The CIO first formed as a committee within the American Federation of Labor, or the AFL, in 1935. It was called the Committee for Industrial Organization, and later it broke away from the AFL and formed a separate organization, the Congress of Industrial Organizations. It was a group of unions that decided it was time to organize on an industrial basis rather than an occupational basis, as most AFL unions had traditionally done. So following a lot of changes in the economy, the way in which companies were organized, and the de-skilling of work that led to the possibility of mass production on assembly lines, the old-fashioned way of organizing people based on their occupational or craft group made less and less sense.

It’s not that the CIO leaders were the first to figure that out. There were lots of experiments with industrial unionism prior to the 1930s, but at this point in the middle of the Depression, and in a period when labor law was changing in a way that was beneficial to union organizing, a bunch of labor leaders decided this was the moment for change.

It goes back to the nineteenth century when mass production didn’t yet exist, and modern labor law didn’t exist either. It was very hard to unionize, a lot like today actually, in that the law was totally stacked against unions. And so one of the few sources of power for workers was skill. If you were a skilled carpenter or machinist, the employer depended on you much the way employers depend on, say, IT workers today. If you stopped working, it would be very disruptive to an employer, and so that became the primary basis for successful organizing in the nineteenth century. If you couldn’t be easily replaced, you had some power, and you could extract concessions from employers, and that was much harder to do if you were not skilled.

There were attempts to organize unskilled workers here and there, and some of them succeeded, though not usually in a sustained way. The most famous is the Knights of Labor in the 1880s, which had a meteoric rise and fall. They organized everybody, but that was difficult to sustain in this period where there was no law protecting the right to collective bargaining.

I should say first that the prior history was not 100 percent failures. For example, the garment workers in the 1910s successfully organized on an industrial basis, starting in 1909 with the big strike in New York City that is sometimes called the “Uprising of the 20,000.” That was industrial unionism. The union involved had previously included only the cutters and the other skilled male workers, and then there was this uprising of the less-skilled masses of women workers. They successfully organized by the tens of thousands and won a union. But it was difficult to sustain. The mine workers also organized on an industrial basis quite early and relatively successfully. And of course, the head of the [United] Mine Workers union, John L. Lewis, was the main leader of the CIO later.

But what changed in the ’30s? Well, a bunch of things. I already mentioned the rise of mass production, which makes focusing on skilled craft workers less and less logical, because suddenly you have these masses of workers gathered together in large factories, and most of them can be easily replaced. They aren’t skilled enough to have the kind of leverage that the AFL unions once had. So more and more people start thinking, well, maybe instead of just organizing people based on skill, we should organize whoever the boss hires, and our power will be in unity, not in irreplaceability.

There’s a lot of experimentation with that idea before the ’30s, but then with the Depression, the New Deal, and the passage of the Wagner Act giving workers the right to collective bargaining, the moment seemed opportune to launch this on a big scale, and that’s what happened. There was a lot of resistance from employers, despite the law, just like today, but they did pull it off.

But of course, it’s during the war that it really takes off. In the late ’30s, there are some big breakthroughs, most famously the 1937 recognition of the United Auto Workers [UAW] after the strike in Flint, Michigan. That’s the big moment when people say, “Yes, we can do this,” and it starts spreading after that. But it’s really when the war leads to the dramatic expansion of manufacturing in the US, as the economy is completely converted to producing military goods, that the big numbers come.

There’s a big debate about that among scholars. I’m a little bit of an agnostic. One side points to the strikes that occurred prior to the passage of the Wagner Act in the early ’30s — the Teamsters strike in Minneapolis, the Ohio auto parts strike, the west coast waterfront strike. Some people argue that worker militancy made FDR and his colleagues concerned that there was going to be a lot of industrial strife and disorder, and this was not good in the middle of a depression when they were trying to get things going again economically. In that view, passing the Wagner Act was a co-optation of worker militancy. That’s one view. The other view is that it was only after the passage of the law that the CIO was really able to take off.

By the way, the AFL unions also grew in the late ’30s, partly because of the Wagner Act, so it was not just the CIO. In my generation, which came of age in the 1970s and in the waning days of the New Left, the dominant view was that the AFL unions were racist, sexist, and anti-immigrant, so we tended to be dismissive of their history. We believed that the only unions worth paying any attention to were the unions that came with the CIO. They had their warts too, but they were much better. That was the orthodoxy. So the fact that the AFL also grew in the 1930s was a surprise when I learned about it. So at least for my generation, there was a distorted view of the two federations that we’re still trying to correct.

Anyhow, the other view is that it was the law that was the key; that Roosevelt, as a progressive, forward-looking guy, concerned about the Depression and how to get out of it, wanted industrial stability and therefore fought along with Robert Wagner and the others for the passage of the Wagner Act against great resistance from employers and their political advocates in Congress.

I think the truth probably lies somewhere in between, that the militancy did contribute to the desire to stabilize industrial relations on the one side. In fact, if you look at the language of the Wagner Act, it actually says that the purpose of the law is to level the playing field, stabilize industrial relations. That’s what they were thinking about. And at the same time, once it passed, it created completely new conditions that enabled unions to really gain traction. Ultimately, they’re not mutually exclusive views, and I myself come down in the middle.

I think you absolutely need both. So for example, take the Flint strike of 1936, ’37. Even with the law on their side, I’m not sure the UAW could have possibly gained a foothold without both the militancy of ordinary workers and the strategic capacity of the leaders. They didn’t just spontaneously rise up — they had a whole plan of action that was extremely successful in how to outwit General Motors. They pulled it off amazingly.

It also really helps to have the law either neutral or on your side. There’s no question about that because labor law has been used as a battering ram against unions throughout American history. So, in the 1930s, when it seems to be tilting in a way that doesn’t favor workers, but at least equalizes the playing field between management and labor, that’s key. That’s not either top-down or bottom-up, but it’s a precondition for successful organizing.

That’s part of why it’s so difficult to organize big corporations today. The employers have all the cards in terms of the legal stuff. The places where unionism is winning a foothold, just like in the days of the old AFL, is where workers have skill. So medical interns and residents, they’re winning. Graduate student workers and adjuncts, who are not easily replaced. Maybe more easily replaced than residents and interns because there are a lot of us academic workers out there, but still, it’s a very highly skilled job. You can’t just fire everybody and start over tomorrow. Journalists are another example. They have a lot of skill and they’re winning.

So in a way we are back to those days where skill really matters. The law is less relevant because you have power outside the law. It doesn’t matter so much if you have other leverage, based on the fact that you’re hard to replace. There is a great article by Howard Kimeldorf, about the pre-CIO period that makes this exact point that workers won strikes when they were not easily replaced, and I think it applies today too.

But back to your question: historians tend to focus on bottom-up over top-down elements. For instance, everybody focuses on the UAW. When I was in graduate school and writing a dissertation, I thought I would study the UAW, the UE [United Electrical, Radio and Machine Workers of America], and the United Steelworkers, who were the three biggest CIO unions at the peak during the war. And I ended up doing just the UAW and the UE. But one thing that was so striking was that there was almost nothing written about the steelworkers relative to the others. Nobody was interested. Whereas there are libraries about the UAW.

Why? Because the United Steelworkers was a top-down union from the very beginning. It did not fit the New Left view of what the CIO was all about. That view was that the CIO unions were democratic and militant and then got crushed with McCarthyism. But in fact they were very mixed all along. In that regard, the CIO wasn’t as different from the AFL as many people assume.

David Brody has made a big deal about that, arguing that there was much more continuity between the AFL and the CIO than a lot of people appreciate. His point has been absorbed now into the conventional wisdom, but when he first wrote about it, there was this romance about the UAW and the communist-led unions too, of which the UE was the biggest.

I think because it’s in this sweet spot between the communist-led unions, which some people also were very enamored of, and the mainstream, more conservative unions. The UAW was led by leftists, some of them communists, some of them socialists, some of them Trotskyists. And because they’re all competing for domination, at least for a while before the [Walter] Reuther faction wins out, the UAW is more democratic internally. I think that’s the kind of thing that riveted people and made it seem like the most attractive organization among the array of CIO unions.

I remember this book by Irving Howe and B. J. Widick called The UAW and Walter Reuther, which is all about internal democracy and the organized factions within the union that kept each other honest. The UAW resisted, at least for a while, what Robert Michels, the Italian theorist, called the “iron law of oligarchy.” “Who says organization, says oligarchy” is the famous quote from Michels’s book. In other words, unions ossify into these top-down organizations, where the leaders have all the power, and they suppress dissent.

Again, the UAW, at least for a while in its early days, was different from that, precisely because there were these organized factions that forced each other to be accountable to the rank-and-file as well as to each other.

I think you would want to distinguish between building unions and sustaining unions. In my view, in the building phase it’s not that necessary. If you can extract union recognition because you’re John L. Lewis and you basically control the coal industry and you can threaten to disrupt the steel industry in a country where that’s the fundamental product that everything else depends on, maybe it doesn’t matter if you’ve got anything democratic behind you. On the other hand, when the employer goes after you later and tries to get rid of the union or weaken it, it really does help to have militant rank-and-file workers there defending it. So, that’s where it becomes more critical than in the initial winning of recognition.

Of course, you can win recognition in a lot of different ways, as we see today as well. Sometimes it is completely top-down, and it’s because somehow there’s leverage from the organization that makes it less costly to the employer to say yes than to keep fighting.

Factory occupations — what better way to disrupt production than to just take over the building? In the case of the Flint strike, it went on for a few months that they occupied those factories. That was not the only tactic, of course, but it was the most publicly visible one. And when they won, most observers, workers, and others attributed the victory to the sit-down strike. We could argue about whether that was really the key thing, but that was certainly the optic. That’s how it looked to everybody.

And there was this epidemic of sit-down strikes thereafter, all over the place, including retail stores in Detroit. There was a famous Woolworth’s strike where the clerks were sitting on the counter and occupying the store. It just inspired copycat organizing all over the place.

Why was it so inspiring? It’s a pretty appealing idea: if you’re a worker who’s treated terribly by your employer every day, you can just take over the whole enterprise and win rights by doing so. I think it’s as simple as that.

I’ll tell you a story about this guy called Clyde Summers, who’s dead now, but who was a very eminent labor law scholar who taught at University of Pennsylvania for many years. He came of age in the ’30s. I met him once at some conference, and he said to me, “You know, Ruth, in my day, labor law was a field for romantics. Now, it’s a field for masochists.” This was probably in the ’80s or early ’90s that we had this conversation, but it really stayed with me.

There was something transformative about the 1930s: the sit-down strikes, the Wagner Act, the New Deal itself, regulation, FDR, all of which emerged against the backdrop of the biggest crisis that capitalism has ever experienced, the Great Depression. This was a thrilling time for the people who participated in it, and I think that romance lives on among people like me who are interested in labor history. There’s something magical about it that we haven’t really seen since — or before, for that matter. It’s when all the stars lined up, which doesn’t happen too often in labor history.

Well, I think sometimes they’re able to seize the moral high ground and win the hearts and minds of the public on the basis of exposing employer abuse and stuff like that. I spent some time studying low-wage immigrant workers and their organizing efforts, which were pretty successful in the ’90s and ’00s. There it was symbolic leverage, some people call it, or the ability to expose abuse — that was key. It didn’t really have to cost employers very much to recognize the union, in fact, and it would benefit the workers greatly. Janitors, arguably among the less skilled occupations out there, were pretty successful in pulling that off. It was through what some people call “public dramas,” figuring out ways to capture media attention and therefore public attention by dramatizing the plight of a low-wage workforce.

After the 1950s, the AFL-CIO was a powerful force. “Big Labor,” everybody called it. You don’t hear that term much anymore. The idea was that unionized workers were pampered and over-privileged, that the unions had so much power that they had won all these great things. Arguably that was true up through the ’70s in some ways. But nobody could say that about janitors who were being paid less than the minimum wage. Nobody could make the “Big Labor” argument faced with the public dramas that Justice for Janitors and other groups like it pulled off. So that is another source of power for labor: moral power or symbolic power.

Well, in recent decades the CIO unions have suffered huge defeats, mostly because they have lost their leverage. In the industries that their efforts focused on, mostly manufacturing industries, the rug has been pulled out from under them by the employers, first in the form of outsourcing, of moving production away from where unions are strong, either within the US to the South or outside the country entirely. That totally undermines the whole logic of industrial unionism. There’s also new technology that has shrunken the workforce in those industries quite a bit over the years. So, the base of the old CIO unions is almost an endangered species.

Now, this may be changing, insofar as the current global conflicts are bringing manufacturing back to the US or making people aware that it’s good to have at least some of it here in the United States. That may create new opportunities for manufacturing unions to make some kind of rebound. I don’t think it would ever be on the scale of the ’30s and ’40s, but there’s something going on there, so we’ll see. I think it’s too early to make any big predictions about that.

There’s a whole new group of people who have a critique of capitalism, just as CIO organizers did in the 1930s, and a lot of them are increasingly interested in labor organizing. So, that is very similar to the 1930s. This is a huge resource for the labor movement.

But it’s going to be really difficult to win CIO-style 35 percent density in the United States, without a change in the labor law. So instead, we have a few highly unionized cities, places where labor has maintained a foothold. Then there are the skilled workers, in sectors like higher education and health care, and maybe tech workers will be part of that mix in the future. But I still think that without a major reform of labor law that undermines the ability of employers to just drag things out forever and defeat union organizing campaigns, it’s going to be difficult.