They exploit the employés to the utmost to compensate themselves for the exactions of the manufacturers and the competition among themselves…. It is the judgment of the employes in the trade, and most emphatically my judgment, that any measure which does not prohibit the manufacture of clothing in any dwelling by any woman or child will wholly fail of its object.” Social reformer Florence Kelley’s sharp denunciation of the sweatshop system in the tenements of Chicago perfectly illustrated the social meaning of “employé” at the end of the nineteenth century. No worker, indeed, could have been further removed from the modern organization of wage work taking shape in the rising corporations than those whose toil she brought to light in her report to the Illinois legislature—an “employé” was still simply someone who was “used.”1
Kelley was a tireless advocate of the regulation of the economy through the intervention of the state, and her report, which was soon to be translated into a law regulating the work of women and prohibiting the employment of children, illustrated the dynamics behind the rise of the labor question in America. Earlier in the nineteenth century, many middle-class Americans, like their European counterparts, saw in vagrants and paupers dangerous unincorporated social figures. By the end of the century, the rapid industrialization of American society and the centrifugal forces it generated had changed the terms of the debate on the social body. Workers, although they were not idle, had come to embody the problem of social disaffiliation, and wage work was the object of renewed attention. The labor question—a mixture of class antagonism, worker poverty, and social strife—encapsulated both a middle-class feeling of alienation from society and definite programmatic responses moving beyond the orthodoxy of acquisitive liberalism.2
An upheaval in social values and norms was at hand in many Western countries, one that came largely, as in Kelley’s report, from the emergence of realism and its ultimate product—the notion of social fact. From Mary Van Horst’s experience as a factory girl and E. C. Moore’s studies of working-class saloons to Robert Hunter’s analysis of the dynamics of poverty in the United States, a large number of reform-minded scholars now tried to use sociological and statistical techniques to move beyond Victorian moral strictures and abstract free-market principles. Probing the ways in which American society and its economy produced social inequality and economic coercion, they laid the groundwork for a reappraisal of the meaning of wage work.
The input of the social sciences thus simultaneously revived and transformed the old ideal of social harmony. Indeed, these reformers did not wish to enter the class struggle on the same terms as the Industrial Workers of the World or European socialist parties. They saw class antagonism as a pathology that could be remedied and dealt with. With the help of Progressive political leaders convinced that the pursuit of private interests could no longer be expected to produce a social good, they worked to redefine employment and wage work as a source of obligations for employers and of social rights for workers. This was a long and sustained effort that consumed the energies of advocates of social reform well beyond the chronological limits through which the “Progressive Era” is usually understood. Only in the late 1930s, when the Federal Labor Standards Act was adopted, did this movement lose its impetus. By then, “employee,” the legal definition of the worker, had become the mainstay of a new and capacious vision of economic citizenship.3
The Wagner Act, which was adopted in 1935, was no doubt the keystone of the legislative edifice that Progressives and New Dealers built to remedy the patterns of coercion and subordination that they identified at the root of the labor question. By democratizing the workplace and allowing an upward shift in working-class purchasing power, the law came to symbolize the politics of economic security that characterized the reconstruction of the wage relationship. Coming as it did after decades of legal injunctions and battles with employers, the Wagner Act seemed to many workers to be nothing short of a revolution. It was, in William Green’s words, “Labor’s Magna Charta.”
This, however, was a somewhat restrictive vision of “labor,” for the adoption of the Wagner Act also gave rise to a debate over the contours of the legal definition of the worker. While many workers enjoyed the benefits of unionism, others were now engaged in a struggle for inclusion and for recognition as “employees.” Indeed, the benefits of the law were not universal, and opposition to the democratization of the workplace now developed along a participatory axis, with employers seeking to define workers out of the group of “employees.” Tracing the political and constitutional genealogy of the Wagner Act back to its Progressive roots, this chapter explains why at the end of the New Deal, the legal definition of workers had become a contested battleground on which employers and unions now waged their struggle over the limits democracy at work.
“The question that forces itself upon us, and imperatively demands an immediate answer, is this: in the great strife of classes, in the life and death struggle that is rending society and its foundations, where do I belong?” Florence Kelley asked in 1887.4 Progressives emphatically responded that they stood on the side of those who were victims of the industrial capitalism, and yet most of them did not deal in the rhetoric of class, preferring to emphasize the pursuit of the public interest, which they identified with the regulation of the employment relationship rather than its abolition.5 As Jane Addams explained in an article published in the American Journal of Sociology in 1899, social reform should not be effected on behalf of one particular social class, but rather with a view toward fostering social cohesion. “The habitual use of ‘the people’ as a phrase practically equivalent to the ‘working classes,’” she explained, “is a constant admission of the fact that the proletariat is not, properly speaking, a ‘class’ at all, but the body of society itself.”6
Indeed, Addams argued that the labor question and the social strife that attended it were really a failure of the American social body, which had grown so distended as to allow for structural inequality and for the existence of economic social classes. Like organized labor and socialist politicians, Progressives like Addams understood that social change was possible on a relatively short term. Yet the task of resolving this question was not to be left to labor unions alone. Social amelioration should not be primarily a working-class movement, for in promoting their interests, unions, like corporate America, might be oblivious to the common good. For Progressives, an objective and positive observation of social facts, not class consciousness, should form the basis of social policies. As Addams explained, “The trades-union movement secures its lower objects best where there is a well-defined class feeling among the proletarians of its country, but it accomplishes its highest objects in proportion as it is able to break into all classes and seize upon legislative enactment.”7
Those Progressives who took interest in the labor question can be seen as pioneers of a “new liberalism” that also existed in England and thrived in a common ground of shared interests with the left. Progressives did assist workers in their struggles, whether it was Florence Kelley in Lawrence in 1912 or Lincoln Steffens and Walter Lippmann joining Bill Haywood at the Paterson pageant organized by John Reed in 1912 at Madison Square Garden, but this support can be misleading. They were legatees of a liberal world whose values they found jarring and hollow. Yet while they often interacted and rubbed shoulders with socialists and unionists, Progressive reformers were not members of the Socialist Party and did not want to live under working-class rule. They saw the empowerment of the working class and the promotion of its welfare as a means toward social cohesion, a stepping stone to a different social order. The interests of the working class and those of the “public” were so enmeshed as to be sometimes undistinguishable in practice, but in theory there was a great difference—Progressives championed the interests of the working class, but they opposed class politics all the same. This led them to deemphasize the socialist ambition to empower the working class and to seek instead the regulation of employment and the protection of “employees” by weighing on the incidents of the workers’ relationship with employers.8
Notably, many reformers in the Progressive Era were steeped in the language of pragmatism. According to men like John Dewey and William James, truth was essentially contingent; it was the product of social inquiry and experience. In this perspective, there was no such thing as an abstract, coherent working class; rather, there were a diversity of workers and a multiplicity of interests. True to this pragmatic approach, in their efforts at social amelioration, reformers did not tackle the labor question in a broad, structural way, but rather aimed at one specific social ill they could attempt to remedy, usually by marshaling the vocabulary of science and efficiency. This often involved building local coalitions of lawyers, journalists, and working-class advocates to accommodate the demands of party machines. However, it also led reformers to focus their reformist efforts on specific groups of workers rather than on classwide reform.
The adoption of the New York Bakeshop Act in 1895—the law that generated the Lochner v. New York case—is illustrative of this early type of “issue politics,” to use historian Daniel Rodgers’s apt phrase.9 This law resulted from the investigations of the muckraker journalist Edward Marshall into the squalid bakeries housed in New York tenements. With the help of Henry Weisman, the secretary of the local Journeymen Bakers Union as well as several “social gospel” groups, Edward Marshall was able to rouse public attention by shedding light on the disastrous consequences of long hours (it was not uncommon for bakers to work more than one hundred hours a week) spent in an unsanitary environment where dampness, gas fumes, and flour dust generated lung diseases. Tellingly, the Bakeshop Act mostly targeted master bakers or “boss bakers,” whose small establishments employed fewer than five workers—very often French, Italian, or Jewish—and did not bring large profits. Larger, more modern bakeries supported the law, hoping that their English, Irish, and German workforces would no longer suffer from the competition of the tenement bakeries.10
With an editorial published in the New York Press, Marshall was able to launch a local campaign, enlist a large number of urban reformers, and secure the support of the local union of bakers, who had grown frustrated with their unsuccessful attempts to win the ten-hour day through collective bargaining. In a state dominated by the Republican machine of boss Thomas Platt, this campaign ended with a unanimous vote for the following law:
No employee shall be required or permitted to work in a biscuit, bread, or cake bakery or confectionary establishment for more than sixty hours in any one week, or more than ten hours in any one day, unless for the purpose of making a shorter work day on the last day of the week; nor more hours in any one week than will make an average of ten hours per day for the number of days during such week in which such employee shall work.11
The New York Bakeshop Act thus helps us lay bare a number of important elements in the way Progressive quest for harmony reshaped the social meaning of wage work and displaced the idea of a political working class. First, through this law, the New York legislature obviously recognized that there were elements of compulsion in the wage relationship, thus changing the social and cultural meaning of “employee.” Although the law was based on a report highlighting the unhealthy sanitary conditions prevailing in the baking industry, it covered only those who were deemed to lack the freedom to protect themselves from the effects of these conditions. “Our aim is primarily to protect the employee,” explained Henry Weissman after the bill had been amended to exclude the employers, who thus remained free to work as much as they wanted in spite of the health hazards.12
Still, the recognition of the lack of freedom of the wage worker came with a drawback insomuch as it did not entail a full recognition of the asymmetry and inequity of the wage bargain that radicals denounced. First, this recognition was limited to specific groups of workers who could be linked to a social problem—whether it was bakers in the New York law or women in the famous case of Muller v. Oregon in 1908 (where the Supreme Court sanctioned a law mandating shorter hours for women), the group of “employees” covered by the Progressive laws was not universal, but rather determined through a specific social purpose, whether it was the regulation of a specific industry (railroads, bakeries, restaurants) or the protection of women and children workers because of preconceived ideas as to their role and place in society. This fragmentation of workers into specific groups of employees was reinforced by the character of the American polity, which made it difficult to contemplate national social laws at first, but also by the Progressive movement itself—a local impulse that only gradually developed a national dimension.13
Second, in protecting some workers from the overweening power of their employers, such laws also, sub silentio, legitimized and objectivized the workers’ lack of discretion on the job and their submission to managerial rule. As the social insurance expert Charles Richmond Henderson explained, the worker “has special claims upon collective consideration because he no longer has any ownership in the materials and instruments of production, nor any voice in management of the process nor control of the conditions under which his mind and body may suffer.” As the legal scholar Jonathan Witt has suggested, some Progressive laws such as workmen’s compensation even strengthened managerial rule by leading employers to increase safety procedures and reinforce their control of the workplace to avoid civil penalties. The protection of the “employee” was thus thoroughly congruent with Taylorism. In fact, Progressive laws strengthened the inequity of the wage bargain by making it the source of a number of social rights that were trade-offs for the worker’s social and technical submission to the new industrial order. In this respect, the path trod by American Progressives was quite similar to the one taken by reformers in European countries such as France, which provides a useful point of comparison.14
In France, it was the solidaristes led by Léon Bourgeois who left their imprint on the legal definition of the worker during the third Republic. Solidarisme was in many ways the political translation of the sociological analysis pioneered at the turn of the century by Emile Durkheim, who argued in De la division du travail social that social interdependence was both inevitable and necessary, and needed to be founded on social justice. As a result, returning modern societies to harmony meant fostering the social bonds that had weakened with industrialization. Law was most important in this respect, because it could engineer social solidarity and integration. Seeking to act upon this idea, solidaristes conceived of society as built on social debts and credits. In La solidarité, Bourgeois explained that every individual is indebted to society in that he or she benefits from the achievements and decisions made by former generations.15 But society was equally indebted to individuals, for its role was to make sure that their expectations for social justice were met. The ideal of solidarité packed a new social contract that justified the adoption of social laws to promote the security of the working class without accepting socialist collectivism.16
Indeed, although they sought to address the same social issues and sometimes collaborated, socialists and solidaristes had different approaches to class politics. Opposing the logic of class, solidaristes marshaled the concept of “society” to push for social laws that would “restore” equality between employers and employees. In their view, the state should now act as an agent of “reparative justice,” for it had become impossible to maintain that employer and employee were always two free and equal individuals.17 Accordingly, solidaristes pushed for a number of legislative reforms such as progressive taxation, but the crux of the rearticulation of social relations that their political and social outlook engineered was a new legal definition of wage work and of “employé” that incorporated the concept of subordination, namely, the contrat de travail or contrat salarial. First used in a 1901 law, the concept of contrat de travail (employment contract) stemmed from the implementation of an 1898 law that enacted the liability of businessmen in all work-related accidents of workers. As in the case of lien laws in the United States, the implementation of this legislation raised questions about the sociological limits of the group of potential beneficiaries—what, indeed, constituted an ouvrier or an employé? As defined in the contrat de travail, an employé or—to use the term that would be more commonly used—a salarié (wage worker) was someone who submitted to the employer, who ceded the employer the right to direct the work that was to be done.
Importantly, this idea of subordination included in the law went beyond affirming the authority of the manager over the worker. This was no doubt one of the goals sought by the proponents of this legal reform, who sought to adapt labor law to the needs of the growing corporations. Yet the contrat de travail was also premised on the idea that there were elements of social and economic compulsion in the wage relationship that needed to be acknowledged. Therefore this contract also enacted the dissymmetry inherent in the wage relationship—the employer and the employee were not equal at the moment when they concluded the contract, and the contract involved more than labor power—it constructed the worker as a person dependent on the employer.18
Language bore the mark of the defeat of the ideal of working-class emancipation. A term singled out by Proudhon for its negative social implications and rejected by workers throughout nineteenth century, salariat (wage work) gradually lost its meaning as wage slavery. Rather, it was now a legal category spelling a number of social rights stemming from the legal definition of subordination included in the contrat de travail. Having renounced his freedom to subject himself to a master or a manager, the worker (salarié) was to regain his or her social identity through social laws providing a specific and well-defined social status. Indeed, it was the recognition of economic subordination that made it possible to adopt laws to progressively emancipateworkers: the ten-hour day was adopted in 1904, in 1906 work was prohibited on Sundays, in 1919 the working day was shortened to eight hours, and the first law on collective bargaining was passed the same year. Significantly, by 1910 labor law—which covered all those who were seen as “socially dependent”— had been separated from civil law.
The evolution of labor law in France and the United States thus proceeded from a similar ideological impulse against social fragmentation and class identification. There was, however, a marked difference. In France, no doubt reflecting the centralization of the country, the new legal definition of the worker was a universal one. By contrast, in the United States the structure of the polity—particularly federalism and separation of powers—made the general recognition of the socially subordinated character of labor extremely difficult to achieve. Indeed, it was in the course of a dialogue between labor reformers and the courts that the American “employee” assumed a distinctive, limited character—it was built on the foundations of free labor, and it resulted in labor laws protecting some workers while leaving others out of its safety net.19
In 1895, the Supreme Court handed down its decision in Ritchie v. People, the case bearing on the constitutionality of the eight-hour law that Florence Kelley, Jane Addams, and the Chicago Federation of Labor had successfully pushed through in 1893. The decision came as big disappointment to the “eight-hour club” that now met weekly at Hull House to discuss labor legislation. Taking its cues from the brief written by the Illinois Manufacturing Association, the Court struck down the eight-hour section on the grounds that it violated the right to make lawful contracts, a substantive property right protected by the Fourteenth Amendment.20 Underlying this decision was the reigning public/private legal distinction—the idea that through their police powers, states could legislate private behavior and activity only when such legislation was clothed within a public interest. The Court found none of it in the regulation of women’s employment.21
The idea that signing contracts was a substantive property right was of recent coinage, but it was not long in finding a national translation, particularly in Lochner v. New York, a case that—in the Progressive milieu—came to symbolize the judiciary’s determination to reify an abstract, universal legal norm in the face of pressing social needs.22 In 1902, Joseph Lochner, a German immigrant from Bavaria and a master baker in Utica, was found guilty of violating the first provision of the New York Bakeshop Act and was fined $50. Notably, Lochner stood in adamant opposition to the union and had already been convicted for violating the statute a few years earlier. Upon his second conviction, Lochner decided to ask the State Association of Master Bakers to help him challenge the law in court. The request came at a propitious time, for during its 1902 convention the association had decided to find a test case to challenge the law, as it believed that it was used as a bludgeon by unions to attack nonunion bakeries. Defending their legal action in the Bakers’ Review, the association reaffirmed the voluntary, free labor roots of the employer-employee relationship: “[The masters] cannot see why the matters of settling hours of work should not be left to employers and employees in the baking trade as it has been settled in any other trade.”23
The Court was receptive to this argument. In the opinion it handed down on April 17, 1905, the majority vindicated Joseph Lochner’s claim and put forward an abstract defense of “individuals, both employer and employee, to make contracts regarding labor in such terms as they think best, or which they may agree upon with the other parties to such contracts.” Tellingly, the Court argued that the New York legislature had actually impaired sui juris individuals in their ability to make decisions “with regard to their means of livelihood.” In its brief, the state of New York had made the argument that the law was necessary because the baking industry mostly comprised immigrants who needed this kind of regulation, but the majority opinion denied the social and economic dynamics structuring the employment relation, and instead reaffirmed the principle that the employer and the employee were strictly on equal footing: each individual was to make his or her own decisions free from interference from the state.24
This principle was sharply criticized by the proponents of legal realism, particularly Roscoe Pound and Oliver W. Holmes, for whom this liberty of contract was nothing more than a social and legal construct. Holmes famously took the majority to task for attempting to read Spencer’s Social Statics into the pithy wording of the Fourteenth Amendment. The public/ private distinction, Holmes suggested, was nothing less than a formal legal reasoning cloaking a political one. As he had explained in an earlier opinion, the Constitution embodied only “relatively fundamental rules of right, as understood generally by all English speaking communities.”25 Beyond their duty to protect this kernel of Anglo-Saxon liberties, the courts were to defer to the legislative process and the definition of the public good that it produced.
Following Holmes, many reformers doubted whether a substantive freedom of contract really constituted a natural right, and were quick to point out that the courts had seized on this concept precisely at the moment when there was an increase in available evidence showing that the freedom of workers was restricted in many ways. In an article lamenting the Court’s “mechanical” reasoning, legal scholar Roscoe Pound argued that the holding was not adequate to modern industrial conditions. “Why is the legal conception of the relation between employer and employee so at variance with the common knowledge of mankind?” he asked, citing the sociological work of Richard T. Ely and Carroll D. Wright.26
Yet the Court took its defense of an abstract, universal employment relationship one step further in 1908 in Adair v. United States, when it invalidated the Erdman Act, a federal law prohibiting yellow dog contracts—contracts whereby workers pledged not to join a union—in the railroad industry. Citing the work of the Jurist Thomas Cooley, who had penned influential defenses of liberty of contract denying that the interests of employers and workers might ever be at variance, Justice Harlan—himself the author of a famous dissent in Plessy v. Ferguson—reaffirmed the sanctity of the principle of employment at will:
The right of the employee to quit the service of the employer, for whatever reason, is the same as the right of the employer for whatever reason to dispense with the services of the employee…. In all such particulars, the employer and the employee have equality of right, and any legislation that disturbs that equality is an arbitrary interference with the liberty of contract that no government can justify in a free land.27
Interpreting this determined defense of laissez-faire and liberty of contract has been no easy task. Progressive historians have long argued that in the Gilded Age judges indulged in their personal faith in laissez-faire rather than legal interpretation. Moreover, progressive historiography insists that decisions such as Lochner exemplify judicial activism, one that stemmed from the judge’s misgivings about the democratic process. Rufus Peckham, the author of the majority opinion in Lochner, was a conservative Democrat and a former corporate attorney who believed, as he noted in his opinion, that many health statutes were passed for “other political motives” rather than the workers’ health. Similarly, justices such as David Brewer spoke for a large number of property-minded conservatives when he declared in 1893 that the role of the judiciary was to protect the individuals and their rights from the majoritarian process, “the danger is from the multitudes—the majority with whom is the power.” From the 1895 Pollockcase—in which the Court had ruled the federal income tax unconstitutional—to the 1897 Debs decision, the Court had evinced a decided interest in negating the very nature of social interactions and conflicts that now ran through American society, refusing to conceive of social relations in any other terms than individualist ones. In point of fact, in Pollock, Justice Field had expressly stated the Court’s determination to not to let the political process drift into an all-out struggle between the rich and the poor.28
More recently, scholars have argued that such judges did not simply read their own political views into the Constitution. Their decisions also made sense within the framework of Jacksonian and Republican principles, which implied a tradition of equal rights and an opposition to monopolies that led judges to oppose “class legislation.” Historians have also pointed out that the Court actually supported more pieces of protective legislation than it actually struck down. In this perspective, the Court’s preindustrial faith in the purposive equality and unity of the “employer” and the “employee” did not prevent it from sanctioning legislation that seemed to serve a legitimate social purpose.29
We will dwell with such laws and cases in just a moment, but before we do so, we need to note that by the early twentieth century, whether it stemmed from an antidemocratic impulse or from a nostalgia for a bygone, preindustrial era, liberty of contract was indeed legal fiction for most workers, particularly those whom protective legislation targeted. The effect of the jurisprudence on freedom of contract was to disconnect the purchase and selling of labor from the worker’s social position. What the Court actually secured is a status quo keeping the employee in a state of subjugation, a reality it halfheartedly acknowledged in 1915 in its Coppage decision as it struck down yet another law prohibiting yellow-dog contracts: “It is from the nature of things impossible to uphold freedom of contract and the right to property without at the same time recognizing those inequalities of fortune that are the necessary result of the exercise of those rights,” the Court tersely concluded.30 The phrase must have rung a bell to reformers, for it was used in the Civil Rights Cases, in which the Court held that the 1875 Civil Rights Act was incompatible with the public/private dichotomy. Like the phrase “separate but equal,” the definition of the employment relationship as one involving two equal parties thoroughly concealed one essential feature of modern societies, in which an individual’s liberty was exercised within a tangled web of relations of power. In this respect, the triad Lochner, Adair, and Coppage can be seen as an industrial Plessy v. Ferguson.
Most important, although these rulings were subsequently overruled—and although substantive due process was abandoned as a standard for reviewing the constitutionality of economic regulation—there was no general statutory reform of the employment relationship. Indeed, the drift and logic of Progressive reform was to carve exceptions to the at-will common-law rule and its underlying principle of equality rather than fully discard it and give all workers the same economic rights.31
To understand the full significance of Adair and Lochner for the history of the definition of the worker in labor relations, it is the dissenting opinion penned by John Marshall Harlan for himself and two other justices that should command most of our attention. For while he stood by the precept that the “employer” and the “employee” were social equals, Harlan believed that laws protecting “employees” might nevertheless pass constitutional muster. In contradistinction to Holmes, Harlan agreed with the majority that the Bakeshop Act may have stemmed from the ill-conceived notion that “the employer and the employee do not stand on equal footing in such establishments”—an agreement largely congruent with his opinion in Adair. Harlan departed from the majority opinion, however, when he contended that the Court had passed too quickly on the possibility that the hours law might constitute a valid exercise of police power in the public interest. Using the public-private distinction, he suggested, required the Court to probe all the available economic and sociological evidence to determine the relationship of a law to the public interest. In a kind of proto-Brandeis brief, the “Great Dissenter” culled evidence from a number of studies produced in the United States and in Europe since the early eighteenth-century plague to demonstrate that the Bakeshop Act was no oddity in the larger international movement to protect the health of workers and of those dependent on them. As such, it constituted the legitimate expression of the belief held in New York that sixty hours a week was the limit of reasonable exertion in bakeries, which may not be the case in other industries.32
Harlan’s dissent bears rereading because of its implications. What he suggested was that it was really possible to adopt social legislation and still retain the nineteenth-century notion that the employer and the employee were juridical and social equals. Under Harlan, the common-law definition of employers and employees as free contractors would remain the default one, with legislatures intervening in this equal relation only to serve the public interest, but without fundamentally affecting the theoretical and legal definition of the worker as a free seller of labor. Far from becoming a new anthropological norm, the recognition of the subordination of the worker would be conditional and would be linked to the pursuit of a specific social and economic end. The definition of employee would be not political but thoroughly instrumental, and it has remained so to this day.
Indeed, it is possible to discern a jurisprudential undertow that, through the public-private distinction, allowed American states to remedy what they saw as the worst evils arising out of the industrial revolution. Thus, in the 1898 Holden v. Hardycase, the Court had sustained a law limiting to eight hours a day the work of miners in Utah. Admitting that “law is, to a certain extent, a progressive science,” the Court drew an important distinction between the fact that the miners were “full of age, and competent to contract,” and the “state’s power to interfere, where the parties do not stand upon an equality, or where public health demands that one party to the contract shall be protected against himself.” In view of the dangerous working conditions of miners, the Court held, the state of Utah could legitimately intervene and regulate the miners’ work. Moreover, the Court noted that the claim that workers should enjoy an unfettered freedom of contract was made not by miners, but rather by mine owners whose interest was in long hours. “The argument would come with better grace and greater cogency if it came from the other class,” the Court concluded.33
This does not mean that the Court admitted that liberty of contract was mere legal fiction. Rather, Holden v. Hardywas important because it implied that the courts were willing, in specific cases, to admit exceptions to the theory that freedom of contract was a model of social interaction that would go of itself. “Freedom of contract is a qualified, not an absolute right,” the Court proceeded to explain in 1911 as it passed upon the validity of an Iowa law that made companies liable for their workers’ injuries even when the latter had signed contracts that said otherwise. At issue in this case was a railroad company relief department which awarded benefits to injured workers provided that they had signed contracts in which they waived the right to sue the company for damages incurred. “There is no absolute freedom to contract as one chooses. Liberty implies the absence of arbitrary restraint, not immunity from reasonable regulations,” the judges concluded.34 In the meantime, the Court had handed down the Muller v. Oregon case, in which the famous Brandeis brief opened the door to regulation of hours and working conditions of women. Notably, the court followed this jurisprudential vein in 1917, when it once more sustained the state of Oregon, this time sanctioning a law that limited the working hours of “employees”—not just women—in mills, factories, and other manufacturing establishments, and required that overtime be paid to those requested to work over the limit.35
Still, the limitations inherent in this reconstruction of the wage relationship could be quite significant, for the question of who counted as an “employee” was essential to the making of Progressive laws. Paradoxically, it seems that it was employers who first challenged the restrictive sociological purview of Progressive laws on equal protection grounds. In 1915, in the case Miller v. Wilson, it was the owner of a Riverside hotel arrested for asking a chambermaid to work nine hours a day who asked the Supreme Court to invalidate the Women’s Eight Hour Law adopted in California in 1911. According to the plaintiff, the law was not simply an unconstitutional violation of freedom of contract; it also needed to be struck down because it violated the Fourteenth Amendment equal protection clause. Indeed, the law omitted from its protective ambit large numbers of women, excluding from its definition of “employee” (1) women employed in “harvesting, curing, canning, or drying of any variety of perishable fruit or vegetable,” (2) women employed in “boarding houses, lodging houses, etc.,” and (3) “stenographers, clerks, and assistants employed by the professional classes, and domestic servants.”36 Earlier on, a similar argument had been made by a businessman opposed to a law regulating the working time of women in Ohio. Yet in both cases, the Court rejected the argument, holding that laws protecting only certain groups of workers were perfectly legitimate.37
A quick look into the personal journey of Charles Evans Hughes, who drafted the opinion of the court in the Miller case, offers interesting insights both into the logic of the Court’s response and into its implications. In 1915, Hughes was reaching the end of what legal historian James Henretta has termed the “19th century transition from Mugwump to Progressivism.”38 Raised according to strict Calvinist principles, Hughes was a hard worker who shared the American bourgeoisie’s concern over democracy and universal suffrage. A Wall Street lawyer, he became a local popular hero in 1905 when he contributed to a high-profile investigation of the corrupt business of the public utilities industry in New York, before turning his attention to the rule of law to conduct an equally devastating investigation of the life insurance business. In the 1906 gubernatorial race, he ran a Progressive campaign against the radical platform of Hearst, condemning “legislation for classes … working classes or any other classes”—words reminiscent of the political outlook of the likes of Jane Addams and many other middle-class reformers. As governor, Hughes proceeded with his struggle against corruption, but he also supported the National Consumers League’s efforts to impose minimum wages on department stores and regulate child labor. Meanwhile, he developed ties with the American Association for Labor Legislation and, using the current distinction between ordinary and hazardous labor, supported barring the employment of young men in some occupations. As James Henretta notes, by the end of his mandate Hughes had moved so close to the principles of the new liberalism as to acknowledge, like Roscoe Pound and sociologists like Arthur Ross and Richard Ely, that society was an organic whole. As a result, once on the Court Hughes was prepared to assert the protective power of states and uphold regulation in the public interest. “The opportunities for Labor (should be) protected and enlarged by state action,” he explained.39
Yet there was a significant difference between this quest for social cohesion and the recognition of the subordinated social and economic position of wage workers. Hughes was willing to admit that in some cases workers and employers did not stand on equal footing (he did so in the case of an Iowa law that nullified contracts limiting workers’ rights), but he really believed that the recognition of this inequality was necessarily contingent on specific industries and well-defined groups of workers. It was a matter of legislative power, not class. As a result, he explained in Miller v. Wilson, legislatures were not expected to couch their laws in “all embracing terms,” “extending the regulation to all cases where it might potentially reach.” Rather, he said, they were “free to recognize degrees of harm,” and may confine regulation to “those cases where the need is the clearest.”40
Such were the origins of the contemporary debate over the meaning of “employee.” Erected on the legal basis of the public-private distinction, the definition of the worker in Progressive laws was a far cry from the Marxian collective worker, a political figure bound by economic pressure to attempt to emancipate himself. Used as it was by reformers and Progressive judges to break down the labor question into a large number of specific socioeconomic problems arising within the framework of the capitalist system, this definition conveyed a vision of the working class that bore two characteristics: it was sociologically fragmented and sought to combine the individualism inherent in freedom of contract with the need to expand the police power of the states. In many ways, the restrictive, contingent character of the legal definition of the worker illustrated the Progressive search for order, a limitation that would prove hard to transcend.41
Indeed, one may argue that by the end of the Progressive Era, most reformers had grown accustomed to thinking about social reform in those restrictive terms, even if talk about “industrial democracy” carried considerable universal appeal.42 To be sure, in the coming two decades the Great Depression would give labor reformers the opportunity to think about American workers on a much broader scale, pushing for an ideal of “economic security” covering all American workers. In his famous Commonwealth Speech in San Francisco in 1932, Roosevelt called for a “new constitutional order,” one that would protect a number of economic rights as well as political ones. No doubt, as a number of scholars have noted, the Wagner Act was a pillar of the attempt to protect a citizenship writ large.43 Yet as they built the right to organize, reformers and labor progressives continued to work within the conceptual regulatory framework that they had established during the Progressive Era, building the state’s visible hand on labor relations by adopting a law that protected some workers while it excluded others.44
Historians of the New Deal often give pride of place to the strikes and movements that swept the country from the mining districts of Pennsylvania to the Port of San Francisco. Such movements are significant because they reveal the role played by workers in the process that led to the adoption of the law recognizing their right to organize. As in France, where working-class mobilization forced obdurate employers to bargain with the leaders of the sit-down strikes in 1936, American society gained in 1933–1934 what the historian Melvyn Dubofsky has called a “practical education in industrial warfare.” Without this mobilization, it is quite unlikely that laws such as the Wagner Act would ever have been adopted.45
Tracing the legal and constitutional logic of the law, however, requires that we take a different approach, one that can tell us what reformers had in mind when they responded to social pressure. When the Great Depression set in, the 1925 Railway Labor Act—the first federal law protecting the right to organize—figured prominently in the minds of labor liberals because of a two-year struggle between the Brotherhood of Railways and Steamship Clerks and the Texas & New Orleans Company.46 In May 1927, the union officials requested a wage increase that the company, although it had traditionally been open to negotiations, firmly refused. When the union petitioned the U.S. Mediation Board, the tripartite agency created by the Railway Labor Act to solve such matters, the company hastily set up a company union with the intent of bypassing the American Railroad Union. Determined to enjoy the benefits of its statutory rights, the union decided to sue the company and, in an irony of history, secured an injunction against it for violation of the Railway Labor Act; the federal court ordered the company to disband the company union.47
By 1930, the stakes were high indeed as the Supreme Court prepared to pass on the constitutionality of the law. Before the Court, the company argued that the Railway Labor Act was unconstitutional on two grounds. First, the law protected an “abstract right” (i.e., the right to organize) with no real legal foundations. Second, because the law prohibited employer interference in the selection of the railroad workers’ delegates, it violated the property rights and the freedom of speech protected by the First and Fifth Amendments. The company therefore contended that the Adair decision should remain controlling—employers and employees being equal in the railroad industry as in others, no departure from free contractual relations was constitutionally permissible.48
The Supreme Court firmly rejected the company’s argument, and once again Charles Evans Hughes was the author of the majority opinion. Relying on a 1921 case in which the Court had ruled that picketing was legal—American Steel Foundries—Hughes rejected the idea that the right to organize was an “abstract right.” In American Steel Foundries, the Court had ruled that while it was limited by other rights and freedoms such as freedom of contract, the right to associate and bargain for better working conditions did exist. To be sure, in and of itself this was no breakthrough. The American Steel Foundries decision was so limited in its endorsement of the right to organize that it had hardly made a dent in the free labor legal apparatus that had construed the employee’s economic citizenship as the right to contract freely.49
However, what Hughes was intent on establishing in this case was different—reaffirming the American Steel Foundries decision was only a logical segue to the main part of its holding, which was that the federal government could legitimately intervene in the employer-employee relationship if it did so to protect the public good. In his dissent in Coppage v. Kansas in 1917, Hughes had contended that “the right to join a union is undisputed … and may be the legitimate subject of the protection of the police authority of the States,” which to him included the power to prohibit yellow-dog contracts. He followed this idea in the present case, ruling that the Railway Labor Act did not infringe on freedom of contract because it did not provide the U.S. Board of Mediation with the power to force an employer to hire or fire a railroad worker. Rather, the law simply protected the railroad worker’s right to organize with the intent of doing away with labor strife in railroads:
We entertain no doubt of the constitutional authority of Congress to enact the prohibition. The power to regulate commerce is the power to enact “all appropriate legislation” for its “protection or advancement”; to adopt measures “to promote its growth and insure its safety.” … The legality of collective action on the part of employees in order to safeguard their proper interests is not to be disputed…. Congress was not required to ignore this right of the employees but could safeguard it and seek to make their appropriate collective action an instrument of peace rather than of strife.50
Following a legal logic that hearkened back to—but went beyond—Harlan’s seminal dissent in Lochner v. New York, Hughes had thus adumbrated the possibility of a new deal in the employer-employee relationship. This new deal was a far cry from the legal recognition of the existence of an American working class with rights stemming from the workers’ social and economic subordination, for the Texas decision left the generic, common-law definition of the employer and employee as equal, free contractors intact. Now, however, the Court allowed the federal government to superimpose on that definition a specific category of “employees” made up of railroad workers only, a category whose existence rested on a more expansive reading of the regulatory rights of the government under the Commerce Clause, not on a novel reading of workers’ rights. As a result, while it was in theory nothing more than a limited right to associate, the right to organize and bargain collectively of a specific group of workerscould be strengthened and actively protected by the federal government if such was the object of a public policy designed in the public interest.
What the Court recognized in Texas, then, was the government’s visible hand on labor relations. And much of the subsequent debate on the legal definition of the worker would stem from the fact that because the New Deal collective bargaining regime was premised on a novel, expansive reading of the government’s regulatory power rather than on the individual rights of workers, it did not have to include all American workers in the category of “employees” it protected.
The Texas & New Orleanscase was not widely noted at the time. However, its significance was not lost on labor reformers and jurists determined to challenge the formalist notion that law and politics were fundamentally distinct. Thus, the Harvard Law Review explained that the case marked “a definite departure from the strict policy of laissez-faire” and insisted on the need to shift the emphasis of law from abstract principles to social engineering.51 Legal realists were not alone, however, in hoping that labor reformers stood on the cusp of a new era. Edward Berman, a liberal labor economist with the University of Illinois, explained approvingly that in the Texas & New Orleans case, the Court had produced a “very considerable extension of constitutional boundaries within which social legislation had been confined.”52 The American Federation of Labor was equally interested in the case, which figured in its brief defending the legality of the Norris La Guardia Act, a 1932 law forbidding the use of injunctions against unions.
Yet only in 1934, when labor progressives were faced with the difficulties associated with the implementation of Section 7a of the National Industrial Recovery Act (NIRA), did the Texas & New Orleanscase become directly relevant to the goal of reforming the workplace. Beginning in 1933, New York’s Senator Robert Wagner—by then the foremost labor reformer in Congress— and other liberals had sought to include American labor in the Roosevelt’s administration attempt to rationalize the economy. This effort stemmed from the numerous proposals for economic planning that were publicized and debated at the time, many of which did not include unions in the reorganization of the economy.53 To Wagner—whom Roosevelt had asked to synthesize the various proposals that circulated for an economic recovery bill—protecting the right to unionize was first and foremost a matter of political principle. Like Roosevelt, he had risen politically by advocating a transition from the sway of political machines to state welfare programs, but unions occupied a much more important place in his variant of urban liberalism. A former judge, he had collaborated very early on with Progressive jurists seeking to challenge the legality of the yellow-dog contract, and in the 1930s, he stood at the core of a network of labor-oriented reformers running the gamut from Taylor Society engineers like Morris Cooke to labor economists trained by John R. Commons and legal realists. Wagner was also close to important union leaders who advocated the rationalization of the economy. Relying on the experiences conducted in the 1920s in the mining and textile industry, where unions collaborated with management to stabilize labor relations, establish production goals, and uphold efficiency schemes, Wagner argued that collaboration and mutual respect between workers and employers were necessary for the NIRA codes to succeed.54 As we will see in the next chapter, in the 1920s and 1930s labor reformers did not abandon, but rather refurbished, the old ideal of industrial harmony.
This said, in spite of the spectacular rise in organizing that followed the adoption of the NIRA, the democratization of industry that Wagner envisioned when he secured the adoption of Section 7a in the NIRA did not take place.55 In many mass-production industries, businessmen countered this organizing tide by setting up company unions or, as in Ford factories, by creating a surveillance system. Far from agreeing to negotiate working rules with workers, some employers took advantage of the large pool of available workers to “speed up” the line and introduce labor-saving technology that reinforced unemployment, a strategy denounced in an NIRA Research Planning Division report in which economists Isador Lubin and Leon Henderson concluded that the automobile industry was “socially inadequate to meet its responsibilities.”56 As for the Roosevelt administration, it saw the workers’ increasing militancy as an obstacle to economic reform and remained doubtful that independent labor organizations were preferable to company unions, to which the president actually lent his support in a number of industries.57 Indeed, FDR’s only concession to the right to organize came in 1933, with the creation of the National Labor Board, which was charged with the responsibility to mediate labor conflicts but had no power to enforce its decisions. A year later, responding to the upheaval that swept through America in the spring of 1934, when 1.4 million workers staged over eighteen hundred strikes, Roosevelt also agreed to the creation of a stronger NLRB, but the defeat of the textile strike in the fall of 1934, with twelve Democratic governors moving against the strikers, revealed how shaky the administration’s labor policy really was. Disenchanted, NLRB Chair Francis Biddle lamented that the right to organize and bargain collectively was merely “a paper right.”58
Faced with employers determined to retain their authority over the workplace, labor reformers sought to reassert the constitutional language of rights and democracy that had animated talk of industrial democracy since the early days of the twentieth century. As they testified in favor of the Labor Disputes Act in 1934 (the first bill introduced by Senator Wagner) and its successor bill in 1935, the National Labor Relations Act in 1935 (also known as the Wagner Act), both of which were aimed at spelling the contours of the freedom to organize by protecting workers against a list of specified “unfair labor practices,” labor reformers couched their defense of the right to organize in terms that came close to making it an individual civil right. Commenting on the meaning of Section 7a a few days after the reintroduction of the Wagner Bill in 1935, Pennsylvania representative and majority whip Patrick J. Boland explained that “the ideas underlying this section are very simple. The worker is treated as a free man. He is accorded the right to associate with fellow workers, to join or refrain from joining any labor organization. He is protected from acts of aggression by his employer. His helplessness as an individual in bargaining with his employer is recognized.” The labor lawyer Francis Haas, then on the board of the NIRA’s National Labor Board, framed his defense of the bill in equally ambitious language: “Section 7(a) of the NIRA declares that workers may exercise rights that nature gives them, the same as our Federal and State constitutions declare that they may exercise rights which nature gives them to elect their representatives in Government, whether local or national.”59
Moreover, labor reformers deployed a Whiggish reading of history to depict the workers’ right to join a union as an essential step in the ongoing struggle for individual freedom against tyranny. “It is the next step in the logical unfolding on man’s eternal quest for freedom,” Wagner explained in defense of his bill in 1935. Echoing Roosevelt’s own allusions to the economic checks hanging on individual freedom in modern industrial society, Wagner continued, “Only 150 years ago did this country cast off the shackles of political despotism. And today, with economic problems occupying center stage, we strive to liberate the common man from destitution, from insecurity and from human exploitation.”60
Indeed, New Dealers argued that in the modern economy the asymmetry of power between the employer and the worker was so great as to foreclose the possibility that the worker might exercise his citizenship through individual bargaining.61 As the labor economist Harry Millis explained, “Of course, if there were perfect mobility of labor and keen competition for labor, and no concerted control of wages and hours by employers the case for collective bargaining would be less conclusive in the modern industry.” “The actualities of present-day life impel us to recognize economic duress,” Wagner concurred. “We are forced to recognize the futility of pretending that there is equality of freedom when a single workman, with only his job between his family and ruin, sits down to draw a contract of employment with a representative of a tremendous organization having thousands of workers at its call.”62 Wagner suggested that in the tense international context of the 1930s, denying this fact meant putting democracy in jeopardy, for disillusioned workers would turn to either fascism or communism to remedy their lack of freedom.
Yet as they built and defended the Wagner Act, labor reformers did not seek to constitutionalize associational freedom.63 Nor did they seek to revert the shift from working-class empowerment to employment regulation that Progressive reform had effected. Rather, following the drift of employment regulation since the late nineteenth century, they insisted on the positive social and economic effects of unionism, not the philosophical rationale that undergirded it. In a sense, this was a consequence of the Texas decision, in which the Court had affirmed the existence of the right to organize, leaving it to reformers to make a case for a public law providing it with additional protection. During the hearings, Senator Wagner and the jurists Robert Hale, Milton Handler, and Francis Biddle all insisted on the importance of the Texas decision and its main consequence—namely that the Adair and Coppage decisions had been overruled sub silentio and were no longer controlling. Consequently, Congress could enact a collective bargaining law without running into a Fifth Amendment due process wall. Wagner even pointed to Hughes’s dissent in the Coppage case, in which Hughes had noted that “the right to join unions … may be the legitimate subject of the protection of the police authority of the States.” The question that remained at this stage was whether the Court would sanction an interpretation of interstate commerce that would include federal regulation of the manufacturing sector.64
But equally important in the case for labor reform was the Progressive and new liberal idea that the regulation of the employment relationship produced positive freedoms, that is, freedoms that did not inhere in the citizen but rather in the public interest for a well-ordered society. Here, the ideology of harmony and the constitutional and legal drift of Progressive reform intersected. As the lawyer Francis Haas explained, the time had come to part with the idea that the “the wage contract concerns only an employer and an employee…. It concerns everybody else,” Haas contended.”65 The right to organize and to bargain collectively, in other words, was to make a contribution to the common good.
Accordingly, in making their case for a policy protecting collective bargaining, labor reformers insisted on the detrimental character of work stoppages for the flow of commerce. During the 1934 debates, the National Labor Board’s Milton Handler provided statistics showing the steady growth in the number of cases handled by his agency. By March 1934, the agency had intervened in over two thousand cases and passed the one million workers mark. The reports judiciously stressed the increasing number of strikes that had been settled, but also the number of strikes averted in places where both workers and employers had chosen the arbitration of the NLRB. Similarly, in the spring of 1935, reformers pointed to the declining militancy of the workforce—by then the 1934 strike wave had ebbed—as a sign of the success of the NIRA procedures, and labor leaders such as William Green testified about their recent efforts to prevent strikes in the automobile and rubber sectors. If the Wagner Act did not pass, Green explained, the menace to “industrial peace could not be exaggerated.” Indeed, the threat of worker militancy remained constant during the debates, giving the labor reformers a definite edge as they pressed the argument that strikes actually derived from the lack of workers’ rights.66 Harmony would come first in the guise of social peace.
Whether this argument would suffice was unclear. According to Leon Keyserling, the main drafter of the bill, the most important point to be made in the preamble was a broader one—that unionism was necessary to protect purchasing power and that a higher level of purchasing power was necessary to absorb the output of the American economy. The argument had been made since the debates over the adoption of the NIRA, but now the third draft of the bill made this point clearly: “[T]he failure of the total wage payments to advance as fast as production and corporate surpluses has resulted in inadequate purchasing power, which has accentuated periodic depressions and disrupted the flow of commerce,” the preamble now declared. Keyserling and Wagner believed that the interstate commerce argument would be stronger than the argument for industrial peace.67
The notion that mass production required mass consumption, of course, was at the heart of the Fordist production system and had led to the famous $5 day as early as 1914. Yet the appeal of the purchasing power rationale among reformers in the 1930s can be explained only by looking at the broader movement of businessmen, economists, and policy makers who had laid the theoretical groundwork for policies protecting purchasing power in the previous two decades. Businessmen such as Lincoln Filene, Eugene Grace, and Waddell Catchings had long argued that it was necessary for business to expand its pool of consumers, while economists such as Paul Douglas had provided statistics showing that productivity was rising faster than wages, thus creating an imbalance that might lead to an economic crisis. In the meantime, a consumer culture had enveloped both middle-class and working-class households, leading to social movements for lower prices, to which policy makers had responded by creating statist structures such as the Consumer Advisory Board.68
But the purchasing power argument mattered to the reconstruction of the wage relationship precisely because it fit squarely in the general pattern of labor reform, and labor reformers used it to win over the members of Congress and legitimize the intervention of the government in the employment relationship. To do so, they could point to America’s Capacity to Consume, a report in which Brookings Institution scholars Maurice Leven, Harold Moulton, and Clark Warburton likewise rejected the notion that the crisis originated in what a banker called the “glorious” overcapacity of the economy, which produced more than what American society actually needed. Their study revealed that 70 percent of American families lived under the threshold of $2,500 a year, which the authors of the study believed represented the poverty line in 1935.69 Whether the American economy actually had the capacity to produce enough to bring those families the earnings that they needed, the authors actually doubted very much.
The crisis thus reinforced the reformers’ tendency to focus on the ultimate economic aims of the regulation they sought rather than working-class empowerment. Indeed, the severity of the Depression and the causes New Dealers ascribed to it—insufficient consumption by the masses of workers— suggested that the labor question could not be solved by simply looking at the place of the worker in the production system. Rather, it was precisely through the worker’s social role as a consumer that the worker’s social inferiority would be solved. In his statement to Congress, Sidney Hillman, the president of the Amalgamated Clothing Workers of America, noted that the average worker did not have the ability to buy one set of clothes per year.70 In making this case, New Dealers thus suggested that the right to organize would buttress economic harmony and prevent future depressions.
By 1935, then, labor reformers had fully charted the path of a via media between socialism and unregulated capitalism. Although they sincerely believed that they were democratizing American industries, they addressed the question of subordinated labor only indirectly. In the Wagner Act, unionism was as much an economic means as a philosophical end. Throughout the hearings, reformers had built the strongest case possible for a public policy fostering unionism. Notably, industrial peace and the purchasing power argument dovetailed because they abutted on the same institutional venue— collective bargaining. Indeed, the law did not simply protect workers’ right to join a union; it also imposed on employers the duty to bargain with the workers’ representatives and sustained collective bargaining through the adoption of majority rule and the decision to give the new NLRB the ability to decide what bargaining units would be most adequate to the goal of bargaining.
On April 12, 1937, in NLRB v. Jones and Laughlin, the Supreme Court lent its support to this reasoning. Following the argument laid out in his own opinion in Texas & New Orleans, Charles E. Hughes, once again the author of the majority opinion in Jones and Laughlin, reaffirmed the idea that the right to organize was a “fundamental right,” one that that the Court had long recognized.71 The main part of the holding, however, bore upon the right of Congress to enact a law fostering the practice of collective bargaining. On this question too, Hughes followed his previous opinion, arguing that the power of Congress to regulate commerce was the power to enact “all appropriate legislation for its protection and advancement.” This power, Hughes concluded, was “plenary.”72 Moreover, the Court said that it refused to “shut its eyes to the plainest facts of industrial life,” and deal with the definition of interstate commerce “in an intellectual vacuum.” Given the size and organization of companies on a national scale, the Court insisted, it had become impossible to continue to hold that their labor relations were in essence a local matter.73 Even if a company’s relation to interstate commerce was only Struggling Against Class 57 indirect, Congress could legitimately decide to steer it toward labor peace rather than strife.
In Jones and Laughlin, the Supreme Court thus completed the slow jurisprudential evolution it had started in the Progressive Era. In validating the Wagner Act, it sanctioned a serious exception to the at-will rule that had come to dominate employment contracts in the second half of the nineteenth century. The Wagner Act, however, did not erase the at-will rule from the nation’s common law. Nor did it construe all workers as subordinated labor. Rather, it built the government’s right to foster collective bargaining in the nation’s factories and offices and left to the government the possibility to decide which workers would be covered by this policy. In doing so, the act, like other New Deal laws, built a definite gap between the worker and the worker’s legal definition—the “employee”—which was now the main legal fulcrum upon which labor relations turned.74
Like its Progressive forebears, the Wagner Act made only a passing reference to “workers” and focused on “employees,” whom it defined thus:
The term “employee” shall include any employee, and shall not be limited to the employees of a particular employer, unless the Act explicitly states otherwise, and shall include any individual whose work has ceased as a consequence of, or in connection with, any current labor dispute or because of any unfair labor practice, and who has not obtained any other regular and substantially equivalent employment, but shall not include any individual employed as an agricultural laborer, or in the domestic service of any family or person at his home, or any individual employed by his parent or spouse.
In a sense, this was an expansive definition of the worker meant to include all workers involved in a dispute rather than only the workers employed by a given company. Indeed, because Wagner and his advisers remembered how the Clayton Act had been eviscerated by the courts through a narrow definition of “employee,” in the National Labor Relations Act (NLRA) they purposely defined “employee” as including not simply workers who stood in a proximate employer-employee relationship with an employer, but also all the workers who had an indirect stake in an ongoing labor conflict (“and shall not be limited to the employees of a particular employer”). Overcoming resistance from members of Congress who argued that the law would allow outside agitators to come and fan the flames of labor strife, Wagner and other reformers suggested that the right to bargain collectively would not be made effective if sympathetic strikes were illegal and more generally if industry-wide bargaining was not established.75 In this respect, the definition of “employee” reflected his view that collective bargaining would be ineffective unless the masses of workers cohered in large enough movements.
Interestingly, the Supreme Court sanctioned this recognition of working-class power precisely because it saw the definition of “employee” as being subservient to the a public policy: the aim of Congress was not to “adjudicate private rights,” it explained in 1941, but rather to give effect to a broad economic policy. Citing the decision a few years later, the NLRB commissioners applauded this expansive definition of “employee,” saying that it was “broad enough to include members of the working class generally.”76
Yet one can say that the Wagner Act protected the “working class” through its definition of “employee” only if one recognizes the thoroughly contingent, contested character of the concept of class and its evolution over time. As the main legal vehicle through which this policy was delivered, the definition of “employee” certainly eased the transition from ethnic to class consciousness that Lizabeth Cohen has described.77 Not only did it bestow a common legal identity on workers that facilitated this transition, but it also fostered class consciousness by allowing them to forge social links beyond individual locales. This explains why to the millions of workers who joined the CIO after 1935, there was no significant difference between the terms “worker” and “employee.” With the Wagner Act, labor progressives had instilled a class dimension in “employee,” one that allowed workers to use either term to refer to the newfound sense of economic security that unionism had nurtured among them.
To others, however, it made all the difference. In 1934 and 1935, civil rights organizations watched helplessly as members of Congress winnowed the definition of “employee” to exclude agricultural workers. In 1933, during the legislative debates over the NIRA, southern politicians such as Huey Long had lamented that Section 7a applied to “all laboring men,” and although the text was not amended, it was implemented in a restrictive fashion, leaving most agricultural workers outside of the law. The question was raised again, however, in March 1934, when the Labor Disputes Act was introduced, for at first it contained no provision excluding agricultural workers and was understood to apply to all workers. When the bill was referred to the Committee of Education and Labor, Alabama Senator Hugo Black—later Roosevelt’s first appointment to the Supreme Court—and Senator Trammell from Florida worked with three senators, from Iowa, Montana, and Utah, and wrote an agricultural exemption in the bill that reflected the ability of what a historian has called the “marriage of corn and cotton” in applying brakes on labor liberalism and containing the reform to the modern factory. Wagner knew fully well that given the number of votes that these men represented in Congress, there was no avoiding the agricultural exception, but “employee” was malleable enough to be interpreted as covering industrial workers while excluding others. Its definition thus registered the conservatism of a large group of states that would launch the right to work movement a few years later.78
For black organizations, this defeat mattered enormously because in spite of the deep-seated racism of many labor organizations, their relationship to unionism was then in a state of flux. The National Urban League had shed its erstwhile opposition to unionism, and, along with the National Negro Congress, it now stressed the economic character of social inequality in America. This “proletarian turn,” as historian Thomas Sugrue has called it, was precisely facilitated by the upsurge in union organizing in transportation, where a clause equivalent to the NIRA’s Section 7a, existed but also among sharecroppers and farm tenants in Arkansas, where the possibility of an interracial unionism could now be glimpsed.79
Indeed, African American organizations were certainly among the first to gauge the importance and the nature of the change that the Wagner Act had wrought in the working-class struggle, which was no longer limited to the defense of labor solidarity on the shop floor or in the farm. To travel the via media charted by the New Dealers, workers now had to be officially recognized as “employees.” Black workers learned this lesson very early because they chafed under the discriminatory practices established by most unions in northern cities. Although black Americans composed 8 percent of the nonagricultural workforce, they accounted for only 1 percent of unionized workers, with half of this membership in the Brotherhood of Sleeping Car Porters. In general, blacks were either kept out of unions altogether or, when they were accepted, remained relegated to the lowliest jobs and were forced into segregated locals. As a result, as the National Urban League’s Arnold T. Hill explained in a letter to Robert Wagner, “It’s been through going in to break strikes that many [black workers] have found employment in coal mines, steel plants, and other important industries.” Indeed, in the 1920s black workers helped defeat strikes in a number of industries such as coal and steel, but also in automobile textile manufacturing and slaughterhouses.80
Yet the first drafts of the Labor Disputes Act in 1934 explicitly excluded “strikebreakers” from the definition of “employee.” This was an attempt to correct earlier National Labor Board decisions, according to which an employer was free to engage replacement workers during a strike and retain them when the strike was over as long as the strike was an “economic” one, and did not stem from a violation of the Section 7a disposition in the NIRA. In excluding strikebreakers from the statutory definition of “employee,” the drafters of the act hoped to strengthen the recall rights of strikers and vindicate their claims to reinstatement.81
When combined with the provision for majority rule that drafters had included to forestall company unionism and facilitate collective bargaining, the bill, although aimed at protecting workers, thus threatened to reinforce the discriminatory pattern from which black workers already suffered. Notably, the Harvard sociologist Kelly Miller indicated that if it were left unamended, the bill would have detrimental consequences in the very few places where black workers had gained an industrial foothold—Ford factories especially.82 The National Negro Congress and the National Urban League already denounced discrimination in New Deal programs, and they pointedly pleaded with Roosevelt to address the question of discrimination in labor unions.83 When the Wagner Act was first introduced in 1934, they promptly reacted, sending numerous requests for an amendment protecting black workers, who were kept outside of the world of labor: “The term ‘employee’ shall not include an individual who has replaced a striking employee, except when the labor organization either by direct constitutional or ritualistic regulation and/ or by practices traceable to discriminatory policies bars an individual from joining such labor organization or restricts rights, privileges, and practices usually accorded members of such labor organizations” (my emphasis).84
Overall demands for amendments to protect African Americans from Jim Crow were usually of no avail because of the concerted opposition of AFL and southern lawmakers, but in this precise case the demands arising from the African American organizations drew support from other sources—not only did corporate America express its opposition to a clause that would severely hamstring its control of the workplace, but even liberal lawmakers balked at the idea that replacement workers would not be able to shed their “nonemployee” status if they retained their new job. Indeed, Professor John Fitch of the New York School of Social Work convinced Senator Wagner that the purposes of the bill would be better served by allowing all replacement workers to bargain with their new employer, a position that can be explained by the fact that the employers’ right to hire replacement workers was not in doubt and that excluding strikebreakers would not directly contribute to developing collective bargaining.85 Accordingly, when the bill was reintroduced and later adopted in 1935, the reference to strikebreakers had disappeared from the definition of “employee.”
The limits inherent in the definition of the worker for labor relations were made obvious a few years later in a case that directly harkened back to the Progressive struggle to root out injustice in the workplace—child labor. In NLRB v. Hearst Pub. Inc., the Court passed on the case of newsboys, whose bargaining rights were denied by their employers on the grounds that they were not “employees” but “independent contractors.” Upholding the NLRB’s decision that the boys were “employees,” the Supreme Court once again reaffirmed the specific character of “employee” under the Wagner Act. Indeed, the high tribunal denied that Congress had imported in the Wagner Act either the common-law definition of “employee” or a social definition of worker whereby anyone hired—that is, all workers—would be protected by the law.86 “Congress did not treat employee as a word of art having a definite meaning,” the Court explained. Rather, “it [“employee”] takes its color from its surroundings, in the statute where it appears, and derives meaning from the context of that statute which must be read in the light to the mischief to be corrected and the end to be attained.”87
No opinion better expressed the promise and limitations inherent in the Wagner Act. While the intervention of the government in the employment relationship liberated millions of workers from the arbitrary rule of management, the Wagner Act raised, but did not solve, the question of which workers would be defined as “employees.” Indeed, what Progressives intent on deflecting class antagonism had really built was the government’s visible hand on labor relations. But the Wagner Act protected no substantive civil right. As the Supreme Court explained in 1944, a pattern of inclusion and exclusion was at the heart of the logic of the reshaping of labor relations.
As a result, included as “employees” were the workers who toiled by the thousands in the mass-production industries on which the Fordist economy relied. For them, the right to organize went hand in hand with the evolution of personnel policies, which sought to foster their long-term attachment and fidelity by offering promotion patterns and social benefits. The growth of unionism and the security it brought, then, paralleled its evolution in management policies. For many others, the Wagner Act laid the groundwork for a struggle for inclusion. To paraphrase E. P. Thompson, the “employee” was to be made and remade, not “given.” But to make themselves “employees,” workers would have to fight more than a cultural struggle—theirs would be a legal and political struggle in the arenas of the NLRB and the courts as much as a social one.
In the 1930s these limitations did not worry labor reformers, who believed that they had fundamentally reshaped labor relations according to the needs of a mass-production economy. Notably, this faith stemmed from the fact that labor reformers such as Robert Wagner and Leon Keyserling had built strong ideological ties with the labor economists trained by John R. Commons in Wisconsin. What undergirded their faith in the Wagner Act was the sociology of labor relations developed by Commons, according to whom collective bargaining could deliver this most wanted social product—harmony and partnership between “employer and employee.”