CHAPTER TWO


The Legitimacy of Identification with Generality

BUREAUCRACY AND POLITICS: A BRIEF HISTORY

The idea of endowing the government bureaucracy with a certain autonomy was first formulated around the turn of the twentieth century. Merely to think in such terms marked a major break with all previous understandings of democratic politics. Historically, democracy rested on the idea that all the institutions of government were strictly responsible to the sovereign people, who alone determined the public interest. The government chosen by the people at the polls was supposed to implement the decisions of the voters, and the bureaucracy was merely an arm of the elected government. In this context, the phrase bureaucratic power had no meaning, unless it was to suggest a culpable usurpation of power that rightly belonged to the people. This was true in both America and France, the first two countries to embrace universal suffrage.

In America, we see this in the “spoils system,” which granted the party in power quasi-ownership of all public employments.1 The spoils system began with Andrew Jackson. On assuming the presidency in 1829, Jackson set out to exemplify a new democratic spirit, a more radical version of Jeffersonian democracy. He sought to rid the country of the Founding Fathers’ “aristocratic” concept of representative government and to end the claims to independence advanced by many public officials. He purged his administration of such men and appointed others who embraced his way of thinking, which had been endorsed by the voters. His goal was to “democratize” the government and combat what he denounced as unconscionable privileges. In his mind, advancing democracy meant placing public officials under the direct supervision of those who had emerged victorious from the polls.

The outlook in France was similar. Already in the revolutionary period people had been obsessed with putting an end to “ministerial power,” which many saw as the very essence of the Old Regime. What the men of 1789 feared was that the bureaucracy might set itself up as an independent power. Any intermediary structure that stood in the way of direct communication between the nation and its elected representatives was a priori suspect of impeding the expression of the national will and conspiring against liberty. More radical still was suspicion of the executive itself, which was understood at the time as a purely delegated power.2 The revolutionaries repeatedly denigrated the executive, insisting that ministers were merely the servants of the legislative power. In 1794 they went so far as to replace ministries by committees directly responsible to the National Assembly, on the grounds that only one voice—that of the elected representatives of the people—was authorized to pronounce the general interest. More broadly, they believed that the role of unelected public officials was simply to execute the laws mechanically, in total obedience to the directives of the political authorities. “The people who clean our streets and light our lamps are delegates of the sovereign,” one commentator ironically remarked in the 1820s.3 In this context, it was impossible to conceive of the administrative component of government as a distinctive part of the governing process.4

In practice, however, the theory of the subordination of the administrative to the political was undermined by the perverse effects of patronage. Hence the desirability of setting objective criteria for the selection of civil servants was widely discussed in the first half of the nineteenth century. All eyes turned to the German states, which led the way in rationalizing the administrative functions of government. In the 1830s and 1840s, many foreign missions came to Prussia and Wurtemberg to observe their bureaucracies. Édouard Laboulaye returned from one such mission with a study that would become a model for Europe, still cited at the end of the nineteenth century.5 Yet it proved difficult to turn the ideas inspired by this study into practical reforms. In 1845, the French Chamber of Deputies rejected the idea that admission to state employment should be subject to a proof of aptitude by test or competitive examination.6 A proposal to set rules for promotion was rejected at the same time. Although nearly everyone recognized that such measures would yield benefits, fear of limiting the freedom of action of government ministers carried the day. All political camps shared this fear. The Vicomte de Cormenin, one of the most celebrated republican writers of the day, wrote: “We do not really see what would be gained by having permanent functionaries serving under impermanent ministers. We want ministers whose hands are free and who can move easily within their assigned sphere of action”.7

Things evolved very slowly. When public opinion in Britain turned against the patronage system in the 1850s, civil servants were granted certain guarantees against political and parliamentary meddling. In the 1870s, a system of competitive examinations was instituted, marking the beginnings of a modern civil service. The United States soon followed suit with the Pendleton Act of 1883. This reform fixed the spoils system around the edges only, however: only about 10 percent of public jobs were subject to meritocratic recruitment.8

Resistance to change was particularly strong in France. To be sure, voices were raised on all sides in protest against the corrupting influence of “favoritism”. In effect, public employments were privatized, as posts were handed out to the protégés and relatives of those in power. In the mid-1880s, parliament still saw itself as powerless to end the practices noted in one report: “The schoolteacher is at the mercy of the prefect, who appoints, suspends, and dismisses teachers as he pleases”.9 But deputies and senators were in no more of a hurry than ministers to give up the opportunity to place their candidates in various jobs.

The idea that the employees of the state ought to be subordinate to the elected powers thus remained quite influential. Elected officials themselves regularly justified this idea as a requirement of democracy. Each change of regime or even each change of majority thus resulted in an administrative purge.10 The most significant of these followed the republican victory in October 1887. Earlier, in 1879 the Conseil d’État had been severely purged, an action justified, according to the majority, by the need for “unanimous adherence within the Council to the republican point of view”.

“Elimination of all elements hostile to the Republic from all public posts, from the highest to the humblest,” was actively encouraged.11 The purge was extended to the judiciary in 1883. Even though these purges exacted a real price, spreading organizational chaos throughout the government, the ruling majority deemed them justified. At Radical Party conventions, it was still common in the early part of the twentieth century to pass motions urging the dismissal of civil servants judged to be reactionary or proclerical and their replacement by men considered to be “sincerely republican”.12 Although the need to modernize the state was in fact recognized, doctrinal resistance always won out.

This view of the relationship between bureaucracy and politics changed sharply at the end of the nineteenth century. In France there were several factors at work. First, the size of the state increased noticeably, bringing the question of its efficiency front and center. More important, though, was the fact that parliament had lost much of its prestige. Once seen as a major prize in the battle for democracy, its symbolic significance had long since diminished. Indeed, it had come to symbolize the perversion of democracy’s original ideals. Antiparliamentarism became a powerful force in the 1890s, fed by proliferating scandals of which the Panama affair was the most prominent. This weakened the justification for the domination of the bureaucracy by the political authorities. Although the law continued to be seen as “the expression of the general will,” the masses had become disenchanted with the idea. In their eyes, the representative institutions of the state no longer embodied the Republic or the rightful rule of universal suffrage. Disappointment on the Left converged with old aristocratic reservations on the Right, leading many people to look anew at ways of expressing the general interest. The belief that the general interest somehow automatically emerged from the polls lost its hold on people’s minds. A distinction of a sort emerged between democracy (as defined by electoral majorities) and republic (the substantive expression of social generality). Evidence of this can be seen in the very sharp decline in the use of the word democracy as a synonym for the political ideal in this period.

A comparable revolution took place in the United States in the same period, 1890–1900. There, it was the political parties that bore the brunt of criticism. They were accused of manipulation, corruption, and prevarication—sins that made them anything but representatives of the general interest. The ravages of corruption were widely acknowledged.13 The Progressive Movement represented an effort to break this pattern and breathe new life into the promise of democracy. Three major avenues were explored simultaneously. The first was to tighten citizen control over elected representatives and develop forms of direct democracy. It was at this time that the system of primary elections was devised as a way to reduce the influence of the party apparatus over the selection of candidates. A number of western states also established procedures for initiatives and referendums, as well as for recall elections to remove elected representatives from office. In addition, independent regulatory authorities, modeled on the Interstate Commerce Commission of 1887, responded to popular demands to curb the excesses of industry (I will return to this later on). A third way was also envisaged, namely, the creation of a more autonomous and more rational public administration.

Why was corruption so toxic to democracy? Because it represented privatization of the public’s business in its most extreme form. At the time, the parties were instruments not for representing the public interest but for enabling private interests to capture public space. In this context, the idea that “politics should be banished” from the management of public affairs was able to gain a hearing. Though at odds with the classical image of democracy, the wisdom of “keeping politics out” took on a life of its own. Many people concluded that a strong bureaucracy could stand as a rampart against efforts to monopolize public institutions for private ends. Independent regulatory agencies provided the material means to withstand partisan pressure. Out of this came a progressive idea of legitimacy vested in objective administrative power.

Thus we find that in both the United States and France, the first two countries to adopt universal suffrage, people began in the 1880s to feel a need for a more concrete embodiment of the general interest. There were two aspects to this development: the emergence of a corporatism of the universal and the constitution of an objective administrative power.14 France was more systematic in its exploration of the corporatism of the universal. The United States was more interested in rationalizing the bureaucracy and making it safe from partisan politics. In both countries, however, the central idea was to create powers within government capable of representing generality. Paradoxically, it was in liberal America at the turn of the twentieth century that the modern doctrine of a democratically legitimated bureaucratic power was first formulated. No less paradoxical was the fact that it was in Jacobin France, ostensibly hostile to corporatism in all its forms, that corporatism was enshrined in the civil service. Both cases show that the true history of democracy is quite different from the way in which it is often presented.

THE CORPORATISM OF THE UNIVERSAL

The period 1880–1914 was a time in which the social sciences flourished in France.15 The new thinking undermined older legal and political theories of the subject and of sovereignty. In sociology, the work of Alfred Fouillée, Alfred Espinas, and above all Émile Durkheim discredited earlier understandings of the democracy of will. This led one of the great jurists of the period, Léon Duguit, to develop a radical critique of theories of the state. Duguit’s recognition that individualist theories of the French Revolution were no longer applicable to the society of his time underlies all his work: the social bond could no longer be conceived as a political contract among individuals who agree to erect a central social power.

Duguit therefore rejected the French theory of sovereignty, which treated the sovereign people as a supreme collective entity and public power as a subjective right. In his view, this theory derived from concepts of modern public law, which, ever since Jean Bodin, had been based on concepts of Roman private law. Thinkers in this tradition had conceptualized political power, or imperium, in terms of individual ownership, or dominium, with all its attributes. From this came the idea of the patrimonial state, which dominated Europe for a time. The theory of the patrimonial state was especially deeply rooted in France, where the Jacobin ideal had merely transferred the attributes of royal power to a collective sovereign.16 For Duguit, the new sociological understanding of social interdependence suggested the need for a more objective concept of the role of the state. “The idea of public service must replace that of public authority,” he concluded.17

For Duguit, the modern state can no longer be defined as a “power that commands”. It should be understood, rather, as “a cooperating array of organized public services”.18 By the same token, those who govern are not merely instruments of an authority that looms above the society that instituted it. They are merely the managers of the public’s business. “If there is a public power,” Duguit concludes, “it is a duty or function and not a right”.19 Central to his work is the idea of “public service,” comprising those activities deemed to be “indispensable to the realization and development of social interdependence”.20 Consequently, public service needs to be reliable and continuous. Pursuing this radical change of perspective, Duguit attempts to rethink the foundations of public law. Instead of a subjective right of command, he sees an objective duty to accomplish certain tasks: “Modern public law,” he argues, “becomes a set of rules determining the organization of public services and ensuring their regular and uninterrupted operation”.21

Duguit’s approach to these issues led him to rethink the concept of political legitimacy: “Public power cannot be legitimated by its origin but only by the services it renders pursuant to the rule of law”.22 As the basis of administrative power he therefore substitutes the notion of general interest for that of general will. In other words, the key to understanding public power is not how it was instituted but rather the purposes for which it acts. From this insight Duguit derived his idea of “objective right”. Rather than view the law as an expression of the general will (as a celebrated revolutionary formula had it), Duguit saw it as a formal statement of the social interests that the general will was supposed to embody. He thus totally rejected the Rousseauian idea of the social contract.

The procedure for constructing a sovereign will by way of elections thus ceased to be central in his theory. In its stead he substituted an objective process for identifying the needs of society based on its nature and structure. These needs then determined the duties of the public authorities, whose only justification lay in their function. Social science therefore played a key role in the new theory. Duguit was by no means the only theorist to pursue this line of thought. He was merely the most eminent representative of a broad movement that is sometimes characterized as “the social law school”.23 His ideas exerted considerable influence, well beyond the law schools. They were received favorably in reformist circles in many countries. For example, Harold Laski, the well-known English political theorist, wrote that Duguit’s mark on his generation “can be compared to that left by L’Esprit des lois in its time. Both disciples and adversaries were obliged to adapt their concepts to the new framework that he developed”. The critical work of the pioneer of objective law, Laski continued, will “be seen by future historians as the dawn of a new era”.24

Duguit saw the state as a federation of public services whose mission was to organize society to serve the public interest. Thus government was no longer defined by the classical “regalian” powers. To be sure, the functions of police, justice, and defense remained essential for the protection of individuals, but the essence of government was now to forge a true society out of existing groups, associations, and other forms of solidarity. In this sense, the state continued to be the “institutor of society” but in a new way. One thought of it not as shaping a formless mass of individuals but rather as a force for coordinating a range of autonomous public services, each of which was charged with defining the general interest in its specific area of responsibility. At the heart of the process by which the general interest was defined was thus the civil servant rather than the elected representative.

In a state defined in this way as a collection of public services, civil servants are not simply employees of the collectivity. They do not simply carry out orders issued by elected officials who supposedly represent the general interest. They are also active agents in their own right, who “contribute to the provision of a service that numbers among the state’s essential missions”.25 In other words, their function is to serve the common good. The modern bureaucrat must therefore enjoy a certain degree of independence: he must “have a stable situation, independent of the whims of government … [and] enjoy what is called a ‘status,’ both in his own interest and in the interest of the service”.26 For Duguit, the civil servant, or functionary—fonctionnaire in French—is a person identified with his function. To be sure, the reality was ambiguous. Maurice Hauriou, the other great theorist of the public service, emphasized the idea that the civil servant “finds himself in a complex situation, responding at times to the instructions of his superiors and at times to the inspirations of his function, and relying not only on authority delegated by the government but also on the power that is intrinsically his in virtue of the autonomy of his function”.27 Like Duguit, however, he insists on the autonomy of the civil servant, which allows him to contribute directly to realizing the goals of society and accomplishing the missions of his department of the civil service.28 In other words, the nature of their role is determined essentially by the objective character of their function.

This new appreciation of the civil servant’s role was not simply the result of a novel theory of the state associated with the transformation of the relationship between state and society. It was also a consequence of a sociological fact: the growing political influence of schoolteachers. There were 120,000 of them in 1914, or more than a quarter of the total number of civil servants. They constituted the single largest contingent of state employees, as well as the most homogeneous. Above all, their place in society was inextricably intertwined with the consolidation of the republican system. Throughout the Third Republic, the “black-clad hussars of the Republic,” as Charles Péguy famously called them, were totally identified with the regime, as much as they were identified with the country itself. They would have been perfectly justified to claim, “L’État, c’est nous,” because that is how they were in fact perceived by their fellow citizens. Significantly, nearly half of them described their work as “a vocation,” instinctively adopting the language of the grands corps of the Ancien Régime to indicate that they identified with their function in a moral as well as a professional sense.29 As one minister of education, Eugène Spuller, put it, “A public function of this sort is not a mere occupation,” adding that that was why the “emoluments” (traitement) paid to teachers were not to be confused with “wages” (salaire).30 These words might have been spoken by a high official of the monarchy two centuries earlier.

To emphasize the connection between the new type of civil servant he had in mind and the requirements of the general interest, Duguit used the terms functional decentralization and decentralization by service.31 The idea was that public service missions were to be delegated to agents with recognized professional skills. Duguit went so far as to speak of a “franchise” (concession), and he envisioned a sort of “patrimonialization” of public service, by which he meant that civil servants should be provided with the resources they need to accomplish their assigned tasks. He also applied the phrase corporatist organization to a system in which “civil servants are managers of the public services”.32 To construct the “corporatism of the general interest,” Hauriou conceived of the idea of a “social office”. This idea, as provocative as it was suggestive, implied that certain officials (judges, military officers, teachers) enjoyed a privileged status conferring property rights to their functions and even their jobs.33 In using this vocabulary, borrowed from the public law and social organization of the Ancien Régime, he aimed to get across the idea that what was needed was a functional equivalent of the orders and corps of the Ancien Régime to enable a republican government to act in the general interest—as if no one could be better equipped to perform a task than a person whose very existence was identified with it. Lawmakers to a certain extent ratified this view. For instance, in 1880 and 1886 laws were passed guaranteeing tenure to university professors, whose careers were subject only to the governing bodies of the faculties to which they belonged. The “teaching corps” was thus the precursor of the republican corporatism that Duguit and Hauriou proposed to expand.

In the first few years of the twentieth century, “status” (statut) became the standard term in France for describing a civilian servant’s perquisites and the duties corresponding to them. Duguit was among the first theorists to investigate this question of civil service status. He sought to show that civil servants were employees like any other, directly subordinate to their employer, in this case the state. Their position had to be understood more broadly in terms of their objective responsibilities as servants of the state: “Although the civil servant reaps the benefits of his status,” Duguit observed, “it was not in fact bestowed upon him to serve his interests but rather to serve the interests of the public service”.34 The civil servant is granted a status because it is assumed that he will identify with his mission all the more strongly if he believes that he is protected by the law. The goal is to align the interests of the service with the interests of the functionary: “The more firmly his situation is protected, the more he will work, and the more effectively as well”.35 The intention is to make the individual one with his function psychologically as well as materially. The grant of status imposes a social ethic on a professional group. Or, to put it another way, it gives civil servants an interest in disinterestedness. Durkheim, who believed fervently in the status principle, thought that the state could transform itself into “a unique group of civil servants, who together develop ideas and projects on behalf of the community”.36 The purpose of the status is to “functionalize” employees of the state, to turn them into a group without particular ties to the outside—in short, an incarnation of the “corporatism of the universal”. The hope was that, with the grant of status, civil servants could realize the Hegelian vision of them as a “universal class”. The goal, in Hegel’s own words, was to constitute a group “having as its immediate aim to posit the universal as the purpose of its essential activity”.37

RATIONAL ADMINISTRATION

Historically, there was also another way of achieving identification with the general interest: through the constitution of an objective administrative power not subject to any particular influence and absolutely identified with its task. Whereas the corporatism of the universal emphasized the crucial role of civil servants dedicated to their mission, here the goal was to establish a power whose very form would wed it to the general interest. Scientific policy and rational administration were supposed to ensure the realization of the common good. The theory of this path to the democratic ideal was developed mainly in the United States by Woodrow Wilson and Frank Goodnow. Wilson, who became president in 1913, published a pioneering article entitled “The Study of Administration” in 1887.38 His goal was to found a “new science” of “practical government”.39 Traditional political science, he felt, was too exclusively concerned with constitutional questions. In a complex society, the question of democracy extends beyond debates over the constitutional text, the legislative process, and the organization of elections. To serve the general interest one must go to the heart of how society works. That, in his view, is where the key distinction between politics and administration comes into play. The general will (broadly defined as establishing the rules governing the organization of the polity) finds expression in the political sphere. In theory, administration is merely the application of principles decided elsewhere. But, Wilson maintains, in the modern world things are more complex. Goals cannot be defined apart from the routine issues that come up as one tries to achieve them. A science of administration is therefore necessary for two reasons: both efficiency and democracy require it. Wilson’s pioneering article raised questions that one of his Columbia University colleagues, Frank Goodnow, would try to resolve.40

Goodnow was the real founder of American administrative law. A progressive as well as an academic of the first rank, this friend of the great historian Charles Beard proposed a new concept of American public administration.41 He stressed the idea that the true power of the executive lies in administration. In his essential work on the subject, Politics and Administration, published in 1900, he revisited the classical theory of the separation of powers and looked at the way the different branches of government actually work.42 Like many others of his generation, he was keen to break away from stereotypical ideas about democracy, too normative for his taste, and focus instead on the living reality.43 For him, the realm of politics is limited to legislative and constitutional work, while the executive operates in the administrative sphere. If the essence of political is thus, by assumption, to express the general will, the essence of administration lies in the pursuit of efficiency and rationality. Indeed, the administration can work toward “executive perfection” only in an internal mode (whereas “legislative perfection” lies entirely in its dependence on the external will of the sovereign people). The administration thus differs from the political branches by virtue of its relation to generality. The administration is primarily concerned with excluding any deviation from the general interest toward special interests; it thus embodies substantive generality. By contrast, the political order, whose goal is to include the largest possible number of citizens in as unanimous as possible an expression of the collective will, embodies procedural generality. For Goodnow, there were two aspects to a more realistic approach to the state: first, the growing role of political parties had to be taken into account and subjected to some sort of regulation (hence the importance of primary elections), and second, the legitimate scope of administrative autonomy had to be clearly delineated. What could efficiency and expertise accomplish? The general will as defined by the ballot box constituted subjective democracy, which needed to be balanced by objective democracy, defined as nonpartisan bureaucratic rationality.

Since no such independent bureaucracy really existed in the United States, Goodnow emphasized the need to develop one capable of serving and protecting the general welfare.44 The focus on efficiency and rationality, he believed, would be enough to guarantee objectivity. Indeed, the early twentieth-century American progressives developed a true mystique of rationality. Reason and efficiency were enshrined among the democratic virtues.45 Political science journals published countless articles on the subject. More than that, a veritable social and cultural movement developed in celebration of these ideas. Publications such as Efficiency Magazine and the Journal of the Efficiency Society were devoted to the subject, and a variety of organizations investigated the sources of efficiency and praised its benefits.

It was in this context that Frederick Winslow Taylor developed his theory of scientific management. The man who was to revolutionize the organization of modern business was both product and symbol of the “fever for rationality”. This accounts for the enormous success that greeted his Principles of Scientific Management when the work was first published in 1911. Subsequently, Taylor’s work helped to spread the ideas that had inspired it. Although the Taylorist ideal of rational management is today associated mainly with its industrial applications, its political origins should not be forgotten. Well before industrial firms were converted to “Taylorism,” the federal bureaucracy sprouted a whole range of new agencies with evocative titles such as Commission on Economy and Efficiency, Bureau of Efficiency, and Commission on Departmental Methods.46 The creation of an impersonal, rational bureaucracy was understood at the time as a way of serving the general interest. It was thus the Americans who proved to be most enthusiastic about applying Max Weber’s idea that improving instrumental rationality by developing a modern bureaucracy was essential to achieving a “deeper” democracy.

American Progressives believed that a scientific government would enhance both democracy and order. This belief had a definite sociological foundation. It was associated with the growing power of a new middle class, a class shaped by “professionalization” in all walks of life and by the professional organizations representing individuals in a variety of fields. Doctors, academics, journalists, accountants—no matter what the field, activities were increasingly professionalized, recognized by the granting of degrees, regulated by professional codes of conduct, and represented by specific organizations and publications. All of these created a more structured social environment in which individuals were identified with their social function, both to protect them and to empower them. In some ways, the calls for rationalization of the administration were linked to this change. But reformers also had broader political goals in mind: a more efficient and autonomous government bureaucracy would yield better policies and better results. Leading figures such as Herbert Croly, the author of Progressive Democracy (1914), and Walter Lippmann, who published Preface to Politics (1913), were able to adapt this praise of expert reason to American political culture by wrapping it in an emotive and mystical sensibility.

Enthusiasm for these new ideas was not devoid of ambiguity. The democratic mystique sometimes went hand in hand with suspicion of “King Demos”. The notion of scientific government was a way of combining both sensibilities without having to choose between them. One even finds both attitudes combined in the mind of a single individual. For example, some of the same progressive reformers who called for states to allow recalls and referendums also backed literacy tests and other strict rules for the registration of voters. (Indeed, such ambiguities account for the ambivalent judgment that many historians have of the Progressive Era.) In essence, though, the construction of democracy was seen as a fight for generality against the powers of particularity and all the distortions they were able to introduce.

The choice of expertise and rationality as central democratic values affected many aspects of the American political system, not only at the federal level but also at the state and local levels. In a federal system with a central government that for a long time remained relatively weak, most people came into contact with the political sphere at the municipal level. It should come as no surprise, then, that many of the problems of democratic politics were concentrated in the cities, where the flaws of democracy were exacerbated. It was in the cities, for example, that the consequences of the spoils system were most pronounced (the Pendleton Act of 1883 required that federal civil servants be recruited on the basis of merit but did not extend to the local level). It was also in the cities that the parties exerted the greatest influence on public affairs. Most cities were controlled by political “bosses” in charge of local party machines. Elected officials such as the mayor were in fact underlings of the political boss, who decided whom to hire and whom to fire and influenced all political decisions. The system fostered widespread corruption. Corrupt city machines symbolized the ills of late nineteenth-century American democracy, as Lincoln Steffens revealed in his landmark work, The Shame of the Cities (1904).

This deplorable situation elicited a positive and constructive response. Everyone recognized the need to stamp out the “virus” of corruption, which left citizens everywhere demoralized. Clearly the need was to unseat the political bosses. How could this be done? The key idea was to make municipal elections nonpartisan, while at the same time establishing procedures to recall elected officials and enable citizens to decide policy through referendums. An early experiment with such reforms took place in Galveston, Texas, in 1901. Subsequently, the movement spread rapidly.

The parties were thus denied control of elections, while at the same time power was vested in a commission with expanded prerogatives and directly responsible to the voters. It became more difficult for party bosses to pull strings than it was back in the old days, when power was dispersed among many departments. The establishment of “government by commission” was only the first step of the reform program, however.47 Whereas the commissions had limited themselves to defining the broad outlines of public policy, city managers, put in place by reformers in a number of cities, actually wielded broad executive powers. These managers were appointed by an elective body but chosen for their supposed professional skills. They were seen as the very incarnation of the “objective power” that many took to be the only hope for the survival of democracy, the only way of ridding it of what many at the time called the “poison of partisan politics”.

Once again, the hope was to promote the general interest by reducing the scope of political power and increasing that of administrative and managerial power.48 Significantly, it was in this period that the neologism technocracy was coined to denote a system of government in which experts organize and control the nation’s resources for the good of all.49 Administrative power was indeed regarded as being in essence substantially democratic.

The idea of an objective power identified with the general interest also took hold in Europe, but there the conditions were different from the United States, as was the scope of the new power. Germany and France are the most interesting cases. The Weimar Republic emphasized the need for a neutral and inviolable administrative sphere. Article 130 of the Constitution of 1919 stipulated that “civil servants are servants of the nation, not of any party”. But the idea remained ambiguous. It was more an extension of the Prussian idea of bureaucracy rather than the expression of a democratic imperative. That is why Carl Schmitt would become such a zealous champion of the objective state, which he saw as a means of combating parliamentarism and the party state.50 He and others celebrated the substantive power of the state as a bulwark against universal suffrage. One finds a similar attitude in France after 1918. The calls for a more rational state reflected lessons drawn from the war rather than a desire to forge a new democratic ideal. The “cult of incompetence,” which Émile Faguet had denounced as early as 1911, was the target of the attacks. With the return of peace, slogans such as “government reform” and “industrialization of the state” gained currency. Henri Fayol, Taylor’s French disciple, published a series of works whose titles epitomize the climate of the time: L’Incapacité industrielle de l’État (1921) and La Doctrine administrative dans l’État (1923). The kind of administrative reform that Fayol had in mind was different from that imagined by the public service theorists of the turn of the century. For the latter, the goal was to create corps of professional civil servants dedicated to their function. After 1918, there was greater emphasis on procedures and organization.51 “Corporatism of the general interest” no longer served as a model. But another difference was that the new reform ideas did not figure within a philosophy of democracy. This also distinguished the post-1918 cult of rationality in Europe from the turn-of-the-century cult of rationality in the United States.

In America, of course, Taylor saw rationalization as essentially a matter of technology and management, but his political success came from the fact that others seized on his ideas as a weapon in the war on corruption and party control of public services. Rationalization was seen as a way of formulating the general interest in objective terms that would make it easier to achieve. This democratic dimension of Taylorism was implicit in the fact that many of the Progressives who celebrated the virtues of efficient and rational administration also favored referendums and popular initiatives. There was nothing comparable in post–World War I Europe: in this respect the French and German cases are typical. The goal was to wipe out incompetence, not corruption.52 Indeed, old prejudices against the masses and wariness of universal suffrage resurfaced at this time. Accordingly, calls for rational administration were often linked to praise of the role of elites and disenchantment with democracy. A striking example of this attitude can be found in the work of Henri Chardon, a member of the Council of State and one of the most ardent early twentieth-century French advocates of administrative power.

In 1911 Chardon published Le Pouvoir administratif.53 Modern societies, he explained, need order and continuity if they are to be governed well. A parliamentary regime is structurally ill-equipped to satisfy these needs because it is divided by partisan conflict and paced by frequent elections. For Chardon, the conclusion is obvious: “The administration must exist in its own right, outside the sphere of politics”.54 Only the administration can satisfy the requirements of permanence and generality necessary for the realization of the common good. Chardon agreed with the public service theorists that civil servants were defined by their “interest in disinterestedness,” but he placed greater emphasis on the technical legitimation of their autonomy: “Each civil servant should be seen not as a person delegated by the minister to provide some public service but as the technical representative of one of the nation’s permanent interests”.55 Indeed, he went so far as to say that the lowliest of civil servants “is himself the government” in the performance of his duties.56 To be sure, political power remains useful and legitimate, but it can play its role only if the legitimacy and independence of administrative power are also recognized. The function of political power must be limited to “sovereign control” of the action of the administration. In Chardon’s view, democracy depends on maintaining a balance between political and administrative power. One is just as important as the other, and each must correct the other. “In a well-constructed system,” he wrote, “the vices of the politicians and those of the administrators neutralize each other. The sovereign control of parliament fosters administrative virtue and efficiency. The vigor of the administration minimizes the inconveniences of elections”.57 As Chardon saw it, the old conflict between opinion and reason, between the masses and the elite, plays itself out anew in the relation between the two powers. Indeed, in celebrating administrative power, he quietly intended to strengthen the influence of the elite.

THE JACOBINS OF EXCELLENCE

The two models we have been discussing—the corporatism of the universal and rational administration—both emerged at the dawn of the twentieth century. It was at the same time that the state—in the generic sense of the term—established itself as a constitutive element of democracy. The full impact of these ideas took time to develop. In the United States, the New Deal marked a turning point in the assertion of administrative power. Nearly everywhere else, the change came only after World War II. Civil servants dedicated to an agenda of modernization then portrayed themselves as the representatives of a new type of legitimacy based on efficiency and competence. The legitimacy they claimed stood in sharp contrast to electoral legitimacy. Over time, their professional culture came to embody both the corporatism of the general interest and the ideal of rational power. To varying degrees in various countries, political power thus found itself, in practice, discreetly disciplined and counterbalanced by this new form of power. Discreetly, because by this point few if any political theorists were working on explicit theories of the new power, as various authors had attempted to do at the turn of the century. Whether in Germany, the United States, or France, there were no more Jellineks or Webers, Goodnows or Wilsons, or Duguits or Chardons seeking to extend democratic theory to institutions of generality other than those that relied on the ballot box as the source of their legitimacy. Doctrinal and political caution carried the day. Hence administrative power, with its corrective and pedagogical components, chose a pragmatic course to ensconce itself within representative regimes. In order to understand the history of democracy in the second half of the twentieth century, one has to take the formative role of these new institutions into account. Indeed, it was administrative power that silently sought to correct many of the deficiencies of electoral-representative regimes.58

The reformers brought into government by John F. Kennedy offer a good example of the phenomenon as it manifested itself in the United States in the 1960s. But the French case offers perhaps an even more striking example of bureaucracy as an agent of modernization. In France, the modernizers drew on a long tradition of state bureaucracies, particularly in technical departments of government: the bureaucracies responsible for mines and roads and bridges had long employed qualified engineers dedicated to serving the public’s needs. Because of this long tradition of providing the popular sovereign with technically competent public servants, France was quick to respond to Keynesian ideas about managing the economy in the post–World War II period.

The debacle of 1940, which put an end to the Third Republic and decimated the old political elite, played an important part in the emergence of an alternative form of democratic legitimacy. The various movements that comprised the Resistance to German occupation repeatedly lam-basted the bankruptcy of the old ruling class and especially the deputies of the National Assembly.59 A new vision of the general interest emerged in contrast to what was stigmatized as “party politics”. No one denied that the parties had a role to play in defining the general will, yet nearly everyone was contemptuous of a regime in which the parties could do no more than express the sum of various special interests. The times lent themselves to favoring substantive notions of the general interest over the merely procedural legitimation of the electoral process, and this created an opening for a group of civil servants who have been characterized as “Jacobins of excellence”.60 At a time when the nascent Fourth Republic exemplified, to the point of caricature, the regime of parties whose deleterious effects everyone deplored, the “Jacobins of excellence” were able to create a relatively independent high civil service. They were men who felt themselves invested with a higher mission. “You joined the administration as you might have taken religious orders, to carry on the fight,” one of them recalled, looking back on the state of mind of the generation that came to power after 1945.61 When they were called on to define themselves, the words that came most naturally to their minds were “vocation of public service,” “mystique of the state,” “servants of the general interest,” and “priesthood”.62

One of the archetypical figures of the group, Simon Nora, expressed the motives and justifications of his generation particularly well: “We were the handsomest, the most intelligent, and the most honest, and legitimacy was ours. Although I’ve just described these sentiments in rather sarcastic terms, it is important to remember that for thirty or forty years the technocratic class thrived on them”.63 Legitimacy was theirs? This is indeed the point of the remark. To be sure, Nora does not challenge the idea that because the political side of government is elected, it must be recognized as preeminent and that the civil service is subordinate to it. But he immediately adds: “Nevertheless, political legitimacy is subject to the rhythms of the electoral cycle. Even the longest terms of office are short compared with the time scale required for dealing with the fundamental issues involved in running a country. … If there were no ‘priests of the long term,’ if there were no one responsible for looking after the structural interests of the nation, beyond the coming and goings of the political class, the country would lack something fundamental”. For Nora, in other words, the top ranks of the civil service are there to watch out for the country’s long-term interests. What justifies this claim? The further claim that civil servants possess two qualities: disinterestedness and rationality.

As we have seen, disinterestedness and rationality are two of the modes in which generality finds expression. The disinterested individual is one who acts as if he were everyman—immediately identified with society as a whole. Social theorists have devoted considerable discussion to the problems posed by this idea, calling attention to its contradictions and equivocations.64 Yet the fact remains that the experience of the war, outside the usual norms, had made it intelligible and cogent for quite a while after the war ended. A person who is prepared to die for his country invests his life in the fate of his community and relegates his own existence to a lesser plane. He begins to identify with “the people” or with “humanity,” as Victor Hugo might have put it. The generation that came of age in the Resistance capitalized on the character traits it had developed in the underground fight. History in a sense vouched for the disinterestedness of this group: they had proven themselves in exceptionally difficult circumstances, hence it was possible to believe that they were new men dedicated to the common good and not needing to give further proof of their devotion.

The other pillar on which their legitimacy rested was the claim of competence. Trained by the new École Nationale d’Administration (established in 1945), they joined the elite ranks of the bureaucracy, where they wielded their knowledge (mainly of economics) as an instrument of power and prestige. Thus their state service was not simply a sacrifice of self akin to entering the priesthood; it was also, in the words of Simon Nora, “a homage to rationality”. Craftsmen and servants of reason, these high civil servants were thus able to press a double claim to legitimacy during the period known in France as “the Thirty Glorious Years” (1945–75), during which they played the role of champions of generality.

COMPETITION AND ELECTION

The corporatism of the universal and rational administration are ideal types. Both have a critical edge, directed against the power of special interests in society. Both share with political democracy an insistence on the idea of equality. Political democracy insists on equality of political voice, whereas administrative democracy insists on equality of opportunity to serve the public. Each imposes a different type of test to select those individuals authorized to serve the general interest: elections in the case of political democracy, competitive examinations in the case of administrative democracy. An election can be defined as a joint expression of qualified wills ending in a choice. A competitive examination is rather an objective selection based on specific criteria. A systematic comparison of these two types of test is therefore an essential key to understanding the difference between the two forms of legitimacy with which they are associated. Much has been written about this distinction since the early nineteenth century. The comments of Édouard Laboulaye and Constantin Pecqueur deserve particular attention.

Pecqueur was one of the founding fathers of French socialism. Marx especially admired him. In the communist society of his dreams, all citizens were to be seen as civil servants.65 In order to take optimal advantage of the talent available in society, Pecqueur envisioned a broad system of competitive examinations: “A fair distribution of social utilities among the members of the association can be achieved only through examinations and competitions”. Hence the power of the people was not limited to choosing who would govern. The people also enjoyed “the right and duty to judge merits, appoint servants, classify individuals, and distribute positions on their own”. The ideal communist society must be governed by a general law of selection and classification. Examinations and elections are different aspects of the same social project of institution and organization. Examinations were to be seen as “scientific or intellectual elections” for the purpose of detecting “relative talent,” or “knowledge, intelligence, and aptitude”. By contrast, voting for representatives is a “civic or political election” for the purpose of identifying a “relative morality” and selecting those with the greatest aptitude for service. Hence both procedures are absolutely necessary for establishing a good society.66

Pecqueur’s contemporary Édouard Laboulaye was one of the great nineteenth-century French political theorists. In the early 1840s the government sent him to Prussia to study how public administration worked there. He produced a voluminous report, several pages of which were devoted to recruitment by competitive examination.67 Laboulaye was interested in the problem of “the political and social organization of democracy”. A jurist and student of German governmental practices, he was one of the few thinkers of his generation to have recognized the importance of the administrative phenomenon. He favored a powerful administration because he believed it was the only real “counterweight to the omnipotence of the Chamber … a counterweight that has been sought, to no avail, in the separation of political powers”.68 But in order to ensure the legitimacy of the administration, he insisted that it must be democratic: “It must establish a firm foundation for itself in the country, and democracy must balance itself by weighing equally on both sides of the scale. In other words, it should manifest itself in the chambers through election and in the administration through competitive examinations”. Though inspired by different philosophies, Laboulaye and Pecqueur thus agreed in seeing an equivalence between the two procedures, elections and examinations.

Their analysis can be extended by comparing the two selection processes. In elections it is the court of public opinion that delivers its verdict. This court draws its members from a natural community of citizens whose existence is given beforehand. In courts of justice, when a jury is constituted, members are drawn by lot to meet the requirements of equality and ability. In a competitive examination, by contrast, the jury is socially constructed. The members have to be nominated, and the criteria of selection must be approved by a social consensus if the procedure is to count as democratic. In other words, society must be institutionally organized in such a way as to sort out the professional qualifications of the examiners as well as the examined if the recruitment process is to be validated through what might be called a “legitimacy feedback loop”. The democratic legitimation of the competitive selection process depends on this.69

The similarity between elections and examinations does not end there, however. Indeed, one can argue that an examination is a specific type of election, one that revives older forms of representative government. First, an examination is like an election in that both are designed to identify qualifications. Elections have always had two related dimensions: on the one hand they symbolize the equality of all citizens, while on the other hand they select certain individuals to serve as leaders. In classical democratic theory, one distinguishes between voting as a right and voting as a function. The two dimensions do not always go together, as the history of universal suffrage shows. What is more, voting as function—the procedure by which leaders are chosen—was long understood in ways that might seem surprising today. For the first modern theorists of representative government such as Madison in the United States and Sieyès in France, elections were in no sense a competition between competing programs or rival personalities. Their only purpose was to identify the best, most qualified candidates. In Madison’s words, elections exist “to obtain for rulers men who possess most wisdom to discern and most virtue to pursue the common good of the society”.70 The required characteristics were thus inextricably moral and intellectual. Intellectual, because governing requires special talents that not everyone possesses. “Citizens,” Sieyès remarks, “choose representatives far more capable than themselves to identify the general interest and interpret their own desires to that end, for the utility of all”.71 But governing also requires specific moral qualities. In this vein, Madison remarks that representatives are “a chosen body of citizens whose wisdom may best discern the true interest of their country and whose patriotism and love of justice will be least likely to sacrifice it to temporary or partial considerations”.72 Elections conceived in this spirit were not narrowly “political”. It is easy to understand why choices were normally expected to be unanimous, since there ought to be no difference of opinion in recognizing talent and qualifications. People firmly believed that the correct choice should be almost self-evident. No one imagined that merit and virtue would not be recognized universally and spontaneously.

In this intellectual universe, the competitive dimension of voting had no place. Elections seen in this light had what might be called an “objective” character. They did not imply a need for contradictory arguments or partisan choice. The goal was to choose a person, not to choose between different options.73 Seen in this way, the purpose of elections was to make an “exemplary distinction”. It was to draw out the intellectual and moral essence of the society, to sketch its idealized portrait. By revealing “general individuals,” elections contributed directly to the realization of the common good. Hence they held considerable potential for legitimating government, because the quality and significance of government action were totally identified with the individuals selected. To the extent that competitive examinations were designed to make the same types of distinctions, they can be seen as having a function equivalent to that of “pure elections”.74

This functional idealization of elections did not survive the reality of politically opposed projects and candidates, which came to be seen as the norm, but the idea of recruitment through competitive examinations in a sense revived the lost ideal. It gave new life to the hope of achieving an objective, unanimous selection. The competitive examination corresponded to the original conception of the functional vote as described by Sieyès and Madison. It was a way of fulfilling the republican promise by simultaneously respecting the principle of equality and honoring the need for superiority, which was to be identified in a nonexclusive way based on individualized criteria. The qualities identified through examinations were indeed radically individual, so that they could not be appropriated by any particular group. Hence the individuals selected constituted a new type of elite. It was not a caste or class but an almost randomly selected group, whose composition could change from moment to moment. Examinations were thus a nondiscriminatory, purely functional means of making distinctions and therefore a procedure beneficial to all and the precise opposite of a privilege. The “republic of exams” was therefore perfectly compatible with the republic of universal suffrage.75

The form of the competitive examination also revived a “cognitive” dimension of the representative process. Early nineteenth-century liberals had stressed this to justify their doubts about universal suffrage. François Guizot, in particular, provided theoretical justification for this view. As he put it, the goal of representative government is

to discover all the elements of legitimate power that are scattered throughout society and organize them into an actual power, or, in other words, concentrate them, in order to give reality to public reason and public morality and summon those scattered elements to power. What is called representation is nothing other than a means for achieving this result. It is not a numerical machine for counting up individual wills. It is a natural procedure for distilling from the bosom of society public reason, which alone is entitled to govern.76

At about the same time, Jean Charles Sismondi observed that “representative government is a welcome invention for discovering a nation’s eminent men”.77 Since elections had not proved themselves capable of achieving “public reason,” it was possible to present competitive examinations as a more appropriate means of achieving the same end.

When procedures and goals initially associated with elections were subsequently linked to examinations, a new history became possible. It also became possible to see the relation between two forms of legitimacy—that of establishment and that of identification—in a more ample context. And finally, it became possible to understand the conflicts of legitimacy that had for a time blocked the introduction of competitive examinations as a tool for civil service recruitment. Examinations had in effect been understood as a potential competitor of elections. Under the July Monarchy, any number of French politicians objected that “ministerial responsibility becomes illusory if civil servants are hired solely on the basis of competitive examinations”.78 Officials selected by examination were long perceived as threats by those elected by popular suffrage, whose legitimacy stemmed from a different kind of trial. That is why France delayed until 1945 the creation of a national civil service academy, the École Nationale d’Administration, which was first proposed at the time of the Revolution.79 At the same time, liberal economists also opposed the examination system on the grounds that it would reduce social mobility and revive the guild system in the form of a public mandarinate.80 We have seen how the history of elections was in fact related to the history of examinations and how the problems and ambiguities of both types of testing illuminate one another.

Finally, the parallel between elections and examinations opens the way to a broader comparative approach to modes of production of generality. The political importance of examinations in classical Chinese culture stands out when looked at in this light. Indeed, this was the way in which the battle against aristocracy was waged in China and a certain sense of equality was established.81 When Sun Yat-Sen tried to describe the distinctive Chinese route to a modern constitutional order, he stressed what he called the “power of examination,” to which he assigned the same importance as the “right of election” and the “right of referendum”.82

Thus in order to study examinations and elections comparatively, both have to be seen in the broader context of tests of generality. This kinship also enables us to understand the systematic relationship that existed between the legitimacy of establishment and the legitimacy of identification with generality.

1 Under the spoils system the party that won the presidential election claimed the right to remove all federal officials from office and appoint replacements of its own choosing.

2 See my discussion of this point in L’État en France: de 1789 à nos jours (Paris: Seuil, 1990), and The Demands of Liberty, trans. Arthur Goldhammer (Cambridge, MA: Harvard University Press, 2007).

3 Pierre-Paul Royer-Collard, quoted in Guy Thuillier, Témoins de l’administration, de Saint-Just à Marx (Paris: Berger-Levrault, 1967), p. 29.

4 The first book to take a different view of this issue was Auguste Vivien, Études administrative, 2d ed., 2 vols. (Paris, 1852).

5 Édouard Laboulaye, “De l’enseignement et du noviciat administratif en Allemagne,” Revue de législation et de jurisprudence, vol. 18, July–December 1843.

6 French distinguishes between the examen and the concours. The former is a test with a fixed passing grade; all who exceed this grade are deemed acceptable. By contrast, the latter fixes in advance the number of candidates who will be accepted, and the performances of the various candidates are ranked. Candidates are selected until the preestablished quota is filled, no matter how well the lower-ranked candidates have performed. Hereafter, “competitive examination” will be used to translate both terms.—Trans.

7 Quoted in Paul Bastid, Un juriste pamphlétaire: Cormenin (Paris, 1948).

8 For an overview of the organization of the civil service, see Françoise Dreyfus, L’Invention de la bureaucratie: Servir l’État en France, en Grande-Bretagne et aux États-Unis (XVIIIe–XXe siècles) (Paris: La Découverte, 2000), and the data contained in L’État en France, de 1789 à nos jours.

9 Report submitted in 1886 by Théodore Steeg on behalf of the Education Committee of the Chamber of Deputies.

10 See the work by several authors, Les épurations administratives, XIXe et XXe siècles (Geneva: Droz, 1977).

11 See “L’épuration de 1879,” in Le Conseil d’État, 1799–1974 (Paris: Éditions du CNRS, 1974).

12 Armand Charpentier, Le Parti radical et radical-socialiste à travers ses congrès (1901– 1991) (Paris, 1913), chap. 13, “Les fonctionnaires”.

13 Richard L. McCormick, “The Discovery that Business Corrupts Politics: A Reappraisal of the Origins of Progressivism,” in Kristofer Allerfeldt, ed., The Progressive Era in the USA, 1890–1921 (Abingdon, UK: Ashgate, 2007).

14 For the sake of completeness, one should also mention a third approach: that of German jurists. Gerber, as early as 1865, and later Jellinek developed an important theory of the moral personality of the state (the two volumes of Jellinek’s The Modern State and Its Law appeared in 1900). The German perspective was not the same as what we find in France or the United States, however. The goal was more limited: to justify a distinction between the state and the monarch, at a time when the monarchy still justified itself in terms of a patrimonial theory of the state. Concerning these doctrines and their reception in France, see the essays in Olivier Beaud and Patrick Wachsmann, La Science juridique française et la Science juridique allemande de 1870 à 1918 (Strasbourg: Presses Universitaires de Strasbourg, 1997). Later, in the early 1920s, Carl Schmitt’s arguments in favor of a “neutral state” to counter the influence of the political parties may be taken as a further distinctive contribution to this tradition.

15 I borrow the expression used for the title of this section from Pierre Bourdieu, Les Règles de l’art: Genèse et structure du champ littéraire (Paris: Seuil, 1992), “Post-Scriptum: Pour un corporatisme de l’universel,” but apply it to a milieu other than that of intellectuals.

16 Léon Duguit, Les Transformations du droit public (Paris, 1925), chap. 1.

17 Preface to Léon Duguit, Traité de droit constitutionnel, vol. 1, 2d ed. (1920; reprint Paris, 1927), p. x. The first formulations of Duguit’s theories can be found in L’État, le Droit objectif et la Loi positive (1901), and L’État, les Gouvernants et les Agents (1903). Dalloz published both works in reprint in 2003, with a preface by Franck Moderne.

18 Léon Duguit, La Théorie générale de l’État (Paris: A. Fontemoing, 1907), vol. 2, p. 59.

19 Ibid., p. 62.

20 Ibid., p. 61.

21 Duguit, Les Transformations du droit public, p. 52.

22 Duguit, Traité de droit constitutionnel, vol. 1, p. ix.

23 A good introduction to the movement can be found in Georges Gurvitch, L’Idée du droit social (Paris, 1932). More recently, see the very illuminating work of H. Stuart Jones, The French State in Question: Public Law and Political Argument in the Third Republic (Cambridge: Cambridge University Press, 1993). The work of Maurice Hauriou, the other great theorist of the public service, should also be looked at in this context. He too understood power primarily in terms of its function in creating order and solidarity, and he drew from this the same consequence as Duguit, namely, that the legitimacy of power derives from its ability to fulfill its function appropriately. See his Principes de droit public (Paris, 1910).

24 Harold J. Laski, “La conception de l’État de Léon Duguit,” Archives de philosophie du droit et de sociologie juridique, nos. 1–2, 1932.

25 Léon Duguit, L’État, les Gouvernants et les Agents (Paris, 1903), p. 413.

26 Duguit, Traité de droit constitutionnel, vol. 2, pp. 67–68.

27 M. Hauriou, Précis de droit administratif et de droit public, 6th ed. (Paris, 1907), p. 60.

28 Hauriou: “Even when civil servants are subordinate to the government, they are subordinate only to a certain extent. In some respects, owing to the autonomy of their function, they are considered to be collaborators of the government and therefore in a management situation”. Ibid., p. 61.

29 Jacques and Mona Ozouf, La République des instituteurs (Paris: Gallimard-Seuil, 1992), pp. 68–73.

30 Memorandum, September 20, 1887, quoted in Maxime Leroy, Syndicats et services publics (Paris, 1909), p. 251.

31 Duguit, Traité de droit constitutionnel, vol. 2, pp. 66–68, and vol. 3, pp. 89–103.

32 Ibid., vol. 3, p. 97.

33 Hauriou, Précis de droit administratif et de droit public, pp. 560–562.

34 Traité de droit constitutionnel, vol. 3, p. 110.

35 Ibid., p. 163.

36 Émile Durkheim, Leçons de sociologie: Physique des moeurs et du droit (1898–1900) (Paris: PUF, 1950), p. 61. On this point see Pierre Birnbaum, “La conception durkheimienne de l’État: l’apolitisme des fonctionnaires,” Revue française de sociologie, April–June, 1976.

37 Georg Wilhelm Friedrich Hegel, Outlines of the Philosophy of Right (Oxford: Oxford University Press, 2008), sec. 303.

38 Woodrow Wilson, “The Study of Administration,” Political Science Quarterly, vol. 2, no. 2, June 1887.

39 Bear in mind that in the United States the word administration can refer to both the government and the federal bureaucracy.

40 On the actual impact of Wilson’s article at the time of its publication, see Daniel W. Martin, “The Fading Legacy of Woodrow Wilson,” Public Administration Review, vol. 48, no. 2, March–April 1988; and Paul P. Van Riper, “The American Administrative State: Wilson and the Founders: An Unorthodox View,” Public Administration Review, vol. 43, no. 6, November–December 1983.

41 Samuel C. Patterson, “Remembering Frank J. Goodnow,” PS: Political Science and Politics, vol. 34, December 2001.

42 There is a recent edition of the work: Frank Goodnow, Politics and Administration: A Study in Government (New Brunswick, NJ: Transaction, 2003), with an introduction by John A. Rohr.

43 On this conceptual shift, see Morton White, Social Thought in America: The Revolt against Formalism (1949; reprint New York: Oxford University Press, 1976), and Edward A. Purcell, The Crisis of Democratic Theory: Scientific Naturalism and the Problem of Value (Lexington: University of Kentucky Press, 1973).

44 Men like Wilson and Goodnow who made such recommendations were strongly attracted to what they perceived as the “European continental model” in this regard. They often referred to France and Prussia in their writing.

45 On this movement the key works are Samuel Haber, Efficiency and Uplift: Scientific Management and the Progressive Era, 1890–1920 (Chicago: University of Chicago Press, 1964); Samuel P. Hays, Conservation and the Gospel of Efficiency: The Progressive Conservation Movement, 1890–1920 (1959; reprint New York: Atheneum, 1969); Robert H. Wiebe, The Search for Order, 1877–1920 (1967; reprint Westport, CT: Greenwood Press, 1980); and Judith A. Merkle, Management and Ideology: The Legacy of the International Scientific Management Movement (Berkeley: University of California Press, 1980).

46 Stephen Skowronek, Building a New American State: The Expansion of National Administrative Capacities, 1877–1920 (Cambridge: Cambridge University Press, 1982), pp. 177ff.

47 For contemporary views of these developments, see John J. Hamilton, Government by Commission or the Dethronement of the City Boss (New York, 1911); Clinton Rogers Woodruff, ed., City Government by Commission (New York, 1911); and the very comprehensive anthology published by the American Academy of Political and Social Science, Commission Government in American Cities (Philadelphia, 1911). For a more recent treatment, see Bradley Robert Rice, Progressive Cities: The Commission Government Movement in America, 1901–1920 (Austin: University of Texas Press, 1977).

48 For an early evaluation of this system, see Harold A. Stone, Don K. Price, and Kathryn H. Stone, City Manager Government in the United States: A Review after Twenty-Five Years (Chicago: University of Chicago Press, 1940). The best recent study is Martin J. Schiesl, The Politics of Efficiency: Municipal Administration and Reform in America, 1800–1920 (Berkeley: University of California Press, 1977).

49 The term seems to have been coined in 1919. See Raoul de Roussy de Sales, “Un mouvement nouveau aux États-Unis: la technocratie,” Revue de Paris, March 15, 1933.

50 For Schmitt’s defense of the neutral state and opposition to the party state, see Olivier Beaud, Les Derniers Jours de Weimar: Carl Schmitt et l’avènement du nazisme (Paris: Descartes et Cie, 1997), pp. 50–72.

51 For an overview, see Stéphane Rials, Administration et organization, 1910–1930: De l’organisation de la bataille à la bataille de l’organisation dans l’administration française (Paris: Beauchesne, 1977). On the practical impact of these ideas, see Alain Chatriot, “Fayol, les fayoliens et l’impossible réforme de l’Administration durant l’entre-deux-guerres,” Entreprises et histoire, no. 34, December 2003.

52 See the emblematic work of Joseph Barthélemy, Le Problème de la compétence dans la démocratie (Paris, 1918).

53 He had made a previous effort to confront the issue in L’Administration de la France. Les fonctionnaires (1908).

54 Henri Chardon, Le Pouvoir administratif (Paris, 1911), p. 29: “The public service is permanent and necessary, while nothing is more fickle and in many cases more futile than political judgments”. Ibid., p. 11.

55 Ibid., p. 55.

56 Ibid., p. 191. “At that moment, each civil servant is, within the limits of his duties, superior to any other authority”.

57 H. Chardon, Les Deux Forces: Le nombre, l’élite (Paris, 1921), pp. 13–14.

58 To complete this picture, one should also draw attention to the overtly antidemocratic uses to which the legitimacy of identification with the general interest was sometimes put. In Asia and Latin America, for example, military “modernizers” often justified coups d’état by emphasizing the goal of establishing a government to serve the general interest in the stead of a civilian government accused of corruption. Note, too, that communist regimes invariably described themselves as expressions of “true democracy,” that is, power identified with the public good.

59 For this point I rely on the articles and manifestos cited in Henri Michel, Les Courants de pensée de la Résistance (Paris: PUF, 1962), pp. 359–366.

60 I borrow the phrase from Jean-Pierre Rioux’s “Prologue” to François Boch-Lainé and Jean Bouvier, La France restaurée, 1944–1954 (Paris: Fayard, 1986), p. 26.

61 The comment is Simon Nora’s, reported in François Fourquet, Les Comptes de la puissance: Histoire de la comptabilité nationale et du Plan (Paris: Encres, 1980). This book, consisting mainly of interviews, offers an excellent introduction to this generation’s motives and methods.

62 I take these phrases from a typical representative of this group, François Bloch-Lainé, Profession: fonctionnaire (Paris: Seuil, 1976).

63 Simon Nora, “Servir l’État,” Le Débat, no. 40, May–September 1986, p. 102. Subsequent quotes are taken from this article.

64 See in particular the work of Jon Elster and Pierre Bourdieu on this point.

65 See esp. La Théorie nouvelle d’économie sociale et politique (Paris, 1842). The distinction between elections and examinations is discussed on pp. 576–585. All the quotes in the text are from these pages.

66 Constantin Pecqueur notes: “The choice among candidates would be greatly facilitated if the mixed mode of election and classification were applied to all spheres, as required by examinations, competitions, and the granting of degrees”. Ibid., p. 362. This theme can also be found in the Fourierist literature on the specific nature of societal election, which was supposed to ensure perfect harmony between democratic choice and the identification of talent. See, for example, Félix Cantagrel, Le Fou du Palais-Royal (1845; reprint Paris: Fayard, 1984), pp. 364–365.

67 Laboulaye, “De l’enseignement et du noviciat administratif en Allemagne”.

68 Ibid., p. 528, for this and the next quote.

69 The composition of juries of examiners is thus crucial. Laboulaye insisted that university professors, whom he saw as the repositories of “objective knowledge,” occupy the top positions.

70 James Madison, The Federalist, no. 57 (Cambridge, MA: Harvard University Press, 1961).

71 Sieyès, “Dire sur la question du veto royal,” Paris, September 7, 1789, p. 14.

72 Madison, The Federalist, no. 10.

73 Recall that in France, during the Revolution, there were no organized candidacies. To ask for other people’s votes was seen as a personal claim of qualification, an insistence on one’s own superiority, and thus a sign of suspect ambitions. In short, a candidacy involved a claim of distinction that was perceived as a sign of suspect aristocratic ambitions. To reject candidacies was thus initially “democratic,” at least in part. On this important point, see my Le Peuple introuvable, pp. 43–49.

74 The “general individual” chosen by an election should therefore be distinguished from the random individual selected by drawing of lots. Equality also figures in the latter choice, as does the desire to keep choice separate from competition. But the primary characteristic of the individual chosen by lot is randomness. This is what is sought in selecting a jury of the people, for example. The goal is to validate the expression of an immediate form of generality, which is reinforced by the notion of proximity in jury selection. By contrast, an ideal election seeks to reveal what might be called “exemplary generality”. For French revolutionaries in 1789, there were thus two complementary ways of conceiving of the democratic institution of generality. The elected representative and the juror both embodied the general interest that the revolutionaries wished to empower.

75 This analysis explains why the revelation of sociological biases in the outcome of examinations (by Pierre Bourdieu, for example) had such a powerful impact in undermining their democratic legitimacy, in conjunction with other critiques of dysfunctions in the representative system.

76 François Guizot, Histoire des origines du gouvernement représentatif (1821; reprint Paris, 1851), vol. 2, p. 150.

77 Sismondi, Études sur la constitution des peuples libres (Brussels, 1836), p. 51.

78 The argument is quoted and analyzed by Laboulaye in “De l’enseignement et du noviciat administratif en Allemagne,” p. 590.

79 G. Thuillier, L’ENA avant l’ENA (Paris: PUF, 1983). On parliamentary doubts about competitive examinations, see also G. Thuillier, Bureaucratie et bureaucrates en France au XIXe siècle (Geneva: Droz, 1980).

80 See, for example, Jean-Gustave Courcelle-Seneuil, “Études sur le mandarinat française,” in La Société moderne: Études morales et politiques (Paris, 1892), pp. 356–384.

81 See Jacques Gernet, “Organisation, principes et pratique de l’administration chinoise (XIe–XIXe siècles),” in F. Bloch-Lainé and Gilbert Étienne, eds., Servir l’État (Paris: Editions de l’EHESS, 1987).

82 Sun Yat-Sen, “Constitution des Cinq Pouvoirs,” appendix to Souvenirs d’un révolutionnaire chinois (1925; reprint Paris, 1933). See also Kong-Chin-Tsong, “La Constitution des Cinq Pouvoirs: Théorie et application” (Ph.D. diss., Paris Faculty of Law 1932).