During the two periods of twentieth-century labor upsurge, collective bargaining was achieved, initially, through worker disruption. This direct action was followed by legislation giving workers the opportunity to vote in state-supervised Labor Board elections for bargaining rights. But it was the employers who demanded and eventually benefited from the establishment of an electoral road to union recognition: in time, union leaders came to rely less and less on members’ power and more and more on the law, and made steep compromises in order to retain the right to bargain. The real story of the past seventy-five years of labor’s journey is the successful subordination of unions. The union contract is a legal vise; the law that is supposedly the worker’s weapon is in fact a double-edged sword. When unions agree to long-term contracts of as much as six years, they are prohibited from striking for the length of the agreement. For the modest gains unions have made in legal guarantees, they have been obliged to surrender important rights. Labor law obliges a union to enforce its contract against illegal worker insurgencies, and in some states penalties are fairly stiff if union leaders sanction such actions. The law once helped unions to grow their membership, but after years of relentless and relatively successful right-wing attacks on workers’ rights, some unionists have come to realize that the Labor Relations laws at the federal and local levels are mostly rigged against labor. In the 1980s, AFL-CIO president Lane Kirkland proposed the repeal of the National Labor Relations Act, because it had become so watered-down, it no longer served workers’ interests. Yet this idea has not been introduced into the dominant vocabulary of union discourse; far from fighting to repeal the law, union leadership is reluctant to challenge or even strategically evade it. Labor remains committed to “reforming” it. But the sorry history of forty years of that effort attests to its futility, even in states where Democrats have had legislative majorities.