For Paul Buhle, as for many on the left, the central fact about the American labor movement is its persistent failure: to support "workers and peasants in the colonized regions of the world," represent women and minorities at home, and perform its predicted historic role of social transformation. In Taking Care of Business: Samuel Gompers, George Meany, Lane Kirkland, and the Tragedy of American Labor, Buhle aims to explain this failure: the triumph of "conservatives" over "radicals" in the labor movement, and the suppression of the rank and file’s spontaneous militancy by a centralized, self-serving, autocratic bureaucracy. Responsibility for all this, on Buhle’s account, lies at the feet of the labor movement’s leaders and staff.

The story begins in the 1880s, when the American Federation of Labor (AFL) was founded by cigarmaker Gompers, carpenter P. J. McGuire, and a few other craft-union leaders, and then quickly overshadowed the more inclusive and politically oriented Knights of Labor. Like many of his colleagues, the young Gompers read Marx and called himself a socialist. But as he aged and tightened his grip on the AFL, Gompers lost interest in politics and became increasingly conservative and autocratic. He began to identify with his businessmen compeers in the National Civic Federation, and favored the interests of a small elite of craft workers to those of the broad working class. As Buhle writes, the craft and racial exclusivity of the AFL "deepened ... as the AFL grew at once more rigidly bureaucratic and ... more successful" throughout the last years of Gompers’s reign.

Gompers, who died in 1924, was succeeded by Mineworker William Green, who served until his death in 1952. Next came George Meany, a product of New York’s ethnically based building-trades unions and Buhle’s preferred representative of all that is wrong with American labor. In two chapters that take up almost half the book, Buhle grimly tracks Meany’s growing "prestige, power and privilege: a chauffeured limousine, a healthy expense account, an official salary that reached six figures by 1977, two homes, and membership in exclusive clubs, where he maintained cordial relations with legislators, governors, cabinet members, presidents, bankers, and industrialists," not to mention his "friends at the CIA."

Buhle’s disappointment is magnified when he considers what might have been. At the AFL’s founding, the Knights of Labor offered an "egalitarian alternative" that, unlike the craft-based AFL, "affirmed a human solidarity based upon common toil." Next, in the first years of the twentieth century, came the Industrial Workers of the World. "In practice and vision," Buhle writes, "the IWW was everything that the AFL refused to be": it organized Chinese workers and others excluded from the AFL, proclaimed "the inevitability of class conflict," and scorned contracts and compromise with the bosses. Then, in the 1930s, workers across the United States "took fate into their own hands" by staging sit-down strikes and organizing industrial unions. In this telling, the Congress of Industrial Organizations (CIO) was, at least in its beginnings, a spontaneous uprising of industrial workers.

But the Knights and IWW dwindled and died, and a decade after the CIO’s creation it "increasingly functioned similarly [to the AFL], as Communist enforcement of wartime no-strike provisions strengthened the hand of the CIO leadership to limit the independence of local unionists." Then the CIO betrayed the radicals who helped to bring it to light by playing along with McCarthyism and expelling Communist-led unions from the organization. Long before the merger of the two federations in 1955, the CIO had ceased to offer a real alternative.

After the war, the picture grows even bleaker. Over the half-hearted protests of Walter Reuther and other old-line CIO leaders, the AFL-CIO essentially gave up on organizing new workers. "Why," George Meany asked, "should we worry about organizing?" In 1979, Meany handed over the reins to the hapless Lane Kirkland, by his own account "the oldest, established, floating heir apparent in history." The highest drama of Kirkland’s administration came at its end, when, in the first contested election in AFL history, Service Employees Union head John Sweeney beat out Kirkland’s designated successor, Thomas Donohue. Sweeney, who has served since 1995, is the fifth president since the founding of the AFL in 1890. As labor journalist Bob Fitch likes to point out, the papacy has changed hands more often.

Compared to most other recent labor histories, Buhle’s approach is decidedly top-down–it presents a story of "misleadership" rooted in bureaucratic organization. The tone is harsh–but the labor movement can learn even from its most unforgiving critics. Buhle’s salvos largely miss their mark, however, because he has inherited a set of categories–radical versus conservative, democracy versus bureaucracy–that aren’t very helpful in making sense of American labor.

Let’s go back to Gompers. It’s true that he called himself a socialist when he was young, and ceased to do so when he got older. But it would be a mistake to dismiss this as merely a youthful flirtation with radicalism. On the key question of the day–whether the main line of struggle in society was between owners and workers–Gompers was a Marxist, and remained one until he died. In the 1880s and 1890s, this question marked the division between the AFL and the populist-tinged Knights of Labor; the latter organization did not exclude small employers, self-employed artisans, and other "producers" from its ranks, while the AFL was from the beginning an organization of and for wage-earners. The wage- and benefit-centered "job consciousness" that defined the conservative AFL of the 1920s was class consciousness transposed into a different key; both were the perspective of wage-workers. Historians have often noted the "more proletarian" cast of the AFL’s leadership vis-a-vis the Knights. The difference was clear to contemporaries too. In 1890, in preparation for the convention at which he delivered his famous defense of "pure and simple" trade unionism, Gompers turned for advice to Friedrich Engels, who agreed that the AFL should be foremost a labor organization. While the Knights continued to see themselves as free, independent producers interacting as equals in the marketplace, the AFL–even if often limited to a relatively privileged group of craft workers–issued a genuine challenge to the values of nineteenth-century capitalism.

By conflating decentralization, democracy, and radicalism, Buhle doesn’t do much to illuminate this complex history. To take another example, many of the unions that still dominate the labor movement came into being in the two decades before World War I as a result of revolts of wage workers against craft organizations, which had been led by small employers and skilled, often self-employed artisans. Without exception, Gompers supported these rank-and-file rebellions, despite a formal prohibition against AFL interference in the affairs of affiliates. In fact, over the first two decades of the AFL’s existence, Gompers managed to overturn the leadership of at least a half-dozen major affiliates; in every case he replaced them with representatives of less skilled, often less ethnically privileged workers.

This little-known story (for which we can credit the sleuthing of labor historians Elizabeth and Kenneth Fones-Wolf) casts the early history of the AFL into a sharply different light. Yes, the AFL shifted towards craft and ethnic exclusivism in the first part of the century, as Buhle says; but the shift was not a product of success or bureaucracy. To the contrary, it was only around 1910–when the AFL’s efforts to organize new, often unskilled workers were stopped cold by an employers’ counteroffensive–that Gompers stopped interfering in the affairs of the affiliates. When the AFL stopped growing, he lost most of his leverage over the Federation’s constituent unions. That meant decentralization, but it didn’t mean democracy, and it certainly didn’t mean organizing the unorganized and the unskilled.

At the same time, the IWW–the lost cause to which Buhle unflatteringly compares the actually existing labor movement–refused to so much as sign contracts. Radical indeed–but it could equally be seen as hearkening back to the traditions of nineteenth-century craft workers. Until the 1870s, craft union practice was to collectively agree on a "rule" for wages and hours and unilaterally impose it, with no negotiation with employers and consequently no need for delegated authority. As historian David Montgomery puts it, "Union methods based on the unilateral adoption of rules … were peculiarly appropriate for the workmen who had progressed but partway down the path from journeyman artisan to factory wage labor…. For unskilled laborers, however, such practices were clearly impossible."

After all, unskilled workers neither formed a well-defined community that could meet and make decisions together in the absence of formal authority, nor did they exercise the local monopoly needed to make those decisions effective. The paid organizers who became one of the pillars of "business unionism" were not needed so much among craft workers, who, then as now, largely organized themselves. In the early years of the century, two-thirds of the AFL’s organizers labored among the unskilled, and the tedious work of signing and servicing contracts, which for Buhle is the "clearest manifestation" of bureaucracy, was an essential part of their work. And it’s significant that in his brief discussion of the CIO, Buhle focuses on relatively marginal industries in Texas and Minnesota, where industrial workers managed to organize themselves without a major commitment of resources from CIO leader John L. Lewis. But he gives short shrift to the far more important industries, like steel and autos, where a highly centralized organization allowed workers without strong organic ties to collectively challenge the country’s largest corporations.

None of this is to excuse the real failings of staff-heavy and autocratic unions, or to suggest that the AFL was ever a model of labor democracy. But it’s not clear that Buhle’s focus on spontaneity and authenticity–his desire for a labor movement where, to quote Staughton Lynd, "We’re all leaders"–suggests a realistic alternative. The truth is that few American unions have suffered from centralized bureaucracy–if anything, they’ve been weakened by a lack of it. In many unions, promotion is based on personal ties, positions are doled out as rewards for loyalty rather than by any objective criteria, and power, far from flowing from the top down, is dispersed among thousands of officials with their own bases and constituencies. The "top" is largely irrelevant to day to day activities. Unlike union federations in Northern Europe, the AFL-CIO plays no role in contract negotiation. Nor, as of now, does it have any say in organizing priorities. Nor do most of the Internationals play such roles. In most American unions, every decision that matters is taken at the local level, where leaders often serve for life.

Some other labor writers are critical of the existing labor movement from a left or radical-democratic perspective, but nevertheless take seriously the need for organizations, and for processes to confer and delegate authority. This outlook informs the arguments of Mike Parker and Martha Gruelle, two veteran labor activists and writers who have 35 years of experience in union reform movements like Teamsters for a Democratic Union. Like Buhle, Parker and Gruelle see the central challenge for the labor movement as returning power to the rank and file. But unlike Buhle, they understand that organizations matter when it comes to strategic planning (which is very difficult on the local level) and, most important, accountability. And though organizations "are not the substance of democracy," they are its prerequisite. Parker and Gruelle know that categorical claims that bureaucracies must be self-serving can be a "powerful weapon for those who want to keep people cynical and disorganized." And they know that people don’t become members of a community by birth or virtue of shared "objective" interests, but only through their own conscious activity–that, in Lizabeth Cohen’s eloquent phrase, a rank and file is something that workers must become.

The new book from Parker and Gruelle, Democracy is Power, is a handbook for reformers hoping to democratize unions from within. Democracy, they point out, isn’t simply a quality that some unions possess and others don’t: it’s a process, something that happens (when it does) through the habits and practices that allow people to see and act on their common interests. Union democracy typically comes down to some seemingly minor details: whether contract votes, say, take place in meetings where members can debate the pros and cons before voting or through mail ballots, which "keep members dispersed and thinking purely in terms of individual gain or loss"; whether contracts negotiated with regular reports to members, or in closed sessions with management; whether workers file their own grievances or they have to depend on shop stewards. They argue that democracy empowers workers, and empowered workers are good for unions–simply put, the most democratic unions are the most effective ones–and hope to spread that insight.

Whether they succeed may help determine whether the story of the American labor movement is a tragedy, after all.