movement in this moment.

The garment unions in particular represented a mix of workers in terms of skill but also in terms of backgrounds, men and women, foreign-born, native-born. The garment unions also were very progressive in outlook.

I think Hillman and other garment leaders fully understood that if you were going to organize mass production and take on the world’s largest corporations that you were going to need the support of the political sector, of the political classes.

They had a vision of political unionism and of the state weighing in on the side of workers or, at least, not weighing in on the side of capital, that made them believe they could take on industrial capital and win. They’d also done it before. They had figured out how to organize industrial workers before World War I, and they saw what happened when the Republicans came back in power in the 1920s and you had a wholesale assault on the labor movement. So they sought a reinvigorated, more worker-friendly Democratic Party. They were pushing for union organization with the support of the state.

There are a lot of things that came together in the ’30s. It was a kind of perfect storm. It’s a moment of capital crisis; business leadership is discredited by the economic collapse. It’s an era of mass unemployment, homelessness, anxiety, insecurity. We’ve got a desperate working class. We’ve also got a working class that’s infuriated by the petty tyranny of foremen and the relentless speed-up of production. But this dream of organizing mass production has been around for a while and there were many other historic moments of economic crisis that didn’t result in the organization of workers.

So how did the CIO win? How did it take advantage of this moment? I’ll just lay out a couple of the key things that I think were important. I would put right at the top of the list the ability of workers to exert structural or positional power. I think of the sit-down strike as extraordinarily effective in terms of stopping production and creating chaos for employers.

It only took a few people to make it effective. You didn’t have to do much except stop work. It was something that people could participate in. But there were other forms of nonviolent direct action too. The recognitional strike, which became illegal after 1947, was actually the primary way that workers exerted pressure on employers and got them to agree to sign contracts.

Richard Freeman, a Harvard labor economist, has done incredible work tabulating the number of recognitional strikes in the 1930s and arguing for how important they were in getting employers to the table. A huge proportion of the contract victories came as a result of this kind of worker pressure from strikes and picketing. Most of the earliest and most important union victories didn’t involve elections: employers agreed to recognize the union in exchange for the pickets or the sit-downers being withdrawn.

Unionism was contagious. Myra Wolfgang, who helped organize the hotel and restaurant workforce in Detroit and other places, always told this great story that reveals the contagious nature of unionism. She would say she’d be sitting in her Detroit office and the phone would ring and she’d pick it up, and the person at the other end would say something like, “Hi. I’m Mary Jones at Woolworths. We’ve thrown out the manager and we’ve got the keys. What do we do now?” Workers saw that the sit-down and other organizing tactics were effective. They believed they could win and they put their freedom to strike to good use. The demand for unions came from below. It wasn’t the CIO imposing it.

There are other things that made a difference as well. The political transformations of the ’30s, the change in the Democratic Party, and the election of a New Deal administration willing to back workers’ right to full association and collective bargaining. There was support from the public and from the political intellectual classes. You had a left that focused on worker power and had a class-oriented vision. It was a democratic left movement that listened to workers, that channeled their needs, that wasn’t afraid of new ideas.

Lastly, one of the things that’s always been most appealing to me about the CIO is that it spoke for the particular needs of the working class, but at the same time it had a vision of how the labor movement could make the whole society better. It practiced what some call social unionism. So multiple factors — some internal to the labor movement, some external — all came together in the 1930s and enabled the transformation and rebuilding of the labor movement.

It’s a mixed record really. It depends on which unions we’re looking at and which leaders. Let me start with some of the admirable things about the CIO in terms of its record toward women and, particularly, toward organizing women.

The CIO reflected the culture at large in many of the ways it thought about gender and the family, but it was more progressive than that culture as well. On the positive side, the CIO stressed that the best way to organize mass production was through unity and solidarity, and to create an organization that was inclusive, that represented all workers, men and women, skilled and unskilled, all races and nationalities. In part, because of that, the CIO organized women in much greater numbers than the AFL.

Also, due to the CIO’s commitment to equal treatment and to wage compression, it benefited women in the sense that the wages of those at the bottom moved up more than at other levels, and women were at the bottom. We know that unionization raises the wages of women more than men, and lower income workers more than those at the top. That was true in the past and is still true today.

It’s important to understand that CIO policies also varied from union to union, in part based on pressure from the rank and file. So in organizations like the UAW [United Auto Workers] where there were large and active women’s departments, you had a more progressive record. The UAW, for example, was at the forefront of the equal pay drives in the ’40s and ’50s, which by the 1960s resulted in new state laws and, at the federal level, the 1963 Equal Pay Act.

But there are areas where I would say the CIO didn’t listen enough to its women members. Probably its biggest failing was not addressing what women activists in the labor movement — a group I called “labor feminists” in my 2004 book, The Other Women’s Movement — referred to as the “double day,” or the problem of long work hours for women in both the household and the labor force. There are a lot of ways to respond to the double day, and women didn’t always agree either on how to fix it, but the labor movement certainly didn’t give the problem sufficient attention, nor did it recognize and sufficiently value women’s unwaged work in the home.

So, on the one hand, the CIO pushed for women’s rights to jobs and to equal pay, and it sought to raise household income and wages overall, but it often assumed that men had a bigger claim to the good jobs that provided a living wage or enough income to support a family. It also assumed that all women would end up married to men, living in households where income was shared equitably. Obviously, that wasn’t — and isn’t — always true.

The CIO needed to focus on policies that would have lessened the double day — childcare, family leave, shorter hours — and those things weren’t always a priority for the CIO.

My argument has always been that we need to see both the virtues and the limits of various forms of unionism. I do think there’s much to admire about the CIO: its willingness to tackle the citadels of corporate power, the courage of its organizers. They faced death threats, and they also died trying to organize unions. It was dangerous and not for the faint of heart. To get a sense of what it took, you just have to read Rose Pesotta’s memoirs about organizing in the rubber industry, or accounts of Operation Dixie and organizing in the South. There’s much more to admire about the CIO as well.

But the industrial model has its limits as well, especially in terms of organizing workers outside of mass production, including service, retail, domestic workers, transportation, and other non-factory workers.

Where you have small shops and a mobile workforce, you need a different kind of approach, so the ballot box election doesn’t make sense. The notion of focusing on individual firms or enterprise-based organizing is a problem. You need a different approach — what I’ve called occupational or horizontal organizing. And in building union institutions, the challenge is to figure out how to create structures that recognize both the identity workers sometimes have with their job — think machinist or dressmaker or professor or actor — with the need for a broader solidarity to confront employers and capital.

The main thrust in my work has been to think about how different sectors of the economy differ in terms of the structure of capital and the work that’s being done and how you might need a variety of approaches to worker power and representation. There’s no single approach to organizing and union-building that fits all situations.

My first book was called Dishing It Out: Waitresses and Their Unions, and it appeared in the early ’90s. I wrote it in the ’80s at a moment when no one thought service workers could organize, especially women service workers. The conventional wisdom was that unionism was a phenomenon of blue-collar male workers, and that with the decline of manufacturing and mass production, we would inevitably see the decline of unionism.

I wondered about that prophecy of union decline based on the rise of female employment and the turn toward a service economy. I became interested in whether service and other non-manufacturing workers had organized historically, and in particular if women had organized and how they did it. I found out that the answer was yes. I didn’t realize that waitresses were going to turn up as one of the most well-organized groups of women service workers, but that’s what I found. Close to a quarter of all food servers were unionized by the ’50s.

The question was, how did this happen? First, it’s important to note that the vast majority were in the AFL hotel and restaurant employees union, and that they relied on organizing approaches often associated with the AFL, what I called occupational unionism. They rarely organized individual restaurants; rather, they sought to secure agreements with all the restaurant employers in a particular labor niche or market or geographic region — what some might call sectoral organizing now.

That way workers could move around in the industry from one union job to another, and employers would compete on the basis of good food or good service rather than on cutting worker wages and lowering standards. To unionize, they relied on what I think of as a carrot and stick approach. In terms of the stick, they used recognitional strikes, picketing, and secondary boycotts. The support and solidarity actions of delivery drivers — those delivering milk and bread to small restaurants — were really important in helping organize hospitality.

Alliances with customers and community groups were also important. Waitresses and other hospitality workers used the old AFL tactic of union bar cards and union house cards. These placards sat in the front window of restaurants that paid the union scale and abided by union rules. They alerted customers as to whether they should come in or not. That kind of positive advertising of high road employers was effective, particularly in working-class and heavily unionized communities.

AFL unions in hospitality and other sectors developed practices that were appealing both to workers and small employers. They were involved in training and providing health benefits, what Sidney and Beatrice Webb called mutual aid unionism. They set up hiring halls to make sure that small employers had a source of labor power, but also that the workers who were dispatched went to situations where union wages were paid.

For me, writing about the history of hospitality organizing and of the AFL more broadly pointed to how our labor histories need to broaden and how we need to think about the multiple ways, and the creative ways, that workers used to organize and exert power in the past. We can learn from the AFL as well as from the CIO.

I think the best of the CIO ideals are still with us. Its commitment to expand into new and uncharted territory. We just have to think about Workers United [Labor Union] and the Starbucks campaign to recognize how important that quality is. Its belief that workers can organize, that they are a source of intelligence and creativity. Its commitment to democracy and education. Its embrace of political and social unionism. Its attention to universalistic policies and raising the living standards for the majority.

And finally, one of the things I admire most about the CIO was that it was willing to listen to workers. It wasn’t afraid of giving workers power. It wasn’t afraid of new ideas. I think the labor movement today needs to fully embrace that approach, to be open to a wide range of union forms and ways of increasing worker power. Contract unionism is important, and we cannot give up that fight. But there are many ways to build worker power — minimum wage campaigns, worker co-ops, worker-owned platforms. So I think openness to the different needs of workers and the different ways they organize is fundamental to rebuilding the labor movement.