Skip to navigation

Skip to main content


Ideas, Institutions, and Civil Society: On the Limits of Immigration Control in France -- James F. Hollifield

Ideas, Institutions, and Civil Society:

On the Limits of Immigration Control in France*

James F. Hollifield

Arnold Professor of Political Economy

Director of International Studies

Southern Methodist University, Dallas, Texas

*Paper prepared (May-June 1997) for a workshop on immigration control in Europe, University of Bologna, Italy, April 1997. Revised, December 1997.


Introduction: Immigration and the Republican Tradition..............................................3

Stopping Immigration......................................................................................................5

Crises of Control and National Identity..........................................................................11

From a "Threshold of Tolerance" to "Zero Immigration"..............................................18

Limiting Rights: Negative versus Positive Freedom......................................................20

Civil Disobedience and the Limits of Control................................................................26

A New "Republican Pact"..............................................................................................31




Introduction: Immigration and the Republican Tradition

Unlike other European states, France has a long history of immigration, dating back at least to the middle of the nineteenth century when industrialization began in earnest. Yet France was not the only European state compelled to import labor to feed the fires of industrialization. Despite long histories of emigration, both Britain and Germany imported labor. British industry relied upon Irish workers throughout the nineteenth century, whereas Germany brought in Poles and others from East Central Europe to work in mining and heavy industry in the Ruhrgebiet, especially during the period of intensive economic development in the 1880s and 1890s. What distinguishes France from many other European states, however, is the early willingness to accept foreigners as settlers, immigrants, and even as citizens. The acceptance of foreigners as potential citizens is part and parcel of what can be called a republican tradition, which stems from the French Revolution at the end of the eighteenth century. Republicanism is strongly egalitarian, anti-clerical (laïque) and opposed to monarchy. It stresses popular sovereignty, citizenship, and the rights of man. It can be nationalist and imperialist, while at the same time stressing universal political values, such as equal protection of all individuals before the law. Republicanism, as an ideology and a form of government, was bitterly contested in France throughout the nineteenth and into the twentieth century.

Even though France has a long tradition of immigration and was the first European state to grant citizenship to Jews, at the time of the Revolution, it was not until the culmination of the Dreyfus affair early in the twentieth century, under the Third Republic, that the main tenets of republicanism--laïcité or separation of church and state, equal protection of all before the law, a universalist conception of human rights, and popular sovereignty--were finally accepted by a majority of the French people. It was also during this period around the turn of the century that the French State began to lay the legal foundations for citizenship and naturalization, which would be based on the birthright principle of soil (jus soli) rather than exclusively on blood (jus sanguinis) as in Germany. Thus the republican tradition found its expression in a more open and expansive notion of citizenship, similar (but not identical) to the birthright principle enunciated in the Fourteenth Amendment of the U.S. Constitution ("All persons born or naturalized in the United States, and subject to the jurisdiction thereof, are citizens of the United States") and in stark contrast to the more narrow, ethnocultural vision of citizenship evolving in Germany of the Second Reich. While Germany was struggling with the issues of national and territorial unification and would continue to do so--one could argue--until 1989-90, France was becoming more comfortable with its revolutionary and republican heritage, as reflected in an increasingly expansive policy of immigration and naturalization.

In contrast with the United States--the other great republic also founded at the end of the eighteenth century--France was not a nation of immigrants. The first period of intensive immigration in France did not begin until the 1850s, long after the Revolution of 1789. Hence immigration in France was never part of any type of founding myth, as it was (and still is) in the United States. Even though immigration and integration are closely associated with the French republican tradition, they are not crucial to French national identity, except for French Jews for whom the Revolution represents political and legal emancipation. Sustaining an open and legal immigration policy is more difficult to do in France than in the United States, but easier than in Germany, which also has a republican tradition, albeit a young one dating from the founding of the Bundesrepublik in 1949.

For much of the post-World War II period, French governments of the Fourth and Fifth Republics pursued expansive immigration policies, essentially for three reasons. The first justification--as can be seen in the various five-year Plans--was primarily economic. During the period of reconstruction of the 1950s and 1960s (sometimes referred to as the trente glorieuses, or thirty glorious years of economic growth and low unemployment), France, like Germany, was in desperate need of labor. The second rationale for an open and legal immigration policy was the longstanding desire to boost the French population. Having gone through its demographic transition much earlier than other industrial societies, France was believed to have a huge demographic deficit and immigration was seen as one way to overcome this weakness. Finally, as I have argued elsewhere, policy makers and politicians had great confidence in the ability of French society to absorb and integrate the newcomers, because of the strength of the republican tradition. Therefore, an expansive, legal immigration policy was coupled with the most liberal naturalization policy in Europe, quite similar in many ways to that of the United States.

Stopping Immigration

The consensus for an open immigration regime held until the early 1970s, when the trente glorieuses abruptly ended in 1973-74 with the first big recessions of the postwar period. Moreover, decolonization and the granting of independence to Algeria in 1962 radically altered the ethnic composition of immigrant flows, as North African Muslims rapidly replaced the largely Catholic flows from Italy, Spain, and (eventually) Portugal. Many in France began to question the strength of the republican model and its ability to assimilate the new Islamic populations coming from the Maghreb. But stopping immigration (l'arrêt de l'immigration) would prove difficult, because the mechanisms and instruments of control had not yet been developed by the French state, and cutting ties with former African colonies would not be easy.

Throughout the 1960s and into the 1970s, the recruitment of foreign workers was carried out largely by the private sector, as had been the case during the 1920s. Businesses recruited directly in the sending countries, bringing their workers to France, training them, then filing a request with the Office National d'Immigration for an adjustment of status (régularisation). This mode of immigration, which bypassed both external (border) and internal (labor market) controls, came to be known as immigration from within (immigration interne). The rate of adjustments of status (taux de régularisation), basically a ratio of immigration processed by ONI in France to that processed abroad, became a standard measure of immigration control, or lack thereof, during the 1960s and 1970s. The rate (or ratio) reached 90 percent by the late 1960s, indicating the extent to which authorities had lost control of immigration. Reestablishing control over the flows of worker immigration (travailleurs permanents) would take many years. Simply decreeing an immigration stop proved insufficient to master all the different flows (workers, family members, seasonals, and eventually refugees).

Apart from worker immigration, French authorities, like their German counterparts, also struggled to deter family immigration, which remained at fairly high levels (over 50,000 per year from 1974-80), even after the immigration stop imposed in 1974 (see Table 1). The justification for stopping worker immigration was clear in both countries: with the decline in economic growth and the rise of unemployment--especially in France--employers should no longer be allowed to recruit foreign labor, and the denial of visas (external control) and work permits (internal control) was seen as a necessary policy response to worsening economic conditions; and as a way to head off a rising tide of xenophobia, which was increasingly evident in France, but also in neighboring countries like Switzerland and Great Britain. This shift in policy reflected a widespread Malthusian impulse in Europe: if only the receiving states can stop immigration, it will solve the problem of unemployment, because there is a limited number of jobs in each national economy; and it will arrest xenophobic tendencies among the electorate.

(Table 1, here)

Notwithstanding this political and Malthusian impulse in France, one of the principal sending countries, Algeria, took steps in 1973 to prevent the free emigration of its nationals to the former métropole, because of the growing hostility towards Algerians in France. Yet immigration continued throughout the 1970s, in large part due to increases in family reunification, which was much more difficult to control. The economic rationale for stopping worker immigration did not apply to family immigration, which was deemed to be more humanitarian and social than economic. Still, French and German authorities tried to impose internal (labor market) controls to slow the influx of family members, by denying them work permits. In both instances, however, the courts ruled these policies to be illegal and/or unconstitutional. The French also had to cope with the continued inflow of so-called seasonal workers (saisonniers), employed primarily in agriculture. From 1974-87, the number of seasonal workers entering each year hovered around 60,000 (see Table 1). Some of these workers came from Spain and Portugal, and the enlargement of the European Community and the extension of the freedom of movement clause of the Rome Treaty to cover Spanish and Portuguese nationals partially resolved the issue of seasonal migration. In the late 1970's and early 1980's, Moroccans made up the bulk of seasonal flows, whereas in the 1990's North Africans were replaced by East Europeans, especially Poles.

What was happening during the presidency of Valérie Giscard d'Estaing (1974-81) was a radical shift away from the open immigration regime of the earlier Gaullist years towards a more closed regime. The methods used to achieve this objective (l'arrêt de l'immigration) were heavy-handed and statist--consistent with the centralized jacobin state--and they produced many unintended consequences. The most important consequence, which was certainly not unique to France, was to freeze the foreign population in place. By simply decreeing an immigration stop, France like other West European states, inadvertently accelerated the processes of settlement and family reunification.

Switzerland offers an interesting point of comparison in this regard. Swiss authorities, unlike the French and Germans, kept in place their guestworker policies of rotation, even though they took steps to limit employers access to foreign labor. But barriers to settlement and naturalization of foreigners in the Conféderation Hélvetique remained high, whereas in France liberal naturalization polices were not changed and the legal status of Algerians was such that it prevented the state from excluding them. Until 1973, Algerians were guaranteed freedom of movement under the Evian Agreements, which ended the Algerian conflict in 1962. Moreover, second generation Franco-Algerians acquired citizenship according the principle of double jus soli: Algerian children were born in France of parents also born in France. It was therefore impossible for French authorities legally to exclude first or second generation Franco-Algerians; and having raised expectations among the French public that the jacobin state could simply decree a halt to immigration, the government of Giscard d'Estaing found that its hands were tied both by the law and by the virtually uncontrollable effects of chain migration. This did not prevent both the Chirac (1974-76) and Barre (1976-81) governments form making covert efforts to stop family reunification by denying visas and deporting family members. The Barre government also tried overt policies to encourage return migration by paying foreigners to leave.

In Germany, the state was caught in a similar predicament, but with more profound repercussions for German politics and society. As in France, the German federal state discovered that it could not simply decree a halt to immigration, nor was it able to entice foreigners to return voluntarily. But unlike France, Germany had no tradition of immigration and its nationality law, dating from 1913, did not allow for quick naturalization of a permanent foreign population. The republican tradition in Germany has little to say about foreigners, with the important exception of refugees who were covered until 1993 by Article 16 of the Basic Law. German political culture has not been strong enough to incorporate de facto immigration, whereas in France the problem of integrating the foreign population has been solved de jure--at least for the moment--by liberal naturalization policies. As we shall see, however, attempts would be made by subsequent right-wing governments under Chirac (in 1986) and again under Balladur (in 1993) to change French nationality law in order to discourage the naturalization of immigrant children. Here again, however, right-wing governments would be frustrated by their inability to exclude Algerian children, who continued to benefit from the double jus soli rule (see above). This type of psychological warfare against immigrants would be used increasingly by a range of OECD states, including the United States in the 1990s, in an attempt to stop immigration. But in France and elsewhere in Western Europe, the principal outcome of the statist efforts in the 1970s to halt immigration was to accelerate the permanent settlement and naturalization of foreigners in the 1980s. The immigrant population in France as a percentage of the total population has remained virtually unchanged from 1975 to 1990 (7.5%), and it has grown in absolute terms to well over 4 million, according to the 1990 census (see Table 2). The same thing has happened in the U.S. in the 1990s.

(Table 2, here)

While the issue of control (immigration policy) would continue to be hotly debated in France throughout the 1980s and into the 1990s, the issue of integration (immigrant policy) surged onto the national agenda. The realization that millions of North Africans were settling permanently in France led governments and political parties to reconsider their approach to immigration and integration. Political parties, the party system, and the electorate were increasingly polarized on both issues. The election in 1981 of a socialist President, François Mitterrand, and the first truly left-wing government since the Popular Front of 1936 set the stage for some important policy shifts, which can be described as a kind of liberal tradeoff, or what some analysts have called a "grand bargain".

The socialists decided to maintain tight (external) control of borders and stepped up (internal) control of the labor market to inhibit the development of a black market for undocumented workers (travail au noir). Regulation of the labor market was easy enough to accomplish by the use of Inspecteurs du travail, who could make snap visits to firms and impose sanctions on employers caught using undocumented workers. But at the same time the socialist government, led by Prime Minister Pierre Mauroy, offered a conditional amnesty to undocumented immigrants and longer (ten-year) residency and work permits for all immigrants. Anyone who had entered France prior to January 1, 1981 was eligible for a temporary residency permit, valid for three months, which would give the individual time to complete an application for an adjustment of status (régularisation exceptionnelle). By the end of the amnesty period (in 1983), over 145,000 applications had been received.

The idea behind this "liberal policy" has been accepted by many high-immigration countries as the best compromise in the battle to control immigration. The U.S. also enacted an amnesty in 1986. In a liberal and republican polity, strict control of entries together with an amnesty for illegals came to be seen as a good way to integrate permanent resident aliens, or as Tomas Hammar calls them, denizens. In addition to the amnesty, to make foreigners residing in France more secure, the first socialist government under Mauroy (1981-84) relaxed prohibitions against associational and political activities by foreigners. The civil liberties of foreigners were protected by prohibiting the police from making arbitrary identity checks of foreign-looking individuals. But no changes were made in the nationality law or in naturalization policy, leaving this key element of the republican tradition intact. Foreigners would be welcome within strict guidelines of labor market rules and regulations; they would be integrated on the (republican) basis of respect for the separation between church and state (laïcité); and they would quickly assimilate.

Having thus reaffirmed the previous right-wing governments' commitment to strict immigration control, while at the same time taking steps to speed the integration of foreigners in French society, the socialists, it seemed, were forging a new consensus on the contentious immigration issue. But the issue literally exploded in everyone's face (on the left as well as the right) in 1984 with the municipal elections in the city of Dreux, an industrial town just west of Paris. The Front National--a grouping of extreme right-wing movements under the charismatic and flamboyant leadership of Jean-Marie Le Pen--won control of this city, on a platform calling for a complete halt to immigration and for the deportation of African immigrants. The electoral breakthrough of a neofascist, xenophobic, and racist movement profoundly changed the politics of immigration, not only in France but throughout Western Europe. For the first time since the end of the Second World War, an extremist party of the right was making itself heard and finding a new legitimacy, garnering support from large segments of the French electorate across the political spectrum. Within a matter of years, it would become, in the words of the political analyst Pascal Perrineau, "the largest working class party in France." From the beginning, the Front National was a single-issue party, taking a stand against immigration, and its leader, Le Pen, called for a physical separation of the races. His discourse mixes xenophobia, extreme nationalism (La France aux français), and anti-Semitism, with appeals to the economic insecurities of the French working class. How did the breakthrough of the Front National affect French immigration policy and the republican consensus?

Crises of Control and National Identity

The rise of the Front National contributed heavily to a sense of crisis in French politics and public policy, with immigrants at the center of the maelstrom. Suddenly immigrants were seen as the cause of the economic and cultural decline of the French nation, provoking a loss of confidence in the republican model, especially on the right. Immigrants were accused of taking jobs away from French citizens, thereby contributing to high levels of unemployment, and Muslims were deemed to be inassimilable and hostile to republican values. The socialist left, under President Mitterrand, and the neo-Gaullist/liberal right (RPR-UDF), led by Jacques Chirac, had very different responses to the Front National's populist appeals to economic insecurity and xenophobia.

Mitterrand added to his mystique as a Machiavellian politician, as he cynically maneuvered to exploit the rise of the Front National for political gain. From his perspective, not only did the Front National divide the right-wing electorate, but by getting many working class votes, it also weakened the Communist Party, another traditional adversary of the socialists. Yet on a liberal note, in 1984 following the elections in Dreux, Mitterrand called for granting voting rights to immigrants in local elections, thereby forcing the parties of the traditional right (RPR-UDF) to take a stand on immigration and immigrant policy. But then, on a more cynical and Machiavellian note, the socialist government led by President Mitterrand and Prime Minister Laurent Fabius, changed the electoral system from a majoritarian, single member district system to one based on proportional representation, just in time for the legislative elections of 1986. The immediate effect of this rule change was to reduce the magnitude of the inevitable victory of the right and to allow the Front National, with roughly 10 percent of the vote, to gain 32 seats in the new Assembly. For the first time since Vichy, the extreme right had representation in parliament, and a new debate over French national identity was under way.

The first step of the right-wing coalition of neo-Gaullists (RPR) and liberals (UDF), led by Chirac, was to change the electoral system back to the traditional Fifth Republic dual ballot system with single member districts. Under the old system, it would be nearly impossible for the Front National to win seats in future elections. But the damage to the right had already been done, and the task remained of recapturing Front National voters. To accomplish this, the government set about reforming immigration and naturalization policy, handing the entire dossier to the tough Corsican Minister of the Interior, Charles Pasqua, whose name would become synonymous with immigration reform over the next decade. Pasqua's approach to immigration control was quite different than any of his predecessors. As Minister of the Interior, he viewed control primarily as a police matter, so he moved quickly to reinforce border controls by giving sweeping new powers to the Police de l'Air et des Frontières (PAF) to detain and immediately deport anyone who did not have proper papers. He also reinforced the power of the (internal) police forces to conduct random (and arbitrary) identity checks of any foreign or suspicious-looking individual. It was also during this time (1986) that there was a wave of terrorist bombings in Paris, connected to the Middle East, specifically Iran. The violence helped further to legitimize the new get-tough policy with respect to foreigners. The immediate effect of these measures was to restrict the civil liberties of foreigners, specifically North Africans, thereby launching a psychological campaign against immigrants and immigration. The policies were explicitly designed to win back supporters of the Front National, and to prevent any further loss of votes to the extreme right on the issue of immigration. But they also heightened the sense of crisis and contributed to the growing debate over a loss of national identity.

If we look at the numbers (flows), which measure the outcome of French control policies, what we find is considerable continuity. Total immigration hovered between 200,000 and 100,000 persons annually, throughout the 1980s (see Table 1). The only noticeable increase in flows, as in other European states, was in the number of asylum seekers, which peaked at 61,372 in 1989. With the end of the Cold War and the gradual implementation of the Schengen Agreement in the 1990s, the rate of rejection of asylum applicants rose from 57 percent in 1985 to 84 percent in 1995 (see Table 3). So, if flows were not raging out of control, what was the purpose of the first Pasqua Law of 1986? The most important and controversial aspect of the reform was the attempt to weaken the birthright principle of jus soli, by putting an end to the practice of "automatically" attributing citizenship at age 18 to the children born in France of foreign parents. In effect, this reform, which was intended to placate right-wing nationalists and win back Front National voters, was more symbolic than real. Having said this, it is important to note that there were some very real effects of the reform. For example, any immigrant who had been sentenced to more than six months in prison was deemed excludable and would not be allowed to naturalize. West Africans were no longer allowed to naturalize under the streamlined procedure known as "reintegration into the French nation"; and spouses of French nationals would have to wait two (rather than one) years before they could file for naturalization.

Nevertheless, the thrust of the proposal was to require young foreigners to affirm their commitments to the Republic by formally requesting French nationality and taking a loyalty oath. What effect such a change might have on immigration flows was unclear; but the message was quite clear: the acquisition of French citizenship is a privilege, not a right, and it should be withheld from those who have not made a clear commitment to the French nation and society. Regardless of the intention of the reform, the first government of cohabitation succeeded in provoking a political firestorm of protest, as various civil and immigrant rights organizations, such as the La Ligue des Droits de l'Homme, GISTI, SOS-racisme, MRAP, and others, mobilized against the reform, leading Pasqua and Chirac eventually to withdraw the bill from consideration. As pointed out above, these changes did not affect the second generation Franco-Algerians, because they were born in France of parents who were born in Algeria prior to 1962--the year of Algerian independence. Both the parents and the children were French by birth (double jus soli), and therefore eligible to naturalize.

(Table 3, here)

The withdrawal of the bill constituted a political failure for the Chirac government, which had unwittingly provided the increasingly active French civil rights movement with a new rallying cry: "Ne touche pas à mon pote!" (Don't touch my buddy!) Thousands marched in Paris under this banner. But, in addition to altering the political landscape, launching a new debate about French citizenship and national identity, and creating new political opportunities for the left, the attempted reform brought the power and prestige of the Council of State to bear. In ruling on the legality and constitutionality of the bill, the Council of State put the government on notice that the rights of individual foreigners and the republican tradition must be respected. This was a lesson in immigration politics and law, which Minister Pasqua would not soon forget. In 1993, he would have a much stronger hand to deal with the judiciary (see below). But in this round of reform, the right-wing government was forced to compromise and the decision was made to appoint a special Commission (Commission des Sages) to hold hearings on the possibility of reform of immigration and naturalization policy. The Commission was composed of political and intellectual elites, and it was chaired by Marceau Long, who, as Vice President of the Council of State, was deemed to have the moral and legal authority to tackle this difficult policy issue.

After hearing the testimony of many experts on immigration, the Commission simply reaffirmed the importance of the republican tradition by defending the birthright principle of jus soli, while at the same time stressing the importance of integrating foreigners into public and civic life. The whole episode of reform during the first government of cohabitation had little discernible impact on immigration flows, which remained well over 100,000 during the late 1980s and early 1990s (see Table 1). The number of naturalizations remained in the average 50,000 range during the same period (see Table 4). However, if we look at changes in the ratio of naturalizations by decree (par décret) to those by declaration (par déclaration) for the period 1984-1995, what we find is an upsurge in 1985-86 and again in 1995. This indicates that a larger number of individuals were filing for naturalization during the key years of the Pasqua reforms, while the number of those qualifying for "automatic" naturalization (by simple act of declaration) remained relatively constant. The exception is 1994, when, with the implementation of the Second Pasqua Law (see below), the number of those declaring themselves to be French shot up to 43,035, twice the average of 19,911 for the period 1973-1992 (see Table 4). In effect, one of the principal unintended consequences of tougher control policies in France was the revaluation of citizenship, speeding up the process of naturalization and integration of the foreign population, and inadvertently reinforcing the republican tradition. The same thing happened again in the mid-1990s in France and in the United States, where anti-immigrant policies were pushed through the Republican-led Congress in 1996, contributing to a wave of insecurity among foreigners and a tremendous surge in naturalizations.

(Table 4, here)

The liberal and republican right (UDF and RPR) lost its battle to eliminate the Front National and it also lost the elections of 1988. Jacques Chirac was defeated in his bid to unseat François Mitterrand, who won a second, seven-year presidential term; and the right also lost the legislative elections, as the socialists, led by Michel Rocard, regained control of the Assembly, albeit with the necessity of forming a minority government. With a score of 14.5 percent of the vote on the first round of the presidential elections, Le Pen continued to cause problems for the right. Since his party again exceeded 10 percent of the vote on the first round of the legislative elections, but gained no seats in the Assembly, Le Pen claimed that the voices of a substantial segment of the French electorate were not being heard, specifically on the issue of immigration. In response, Charles Pasqua, now the former Minister of the Interior, tried to reassure Front National supporters that the RPR shared many of their "concerns and values" with respect to the impact of immigration on French national identity. Public opinion polls at the time (1988-89) showed that approximately one third of the electorate had sympathies for the Front National's position on immigration.

The new left-wing government, led by the two old socialist rivals, Mitterrand and Rocard, essentially returned to the policies of the early 1980s, increasing regulation of the labor market, campaigning against illegal immigration, and taking steps to help integrate immigrants. To this end, Rocard created the Haut Conseil à l'Intégration, to study ways of speeding the integration of the foreign population, which still constituted over 6 percent of the total population. For the period (1988-93), socialist governments fell back on a "grand bargain" strategy of strictly controlling inflows in order to integrate those foreigners already in the country. The hope was to depoliticize the whole issue and defuse the national identity crisis. But no sooner had the left returned to power than it found itself confronted with a highly symbolic controversy, which struck at the heart of the republican tradition and risked splitting the Socialist Party itself into competing factions.

The controversy arose when three school girls of Moroccan descent came to a public lycée, wearing Islamic scarves (foulards), in direct violation of the principles of separation of church and state (laïcité), one of the core principles of the republican tradition. The event immediately became a cause célebre for the anti-immigrant right, as well as the republican left, with more liberal (or pluralist) elements of the political and intellectual elite, including Rocard, caught in the middle. Allowing the girls to wear the scarves was bound to offend both the left and the right; but forcing them to remove the scarfs could open a new pandora's box, concerning the dividing line between the public and private sphere, including the wearing of other religious symbols in the classroom, such as the crucifix or the Star of David. The event also heightened the sense of crisis with respect to immigration control, because of the widespread fear that the new immigrants from North Africa, as well as the second generation, were increasingly prone to Islamic fundamentalism and therefore inassimilable in a secular, republican society, where the individual should keep his or her private life and religious beliefs completely separate from the public sphere. One of the leaders of the Socialist Party who was most adamant in his opposition to such overt violations of the sacred republican principle of laïcité was Jean-Pierre Chevènement, who held ministerial posts in various socialist governments.

Prime Minister Rocard and then Minister of Education, Lionel Jospin, took the decision to allow the girls to wear their scarves, so long as they agreed not to proselytize or in any way disrupt classes. As happened frequently with the issue of immigration and integration, when the rights of individuals vis-à-vis the State were in question, the Conseil d'Etat was called upon to help resolve the controversy. In this case, the Conseil d'Etat simply ratified the compromise position taken by the Rocard government. But the compromise did little to allay the growing fears of Islamic fundamentalism among the French public, and the foulards affair, as it came to be known, raised a new specter of multiculuralism (à l'américain), seen as yet another threat to French unity and national identity, exemplified by the "One and Indivisible Republic." At the same time, le droit à la différence (the right to be different) became the new rallying cry of those defending the rights of immigrants.

Despite the almost continuous atmosphere of crisis in French politics over immigration, integration, and national identity, dating back at least to the early 1980s, very little had changed, either in terms of policy outputs (actual policies for controlling immigration) or in terms of policy outcomes. In the end, the first experience of cohabitation (1986-88) did little to alter the republican model and the rules of the game, as spelled out in the ordonnances of 1945. France continued to be open to legal immigration, with no quotas or ethnic/racial preferences (in contrast to the American model), even though everything possible was done by the left and the right to discourage purely economic (or worker) immigration. Flows, which are the best measure of policy outcomes, continued at the level of 100,000 or more a year (see Table 2), and the liberal nationality code allowed for the relatively quick naturalization of the foreign population (see Table 4).

In terms of immigration control, perhaps the most important change came in the area of refugee policy, with the conclusion of the Dublin Agreement, negotiated by the E.U., and the establishment in 1990 of the Schengen group. In both instances, France committed itself to refuse entry to any asylum seeker who had passed through a "safe" third country, thereby clearing the way for the refoulement of many asylum seekers. The Schengen Agreement also engaged France in the construction of a common European territory--a Europe without internal borders--requiring the harmonization of visa and asylum policies within the Schengen group, as well as increased policing of external borders. These new European initiatives in some ways would come to represent a challenge to the republican tradition, because of the limits imposed on due process and the attempt (via the Maastricht Treaty) to create a semblance of European citizenship and grant voting rights in local elections for permanent resident aliens (denizens).

From a "Threshold of Tolerance" to "Zero Immigration"

When asked about immigration policy in 1991, President Mitterrand suggested that every society, including France, has a "threshold of tolerance" (seuil de tolérance), beyond which instances of xenophobia and racism are likely to increase. But he refused to specify what exactly that threshold might be in the case of France. On the other hand, Charles Pasqua, soon to be (for the second time in his career) Minister of the Interior, stated bluntly that "France has been an immigration country, but she wants to be no longer." Like any good nationalist and populist, Minister Pasqua claimed to be speaking in the name of the French people. However, as a powerful member of the second government of cohabitation, elected by a landslide (the coalition of Gaullists and liberals won over 80 percent of the seats in the Assembly) in the Spring of 1993, Pasqua made clear what the immigration policy of the new government would be: "the goal we set, given the seriousness of the economic situation is to tend towards zero immigration." This explicit linkage of immigration to the severe economic recession--which began in 1991-92 and would push unemployment in France to postwar highs of well over 10 percent--was again aimed to appeal to the 12.4 percent of French voters who supported the Front National in the first round of the 1993 parliamentary elections. Immigration and integration policies were still very much at the center of French politics, and would remain so throughout the 1990s.

Faced with a badly weakened, divided, and demoralized socialist opposition, and having won an overwhelming majority in Parliament, the new right-wing government, headed by Edouard Balladur, had a virtually free hand to pursue draconian policies for (1) stopping all immigration, (2) reducing the number of asylum seekers to an absolute minimum, and (3) reforming the nationality code to block naturalization of as many of the resident foreigners as possible. These new policies represented a clear break from the old socialist "grand bargain." Even though Mitterrand was still President (and would be until 1995), he was clearly a lame duck and quite ill with prostate cancer. Hence, he was in no position to oppose what looked to be a truly dramatic shift in immigration policy. Only the courts potentially could block the change, therefore the Balladur government wasted no time in launching a sweeping reform of immigration and refugee policy, designed to move France as close as possible to zero immigration. To discourage further settlement of foreigners, the nationality law also would be changed.

What distinguishes this round of reform (in 1993) from earlier attempts to limit immigration (in 1974 or 1986, for example) is the clear focus on rolling back the rights of foreigners across the board. The second Pasqua Law presented a direct challenge to the republican model, as defined by the Ordonnances of 1945. Equal protection and due process (civil rights) were denied to foreigners by cutting off possibilities of appeal for asylum seekers and by giving the police much greater powers than ever before to detain and deport foreigners. Social rights also would be severely restricted by denying foreigners access to the benefits of the social security system, especially health care. On this point, however, a rift developed within the government between the Minister of the Interior, Pasqua, and the Social Affairs Minister, Simone Weil, who argued for maintaining emergency health care for foreigners.

The debate in France over social rights for immigrants parallels a similar debate that was gathering force in the United States, especially in California, where voters approved a measure (Proposition 187) in November 1994 to cut public and social services for illegal immigrants. Barely two years later (in 1996), the U.S. Congress, under Republican control, would adopt similar laws to cut social services for legal as well as illegal immigrants, and the rights of appeal for illegals and asylum seekers would be sharply curtailed. Also in the U.S., proposals were made by prominent right-wing politicians, such as Governor Pete Wilson of California, to limit birthright citizenship, so that the children born of foreign parents would no longer be automatically entitled to American citizenship. Similarly in France, the second Pasqua Law (like the first) sought to change naturalization procedures by requiring children born in France of foreign parents to make a formal request for naturalization, between the ages of 16 and 21.

Limiting Rights: Negative versus Positive Freedom

What we can see in these two old republics in the 1990s is a not-so-subtle shift in strategies and tactics for restricting immigration, away from a reliance on the classic instruments of (external) control of borders and (internal) regulation of labor markets to a new strategy of attacking and limiting the rights of foreigners. In liberal societies, external control of borders--with an emphasis on territorial sovereignty and the sanctity of law--is preferable to internal control of society, which may entail infringements of individual, civil liberties. All things being equal, liberal states will opt for external strategies of control, placing the most stress on border control, or control of territory. The reason for this is simple: territorial closure and sovereignty are essential to the maintenance of the social contract and the rule of law, and this cannot be questioned without questioning the authority and legitimacy of the state itself.

In the 1990s, neither the French (led by Gaullists and right-wing liberals) nor the American governments (led by a Republican Congress and a Democratic President) have abandoned the classic instruments of border control. On the contrary, they have reinforced them, especially in the American case in recent years where the liberal concept of negative freedom (implying a minimalist state and freedom to do whatever one wants within the broad confines of the law) has had a much more powerful influence on politics and public policy. The Clinton administration has placed great emphasis on (external) control of borders, as well as (internal) regulation of labor markets. If control cannot be easily externalized, then a series of internal control policies will come into play and the question then becomes: how far can a liberal state go in imposing such controls on individuals in (civil) society? Should foreigners and immigrants be considered members of civil society? The limits on internal control are imposed by ideas, institutions, and culture, as well as certain segments of civil society, which may resist encroachments by the state on negative and/or positive freedoms.

The powers of the American federal state to regulate the domestic labor market and control illegal immigration are constrained by the lack of a national identification card. Precisely because of their attachment to negative freedoms, Americans are wont to give the State such sweeping power to intrude in their daily lives. In France, on the other hand, the State has long had this type of regulatory power, giving it the capacity to intervene directly and forcefully in the labor market. In countries like France, which have a strong statist, administrative, and/or social democratic tradition, regulation of the labor market is likely to be the first strategy for internal control. Conversely, in countries like the U.S., which have a strong liberal/laissez- faire tradition and a weaker state, regulation of the labor market is not a viable strategy, because of the weak capacity of the state, the resistance of civil society, and the reluctance of politicians to impose limits on negative freedom, which arises in part from free markets. In countries like the U.S., which have no social democratic tradition and a weak welfare state, the first strategy for internal control is likely to be a roll back of social rights, i.e. the restriction of positive freedom. Conversely, in countries like Sweden or Germany, which have a strong social democratic tradition, social rights are likely to be preserved for all members of society, denizens as well as citizens; and the preferred strategy for control will be external control of borders, strict regulation of labor markets, and limits on negative freedom. Yet in nowhere in the OECD world has the State been able to stop immigration, at least not to the satisfaction of the government or the people (if public opinion polls and election results are to be believed).

But there is evidence in the recent reforms in France and the U.S. of the willingness on the part of politicians and the public to allow the erosion of negative freedoms, so that the state can better manage legal immigration, and stop illegal immigration. The increased power of the police (both internally and at the border) to detain and deport individuals has eroded civil liberties in France and the U.S. Limiting appeals by asylum seekers and others threatened with deportation further strengthens the hand of the State in dealing with illegal immigration, but it is difficult for liberals to accept, because it means more state control and less (negative) freedom. In countries with a strong liberal-republican tradition, such as France, the U.S., and Germany, rolling back civil rights (due process and equal protection) as a strategy for internal control and limiting negative freedom will be extremely difficult and contentious. The extent to which this is a viable strategy will depend heavily upon the strength and independence of the judiciary, which may act as an institutional constraint on both the populist impulses of the legislature or parliament, as well as the administrative or police powers of the executive. Separation of powers can act as a constraint on the ability of states to roll back civil rights and take away negative freedoms. In this respect, Britain has both an institutional advantage (no separation of powers) and an ideological/cultural advantage (no republican tradition) in pursuing internal control strategies. In both France and the U.S. the state is more powerful and the individual is weaker because of these reforms. Liberal lobbies like the ACLU in the U.S. and the GISTI in France have fought restrictive immigration reforms on the grounds that they erode the civil liberties of all members of society.

The GISTI has had a much more difficult time blocking reform in France, than the ACLU, which has effectively stopped the adoption of national ID cards in the U.S. To constrain worker and family immigration in France, the second Pasqua Law (1994) required workers and foreign students to wait two (rather than one) years before bringing any family members to join them. To inhibit permanent settlement of foreigners and to control illegal immigration, the Law prohibited adjustments of status (régularisation) for any undocumented individual who marries a French citizen. Mayors were given the authority to annul any suspected marriage of convenience (mariage blanc). In this case the State inserted itself directly into the private lives of French, as well as foreign, citizens. Finally, under the second Pasqua Law, any foreigner expelled from France would be denied reentry into French territory for one year.

The second Pasqua reforms, together with the examples from recent changes in American policy (the 1996 laws), indicate the lengths to which liberal states are willing to go in rolling back the rights of foreigners (and thereby abandoning some aspects of negative freedom and the minimalist state) in order to restrict immigration. A somewhat easier target--more so in the U.S. than in France--is positive freedom revolving around the welfare state and flowing from laws designed to help the individual take advantage of the opportunities afforded by negative freedom in a liberal society. The whole range of welfare benefits, from education to health care and pensions, has become a target for those wishing to restrict the rights of foreigners as a way of controlling immigration. Such actions taken by liberal states against foreigners would seem to be less threatening to citizens, depending upon the extent to which citizens are attached to social rights and determined to protect them for all members of society, even for the most marginal, disenfranchised groups, like children, immigrants, and foreigners. The French are certainly more attached to social rights (les acquis sociaux) than are Americans, but less so than the Germans, Dutch, or Scandinavians. Thus it is somewhat easier for the Americans and the French to cut welfare benefits for foreigners than it would be for the Germans or the Swedes, for example.

Looking at T.H. Marshall's classic trilogy of rights (civil, social, and political), we can see that France--like other liberal, republican, and social democracies--has acted to constrain the civil (equal protection and due process) and social (welfare) rights of immigrants and foreigners. If we follow this policy (of limiting the rights of non-citizens) to its logical conclusion, then the ultimate rights that can be denied to foreigners are political (or voting) rights, which are tied to naturalization. To roll back these rights in a liberal republic requires tampering with the nationality law, and we already have seen evidence of this in both the first and second Pasqua Laws. Birthright citizenship, enshrined in the Fourteenth Amendment to the Constitution, also has been challenged in the U.S. in the 1990s. But changing nationality laws in a republic in order to deal with immigration or integration is exceedingly difficult and fraught with many political dangers, as we have seen in each case (France, the U.S., and especially Germany). Reform in this area often means opening up difficult, moral debates about national identity, with many political, historical, and constitutional overtones. The relative ease with which the Thatcher government changed British nationality laws in 1981 stands in stark contrast with similar attempts at reform in France and the U.S.

In the case of France, amending the nationality code means changing the universalist and nationalist republican model, as it has evolved since 1945. In the case of the U.S., it would mean amending (or at least reinterpreting) the Fourteenth Amendment to the Constitution, adopted in the wake of the Civil War, and designed to overturn the Supreme Court's Dred Scott decision by granting automatic citizenship to former slaves. In Germany (the Bundesrepublik Deutschland), the problem of control is inextricably linked to the problem of integrating a large, permanent foreign population. To solve the problem of integration in Germany, it will be necessary to change the 1913 nationality law, thus abandoning a key aspect of German national identity, that is a community based on the ethnocultural principle of descent (jus sanguinis) rather than on the more universalistic and territorial principles of birth and soil (jus soli). When this change in the German nationality law is made, it will bring Germany more fully into the club of liberal republics.

Infringing individual and group rights or tampering with the social contract, the constitution, and national identity as a means of controlling immigration are fraught with danger for liberal and republican states, because the state or government runs the risk of undermining its own legitimacy, and alienating and/or endangering its own citizens. Moreover, otherwise liberal politicians may inadvertently (or unwittingly) provoke a nationalist, xenophobic, and even racist backlash, which could redound against these same politicians and undermine the state (and rule of law) itself. Fortunately, in states like France and the U.S., which have long histories of liberal and republican governance, there are institutional and ideological checks that work to protect the state and politicians from themselves. In France, as in other liberal states such as the U.S and Germany, the courts play a crucial role in this regard. As it did with the First Pasqua Law (in 1986), the Conseil d'Etat--which functions in part as an institutional watchdog for any infringements of the rights of the individual by the State--warned the Balladur government that certain aspects of the proposed reform were illegal and possibly unconstitutional. The Conseil d'Etat was especially concerned with the impact of the Second Pasqua Law on the (constitutional) right of families to live together, and with the provisions of the Law limiting the right of appeal for asylum seekers. But the Conseil d'Etat has no real powers of judicial review. Its opinions are only advisory, even though it is one of the most powerful grands corps--second only to the Inspection des Finances. Still it has great moral, political, and legal authority, and governments ignore its views at their peril. Moreover, decisions of the Conseil d'Etat may presage a ruling by the Conseil Constitutionel, which does have powers of judicial review and may stop the implementation of any law deemed unconstitutional.

This is precisely what happened in August 1993 when the Conseil Constitutionel ruled that certain provisions of the Second Pasqua Law were unconstitutional. The judges rejected the one-year ban on reentry imposed on anyone deported from France. Also found to be unconstitutional were the provisions dealing with family immigration, namely (1) the longer waiting period imposed on foreign students and workers seeking to bring immediate family members to join them, and (2) the restrictions imposed on marriages between French citizens and foreigners. In rendering these decisions, the Conseil Constitutionel relied specifically on the Declaration of the Rights of Man and the Citizen, referring to the universalist and egalitarian principles of this republican document, especially equal protection. Moreover, citing the Preamble to the 1946 Constitution--which requires that due process be accorded to all asylum seekers--the Conseil Constitutionel ruled that restrictions on the right of appeal and provisions in the Law for the automatic refoulement of refugees were unconstitutional. This ruling seemed to jeopardize France's participation in the Dublin Convention as well as the Schengen Agreement, both of which require European states that are parties to these agreements to refuse asylum to any individual who has passed through a safe third country.

The efforts of the Balladur government to move France to "zero immigration" did little to calm the national identity crisis. If anything, the Second Pasqua Law heightened the sense of crisis, and fanned the flames of xenophobia, leading to a constitutional debate. But one objective of the reforms appears to have been met: immigration flows fell precipitously from a high of 135,000 (including asylum seekers) in 1992 to 68,000 in 1995. The average annual rate of immigration for the period 1993-95 plummeted to its lowest levels (89,700) since the late 1940s and early 1950s (66,400--see Table 1). But, as we shall see--in the crisis over the Debré Law in 1997--many of the heretofore legal flows were simply pushed underground, raising the size of the undocumented population (sans papiers), and increasing the level of insecurity among the foreign population as a whole. The numbers of individuals caught trying to enter the country illegally (reconduites à la frontière) rose steadily from 1993 on, jumping from 8,700 in 1993 to 10,100 in 1995 and over 12,000 in 1996, providing (indirect) evidence of increased illegal immigration.

Nevertheless, to combat the judges and complete his reform, the Interior Minister, Pasqua turned one aspect of the republican tradition (popular sovereignty) against another (birthright citizenship). Claiming that the people, having spoken through their representatives, want immigration reduced, he called for a constitutional amendment that would prepare France for entry into a border-free Europe and give the State the power to turn back asylum seekers without hearings or appeals. As provided by the Constitution, the amendment was voted in an extraordinary congress of the Parliament (the Assembly and the Senate) at Versailles in January 1994. Pasqua proclaimed that there would be no "government by judges" in France, as in the U.S. where many anti-immigrant measures, such as Proposition 187 in California, have been blocked by the Federal judiciary. Ultimately the U.S. Supreme Court can rule on the constitutionality of American immigration control policies, especially when they impinge upon basic civil rights and liberties. The question remains open, however: how far can a liberal-republican state go in rolling back rights of individuals in its effort to control immigration? At what point does the liberal-republican model begin to break down? What is the appropriate balance between internal and external controls? Between negative and positive freedom?

Civil Disobedience and the Limits of Control

Immigration continued to agitate French politics and society into the mid and late 1990s, during the presidential election of 1995 and especially during the legislative elections of 1997. The election of Jacques Chirac as President of the Republic by a narrower-than-expected margin over the left candidate, Lionel Jospin, did little to change French immigration policy, even though Le Pen received a record number of votes (15 percent) on the first round of the presidential elections. The new government, led by Chirac's lieutenant, Alain Juppé, had the support of the same crushing right-wing majority in the Assembly. The UDF and RPR still controlled 80 percent of the seats. But one big difference was the absence of Charles Pasqua from the Jospin government. Pasqua had supported Chirac's rival, the former Prime Minister Edouard Balladur, for the presidency. He was replaced as Minister of the Interior by Jean-Louis Debré, son of Michel Debré, the first Prime Minister of the Fifth Republic and author of the Constitution. Debré fils would quickly make a name for himself by proposing further, draconian steps to limit the rights of foreigners in France and crack down on illegal immigration. The Debré Law of 1997 would test the limits of strategies for (internal) immigration control, leading to civil disobedience, more court rulings, new elections (thanks to the political blunders of Chirac and Juppé), and finally a resurgence of the republican left.

In the summer of 1996, the tough control policies (described in the preceding section) were challenged by a group of Africans, mostly from Mali, who were caught in the web of the Second Pasqua Law (unable to obtain a residency permit, even though many of them had resided in France for many years and could not be legally deported) or whose applications for political asylum had been rejected. The sans papiers, as they were called, occupied a church in Paris, demanding that they be given an adjustment of status (régularisés), and several of them launched a hunger strike. The highly public épreuve de force with the new government was indicative of the willingness of immigrants openly to resist the government's policy and of the sympathy they were able to generate among certain segments of French civil society. Over 10,000 people marched in Paris in solidarity with the sans papiers, and even more embarrassing for the government were appeals by the clergy not to remove the immigrants from the church by force. Nonetheless, the police were ordered to storm the church, arrest the protesters, and break up the hunger strike. The government also proudly published statistics indicating that deportations for the first six months of 1996 were up substantially (by about one third) over the similar period for 1995. Any sign of weakness or wavering by the government in the face of immigrant resistance and civil disobedience was immediately condemned by Le Pen and the Front National.

Apart from occasional acts of civil disobedience by the African sans papiers, which continued throughout 1995-97, whether in the form of occupying churches or, in one case, the offices of UNESCO, the civil war in Algeria also had an impact on French control policy. Since the abrupt cancellation of the Algerian elections in 1992--which Islamic fundamentalists were poised to win--a civil war has raged in the former French colony. The conflict pitted the Islamic radicals against the long-ruling revolutionary party, the FLN, which controls the military. The elections were canceled with the blessing of the French government, which made no attempt to hide its support for the Algerian military. French involvement in Algerian politics led to a number of terrorist attacks by Islamic militants against public targets in France. These attacks forced the government of Alain Juppé to increase security throughout the country. The security sweeps by the police and the military, known as operation vigipirate, focused public attention on the Muslim (and African) communities in France, bringing the full power of the French state to bear in an effort to catch the perpetrators. In October, 1995, the police shot and killed one of the bombers--a young second-generation Algerian man--in the outskirts of Lyon. In the press, his life story was covered in detail and held up as an example of the failure of French society to integrate some segments of the young, Maghrebi population. These dispossessed youths, with no loyalty to the land of their birth, had joined radical Islamic groups and turned against the French Republic.

As in the 1950s, French foreign policy and relations with former colonies, especially Algeria, have become a driving factor in immigration and refugee policy in the 1990s. The government has felt compelled to grant asylum (or at least temporary residence) to many members of the Algerian political and intellectual class, while at the same time stepping up pressure to keep other Algerians out, and carefully to surveil the established Algerian community in France. This atmosphere of crisis and public insecurity together with continuing pressure from the Front National led the Juppé government late in 1996 to propose a new law, which came to be known as the Debré Law, designed to resolve the ambiguous status of some of the sans papiers, particularly the French-born children of illegal immigrants and the foreign spouses of French citizens. These groups could not be deported, but under the Pasqua Law they were not eligible for an adjustment of status. Under the proposed new law, the "foreign" children under 16 years of age would have to prove continuous residence in France for ten years, and "foreign" spouses would have to have been married for two years in order to be eligible (like the children) for a one-year residence permit.

But, even though the Debré Law had some liberal intent, it got much more publicity and became the focal point of controversy and protest because of a provision added to the bill by the conservative National Assembly. The provision required all private citizens to notify local authorities whenever they received in their homes any non-EU foreigner. Moreover, mayors would be given the authority to verify that a foreign visitor had left the private citizen's home once the visitor's visa had expired. What is most interesting about the Debré Law is not so much the effect (or lack thereof) that it had on immigration control, but the response it received both from French certain groups in civil society and institutions of the liberal-republican state. Minister Debré, paraphrasing his predecessor Pasqua, stated that "I am for zero illegal immigration... The State must be given the means to deter foreigners who want to enter France without resources, papers or jobs." The focus in this statement is on those clearly outside the law, i.e. illegal immigrants; but public attention was focused on the effect that the Law would have on private French citizens, who w