email icon Email this citation

Losing Control?
Sovereignty in an Age of Globalization

Saskia Sassen

New York

Columbia University Press


3. Immigration Tests the New Order

Economic globalization denationalizes national economies; in contrast, immigration is renationalizing politics. There is a growing consensus in the community of states to lift border controls for the flow of capital, information, and services and, more broadly, to further globalization. But when it comes to immigrants and refugees, whether in North America, Western Europe, or Japan, the national state claims all its old splendor in asserting its sovereign right to control its borders. On this matter there is also a consensus in the community of states. One of the questions I want to examine here is the interaction between the denationalizing of key economic institutions and spaces, on the one hand, and the renationalizing of politics on the other--economic denationalizing and political renationalizing, in short. How can the state relinquish sovereignty in some realms and cling to it in others?

Beyond the facts of economic transnationalization, in dealing with immigration the state confronts the ascendant international human rights regime. Immigrants and refugees bring to the fore the tension between the protection of human rights and the protection of state sovereignty. This tension is particularly sharp in the case of undocumented immigrants, because their mere existence signifies an erosion of sovereignty. At least in part the tension originates in the state itself, in the conflict between its authority to control ingress and its obligation to protect those in its territory. Furthermore, insofar as immigrants and refugees have gained considerable rights, this can be read as a devaluation of citizenship as a condition for access to rights. Yet, as I will argue, this is a two-way process. The state has been a participant in the emergence of human rights regimes, and various components of the state have been key agents for the incorporation of human rights in domestic law, as in the United States and Germany, for instance. Thus, as I argued for the case of economic globalization, here, too, the state, especially if under the rule of law, has been transformed by its implementation of human rights regimes, both in domestic law and in international agreements. The tension between state sovereignty and international human rights should not be seen as involving an internal and an outside base: the international human rights regime operates partly inside the national state. And, as is the case with the new legal frameworks for global capital, it is this partial grounding of a transnational regime in national institutions and practices (at least in countries under the rule of law) that lends it a distinctive power and legitimacy.

In chapter 1, I posited that exclusive territoriality--a marking feature of the modern state--is being destabilized by economic globalization and that a denationalization of national territory is now in progress, though in a highly specialized institutional and functional way. In both the previous chapters, I argued that the particular combination of power and legitimacy we call sovereignty is being decentered, partly redistributed onto other entities, particularly supranational organizations, international agreements on human rights that limit state autonomy, and the emergent private international legal regime for business transactions. With all this happening, what is the basis for the usual presumption that the state has exclusive authority over the entry of non-nationals? Is the character of that exclusive authority the same as it was before the current phase of globalization and the ascendance of human rights? have gained considerable rights, this can be re

Immigration is an interesting site for such a research question because, unlike electoral politics or political rhetoric, it is an arena for both deploying ideas and pursuing the day-to-day implementation and practice of policy. The idea of the nation--or, rather, the renationalizing of the nation as an idea--can be explored freely in political rhetoric, and it clearly is, more and more, all over the developed world. Such explorations can be carried pretty far as rhetoric without interfering in the day-to-day operation of government. When it comes to immigration, however, there is a daily need to process applications, accept, reject, drive to detention centers, drive from detention centers, patrol the borders, catch, fail to catch, rescue from leaking ships in the high seas and in the low seas . . . in other words, a vast, unmeasurable multiplicity of realities that unfolds on many different terrains, from borders to courthouses, from legislatures to workplaces. Unlike the idea of renationalizing the nation, immigration reality is continuously intersecting with government practices.

Unlike much of domestic politics, immigration and its associated government practices always carry the potential of becoming an item in international relations. This is reinforced in many cases by the presence of so-called ethnic lobbies, particularly in the United States but also, under different guise, in European countries and Japan. (For instance, before the current conflict in Algeria, the treatment of Algerians in France was a regular item on the agenda between France and Algeria.)

Immigration, then, has the dual property of being a central object in and a tool for the renationalizing of political discourse and being the object of government policy and practice. This can't be said about many of the ideas current in political discourse. Immigration can be seen as a strategic research site for the examination of the relation--the distance, the tension--between the idea of sovereignty as control over who enters and the constraints states encounter in making actual policy on the matter. Immigration is thus a sort of wrench one can throw into theories about sovereignty.

Let's begin with a few figures. Today there are about 120 million immigrants worldwide, an estimate that likely excludes many undocumented aliens. Only about half of these immigrants are in the rich developed countries. Similarly, of the 20 million estimated refugees, only 30 percent are in the rich developed countries; half of these are from the former Yugoslavia and live mostly in neighboring countries. Overall figures, then, are not particularly large as a share of the population of the highly developed world. I could devote a whole chapter to discussing whether open borders would markedly raise these numbers. For now, though, I will just say that large-scale international migrations are embedded in complex economic, social, and ethnic networks; highly conditioned and structured, they are by no means a free-for-all. Reality thus belies the popular dramatic images of massive invasion by the poor. Immigration is really more of a management problem than a crisis.

My analysis focuses largely on immigration in the highly developed receiving countries. 1 I use the notion of immigration policy to refer to a wide range of distinct national policies. Note, too, that it is often difficult to distinguish between immigrants and refugees, but there are separate regimes for refugees in all these countries and an international regime as well, something that can hardly be said for immigration.

The Border and the Individual as Regulatory Sites

Policies about immigration differ widely in the highly developed countries. This is perhaps most sharply captured in the criteria for naturalization. Some countries--for instance, Germany--have naturalization policies based on jus sanguinis, or descent, while others--France, for one--base theirs on jus solis, or place of birth. Some countries, among them, the United States and Sweden, facilitate citizenship acquisition; others, such as Switzerland and Japan, do not. Some, like Germany and France, have instituted explicit return migration policies, including monetary incentives, while other countries, notably the United States, hardly register the fact of return migration. Some countries--Canada and the United States, for example--possess political cultures and identity formation processes that incorporate the fact of immigration, while in others, particularly Germany and Japan, this is not the case whatsoever.

In my reading, however, a fundamental framework roots all the immigration policies of the developed countries in a common set of conceptions about national borders and the role of the state. The purpose here is not to minimize the many differences in national policies but to underline a growing convergence in various aspects of immigration policy. The particular issues of interest are, first, this growing convergence at a time when the developed countries are finding their immigration policies increasingly ineffective, and, second, the de facto transnationalizing of immigration policy making over the last few years.

Key elements in this fundamental framework are (1) the sovereignty of the state and border control as the heart of the regulatory effort (whether on land or at airports or consulates in sending countries); and (2) an understanding of immigration as the consequence of emigrants' individual actions (the receiving country is taken as a passive agent, one not implicated in the process). Refugee policy, in contrast, recognizes additional factors as leading to outflows. 2 The framework for immigration singles out the border and the individual as the sites for regulatory enforcement.

When it comes to power over entry, the sovereignty of the state is well established by treaty law and constitutionally. The Convention of The Hague of 1930 asserted the right of the state to grant citizenship; the 1952 Convention on Refugees, which asserted that the right to leave is a universal right, remained silent on the right to entry (better silence than evident contradiction, I suppose). The status of refugees and their right not to be forcibly returned are established in international law, but there is no corresponding right of asylum; that is at the discretion of the receiving state. 3 Various human rights declarations and conventions urge states to grant asylum on humanitarian grounds, but they all recognize the absolute discretion of states. A few states--notably Germany--give those formally recognized as refugees a legal right to asylum, though these provisions are now under revision in Germany. More recently, the various agreements toward the formation of the European Union keep asserting the right of the state to control who can enter. This is quite a contrast with the assertions in the GATT, nafta, and the eu about the need to lift state controls over borders when it comes to the flow of capital, information, and services, as well as state controls over the domestic financial markets. While current international law imposes important limitations on the exercise of the sovereign power to control entry, 4 overall there is little disagreement as to the state's authority in this matter. 5

On the issue of the individual as a focus for enforcement, two different operational logics are becoming evident. The one embedded in immigration policy, particularly in developed countries, places exclusive responsibility for the immigration process on the individual and hence makes the individual the site for the exercise of the state's authority. Yet it is now increasingly recognized that international migrations are a function of larger geopolitical and transnational economic dynamics. The worldwide evidence reveals that there is a pattern in the geography of migrations and shows that the major receiving countries tend to get immigrants from their zones of influence (this holds for countries as diverse as the United States, France, and Japan). This suggests that, because of economic internationalization and the geopolitics resulting from older colonial patterns, the responsibility for immigration may not be exclusively the immigrant's. Analytically, these conditions can enter into theorizations about the state and immigration only when we suspend the proposition implicit in much immigration analysis that immigration is the result of individual action.

The other operational logic is centered on the international human rights regime, where the individual is not the site for punishment but for legitimate rights. I return to this subject in a later section.

Beyond Sovereignty: Constraints on States' Policy Making

But how much control does the state have over immigration? An emerging de facto regime, centered in international agreements and conventions as well as in various rights gained by immigrants, limits the state's role in controlling immigration. An example of such an agreement is the International Convention adopted by the General Assembly of the un on December 18, 1990, on the protection of the rights of all migrant workers and members of their families [Resolution 45/158]. 6 Legal authorities now widely uphold a set of rights for resident immigrants. For instance, administrative and constitutional courts blocked attempts by the French and German legislatures to limit family reunification, on the grounds that such restrictions would violate international agreements. The courts have also regularly supported a combination of rights that have the effect of limiting governments' power over resident immigrants. Similarly, they have limited the ability of governments to restrict or stop asylum seekers from entering the country. 7 Over the last three decades civil and social rights have gradually expanded to include marginal populations, whether women, ethnic minorities, or immigrants and refugees.

Lobbying by interest groups has become a key ingredient in immigration policy making in the United States. The 1986 Immigration Reform and Control Act (irca) represents a balancing of interests that ended up limiting the effectiveness of the law. Interest groups influenced the drafting of the law and its subsequent implementation. Some ins regulations were liberalized--for example, those regarding legalization conditions--as a consequence of court action by interest groups.

Another form of this confinement of state action is illustrated by the impact of the 1975 Helsinki Accords requiring that barriers to the free movement of people and ideas be lowered. The United States had to soften regulations that were seen as violating the accords. Thus the Foreign Relations Authorization Act, passed in 1977, required the secretary of state to recommend that aliens excluded for political reasons be admitted unless this was determined to be contrary to U.S. security interests. The Moynihan-Frank Amendment of 1987 prohibited the exclusion or deportation of aliens because of political beliefs and actions that, if engaged in by a U.S. citizen, would be protected by the Constitution. Yet another example is the 1980 Refugee Act, which was predicated on the un protocol on refugees and greatly opened the country to refugee claims.

The numbers and kinds of political actors involved in immigration policy debates and policy making in Western Europe, North America, and Japan are also far greater than they were two decades ago: the European Union; anti-immigrant parties; vast networks of organizations in both Europe and North America that often represent immigrants--or claim to do so--and fight for immigrant rights; immigrant associations and immigrant politicians, mostly in the second generation; and, especially in the United States, so-called ethnic lobbies. 8 The policy process for immigration was probably never confined to a narrow governmental arena of ministerial and administrative interaction, but today more than two decades of public opinion and public political debate have become part of the arena wherein immigration policy is shaped. 91 Whole parties position themselves politically in terms of their stand on immigration, especially in some of the European countries.

The overall effect of these developments is to constrain the sovereignty of the state and to undermine old notions about immigration control. This is particularly evident in the European Union, 10 where the single-market program has raised various issues associated with the free circulation of people as an essential element in creating a frontier-free community and the attendant problems for national immigration laws regarding non-EC nationals in EC member states. Though lacking the legal competence to deal with many of these issues, EC institutions had to begin to address them. Gradually, they became more deeply involved with visa policy, family reunification, and migration policy. Governments first resisted EC involvement in these once exclusively national domains. But now, notwithstanding many public pronouncements to the contrary, both legal and practical issues have made such involvement acceptable and inevitable, especially in the aftermath of the collapse of the socialist bloc and the rapid increase in refugees. Though very slowly, the general direction has been toward a closer union of member states' immigration policies.

In the United States, the combination of forces at the governmental level is quite different yet has similar general implications about the state's constraints in immigration policy making. When the Department of Labor (DOL) was created in 1914, it assumed responsibility for immigration policy. In June 1933 President Roosevelt combined those functions into the Immigration and Naturalization Service within dol. The advent of World War II brought a shift in the administrative responsibility for immigration policy: in 1940 Roosevelt recommended it be transferred to the Department of Justice, because of the supposed political threat represented by immigrants from enemy countries. This was meant to last only for the duration of war. But in the late 1940s and 1950s there was great concern over how immigration policy could be used to advance foreign policy objectives. The INS was never returned to the DOL. This meant that in Congress immigration wound up in committees traditionally reserved for lawyers, such as the Senate and House Judiciary Committees. It has been said that this is why immigration law is so complicated (and, I would add, so centered on the legalities of entry and so unconcerned with broader issues).

Today jurisdiction over immigration matters in the U.S. Congress continues to lie with the Judiciary Committee, not with the Foreign Affairs Committee, as might have been the case. Congressional intent is often at odds with the foreign affairs priorities of the executive; there is a policy-making tug of war. 11 Because immigration policy in the United States is largely debated and shaped by Congress, it is subject to a vast multiplicity of local interests, notably ethnic lobbies--and we all know how sensitive congressmen and women are to the demographics of their districts. This has made the process very public, quite different from other processes of policy making. 12

That immigration in the United States has historically been the preserve of the federal government, particularly Congress, assumes new meaning in today's context of radical devolution: the return of powers to the states. 13 Conflict has emerged between several state governments and the federal government over the particular issue of federal mandates concerning immigrants--such as access to public health care and schools--without mandatory federal funding. States with disproportionate shares of immigrants are asserting that they are disproportionately burdened by the putative costs of immigration--although the latest study by the Washington-based Urban Institute found that immigrants contribute 30 billion U.S. dollars more in taxes than they take in services. (I should note, however, that in the United States the costs of immigration are an area of great debate and wide-ranging estimates.) At the heart of this conflict is the fact that the federal government sets policy but does not assume responsibility, financial or otherwise, for the implementation of many key aspects of immigration policy. The conflict is illustrated by the notorious case of the state of California and its 377-million-dollar lawsuit against the federal government.

The question raised by these developments is not so much about how effective a state's control is over its borders--we know it is never absolute. Rather, it concerns the substantive nature of state control over immigration given international human rights agreements, the extension of various social and political rights to resident immigrants over the last twenty years, and the multiplication of political actors involved with the immigration question. There is, first, the matter of the unintended consequences of policies, whether immigration policies as such or other policies that have immigration impacts. For instance, the 1965 U.S. Immigration Act had consequences not intended or foreseen by its framers; there was a generalized expectation it would bring in more of the nationalities already present in the country--that is, Europeans--given its emphasis on family reunion. 14 (It did not.) Other kinds of unintended migrations are those related to the internationalization of production. 15 And although immigration policy has rarely been an explicit, formal component of the foreign policy apparatus in the United States, foreign and military aid have rarely deterred emigration. On the contrary. Refugee flows from Indochina are an obvious example, but consider, too, the case of El Salvador in the 1980s: billions of dollars poured in, and hundreds of thousands of Salvadorans poured out as U.S. aid raised the effectiveness of El Salvador's military control and aggression against its own people. 16 Or take the Philippines, a country that received massive aid and has had high emigration. In both cases, the foreign aid was dictated by security issues. U.S. economic and political interventions also played a role in the Dominican emigration in the 1960s and in the emigration from India and Pakistan. (I have long argued that policymakers should attach migration impact statements to various policies having to do with overseas activities, from foreign direct investment to military aid.)

Domestic U.S. policies that have impact overseas have also contributed to promote emigration to the United States. A good example is the notorious sugar price support provision of the early 1980s: Taxpayers paid 3 billion dollars annually to support the price of sugar for U.S. producers. This kept Caribbean Basin countries out of the competition and resulted in a loss of 400,000 jobs there from 1982 to 1988 (the Dominican Republic alone lost three-quarters of its sugar export quota in less than a decade). Predictably, the 1980s was also an era of large increases in immigration to the United States from that region.

A second condition illuminating the issue of the substantive nature of state control over immigration involves a zero-sum argument: if a government closes one kind of entry category, recent history shows that numbers will increase in another. A variant on this dynamic is that if a government has, for instance, a very liberal policy on asylum, public opinion may turn against all asylum seekers and close up the country totally; this in turn is likely to promote an increase in irregular entries. It is becoming increasingly evident that unilateral policy in major immigration countries is problematic. Germany is a dramatic example: that country began to receive massive numbers of entrants as the other European states gradually tightened their policies while its asylum policy remained very liberal. And the EC today stresses the importance to the eu as a whole of the Mediterranean countries--Italy, Spain, and Portugal--maintaining control over their borders to exclude non-EC entrants.

All these constraints on the state's capacity to control immigration should not be seen as a control crisis. In fact, it seems to me that the fears of such a crisis common in many highly developed countries today are in some ways unwarranted, even though states have less control than they would like. A look at the characteristics of immigrations over time and across the world reveals that flows are highly patterned, contingent on other dynamics that contain equilibrating mechanisms; that they tend to have fixed durations (many immigrations lasted for fifty years and then came to an end); that there is more return migration than is generally recognized (for example, Soviet engineers and intellectuals going back to Moscow from Israel; Mexicans returning home after becoming legal U.S. residents through the irca amnesty program because they could now move freely between the two countries). Examination of earlier historical periods when there were no era of large icontrols also shows that most people then did not leave poorer areas to go to richer ones, even though there were plenty of such opportunities in Europe within reasonable travel distances. 17

Yet another set of conditions, mentioned earlier, reduces the autonomy of the state in controlling immigration: Large-scale international migrations are highly conditioned and structured, embedded in complex economic, social, and ethnic networks. States may insist on treating immigration as the aggregate outcome of individual actions, but they cannot escape the consequences of those larger dynamics. A national state may have the power to write the text of an immigration policy, but it is likely to be dealing with complex, transnational processes that it can only partly address or regulate through immigration policy as conventionally understood.

Beyond the Individual: Economic Internationalization and Geopolitical Links

Each country is unique, and each migration is produced by specific conditions of time and place. But to theorize about the impact of economic internationalization we must step back from these particulars to examine more general tendencies in economic dominance and the formation of transnational spaces for economic activity. The goal is to grasp the impact of the internationalization of economies on, first, the mechanisms connecting emigration and immigration countries and, second, the organization of labor markets in both types of countries.

These two in turn have an impact on the formation and direction of migration flows. They produce conditions under which poverty, unemployment, or lack of opportunities for advancement can become activated as migration push factors. For example, the development of commercial agriculture and export-oriented standardized manufacturing have dislocated traditional economies and eliminated small producers. They also contribute to the conditions under which immigrants can enter the labor markets of receiving countries. For example, increased competitive pressures from the internationalization of production cause businesses to favor low-wage workers at the expense of unions, in order to remain competitive with cheap third-world imports.

The mechanisms binding immigration countries to emigration countries can assume many forms. But two appear to be dominant. One is past colonial and current neo- or quasi-colonial bonds, which can generate the types of military actions the United States has taken in El Salvador or the Philippines. The other mechanism is the economic links brought about by internationalization, ranging from the off-shoring of production, to the implantation of export-oriented agriculture by means of foreign investment, to the power of multinationals in the consumer markets of sending countries.

A third type of link, characterized by greater specificity and a variety of mechanisms, is the organized recruitment of workers, either directly by a government (within the framework of government-supported employer initiatives) or through kinship and family networks. Ethnic links established between communities of origin and destination, typically by transnational households or broader kinship structures, are crucial after a flow has begun, and ensure its persistence. These recruitment and ethnic links tend to operate within the broader transnational spaces created by neocolonial processes and/or economic internationalization.

It is a little-known fact that, during the 1800s as well as today, some form of organized recruitment by employers or governments often stimulated immigrant flows. But eventually most migration flows become independent of organized recruitment. Although organized recruitment, and with it the constitution of certain countries as labor-exporting, is radically different from the migrations engendered by former colonial bonds, similarities also exist.

The mass migrations of the 1800s formed part of a trans-atlantic economic system binding several nations by economic transactions and wars. These mass migrations were highly important to U.S. development. Massive flows of capital, goods, workers, and specific structures produced the American economic system. Earlier movements of labor across the Atlantic had largely been forced, notably through slavery, and mostly from colonized African and Asian territories. Similarly, the migrations to England of the 1950s originated in what had been British territories. Finally, the migrations into Western Europe of the 1960s and 1970s were initiated by recruitment (in a context of European regional dominance) in the Mediterranean and some Eastern European countries. Few if any passive bystanders can be found among countries receiving large influxes of laborers.

The renewal of mass immigration into the United States in the 1960s, after a hiatus of five decades, occurred during a period of expanded U.S. economic and military activity in Asia and the Caribbean Basin. The United States is at the heart of an international system of investment and production that connects these various regions. In the 1960s and 1970s it played a crucial role in the development of a world economic system. The U.S. passed legislation aimed at opening its own and other countries' economies to the flow of capital, goods, services, and information. This central military, political, and economic role contributed, I argue, both to the creation of conditions that mobilized people into migrations, whether local or international, and to the formation of links with the United States that subsequently served as often unintended bridges for international migration. Measures commonly thought to deter emigration--foreign investment and the promotion of export-oriented growth in developing countries--seem to have had the opposite effect. Among the leading suppliers of immigrants to the United States in the 1970s and 1980s were several newly industrialized countries of South and Southeast Asia whose high growth rates are generally recognized to be a result initially of foreign investment in export manufacturing.

That migrations are patterned is further reflected in figures on the U.S. share of global immigration. Though inadequate, the evidence compiled by the United Nations Demographic Yearbook and World Population Prospects shows that in the mid-1980s the United States received about 19 percent of global emigration. This figure is derived from data on permanent settlement, which excludes illegal immigration and unofficial refugee flows between countries, a growing category. A breakdown by region and country of origin shows a distinct pattern. The United States received 27 percent of total Asian emigration but 81.5 percent of all Korean emigration and nearly 100 percent of emigration from the Philippines. It received 70 percent of Caribbean emigration but almost 100 percent of emigration from the Dominican Republic and Jamaica and 62 percent from Haiti. And it received 19.5 percent of all emigration from Central America but 52 percent of emigration from El Salvador, the country in the region with the greatest U.S. involvement.

One could generalize from these tendencies that immigration flows take place within specific systems. The economic ties outlined here represent one possibility. In other cases, the system within which immigration takes place is political or ethnic. One could ask, for example, if systemic links underlie the current Central European migrations to Germany and Austria. Rather than the push factors of poverty and unemployment and the general failure of socialism, other links may operate as bridges. This seems likely given that both Berlin and Vienna before World War II received large migrations from a vast eastern region. Thus migration systems were produced and reproduced. In addition, the aggressive campaign during the cold war to depict the West as a place where economic well-being is the norm and well-paying jobs are easy to get must also have induced people to migrate westward. A more accurate portrayal of conditions in the West might have deterred migrants who were not determined to come at any cost. These historical and current conditions define the systems within which post-1990 eastern migration to Germany and Austria is taking place.

The claim that there is a geopolitics of migration is supported by some of the immigration patterns in Europe. Sixty percent of foreign residents in the United Kingdom are from Asian or African countries that were former dominions or colonies; European immigrants are relatively few, and almost three-quarters of them come from Ireland, another former colony. Almost no immigrants come from Turkey or Yugoslavia, which provide the largest share to Germany. Almost all immigrants to Europe from the Indian subcontinent and from the English Caribbean live in the United Kingdom.

In the ten years following World War II, a vast majority of so-called immigrants to West Germany were the eight million displaced ethnic Germans who resettled there. Another three million came from the gdr before the Berlin Wall was erected in 1961. Almost all ethnic Germans went to Germany, and those who did not went overseas. But 86 percent of Greek immigrants in Europe, almost 80 percent of Turkish immigrants, and 76 percent of Yugoslavs, also reside in Germany.

Almost all Algerians in Europe reside in France, as do 86 percent of Tunisians and 61 percent of Moroccans. Almost all immigrants in Europe from overseas territories still under French control--such as the French Antilles, Tahiti, and French Guiana--live in France. But so do 84 percent of Portuguese

and Spaniards who dwell in Europe outside their countries of origin. France has a long history--going back to the 1800s--of recruiting and receiving migrant workers for its vineyards from these countries.

The Netherlands and Belgium both hosted significant numbers of immigrants from their former colonial empires. They also received foreign workers from what had emerged as labor-exporting countries such as Italy, Morocco, and Turkey. Similarly, workers immigrate to Switzerland from traditional labor-exporting countries: Italy, Spain, Portugal, Yugoslavia, and Turkey. All three countries originally recruited these workers, until eventually a somewhat autonomous set of flows was in place. Sweden receives 93 percent of Finnish immigrants and is expanding its recruitment area to include workers from the traditional labor-exporting countries on the Mediterranean.

As a labor migration flow ages, its destinations tend to become more diverse. A limited autonomy from older colonial and neocolonial bonds develops. Immigrants from Italy and Spain are now distributed among several countries. One-third of Italian immigrants in Europe reside in Germany, 27 percent in France, 24 percent in Switzerland, and 15 percent in Belgium. That destinations continue to be limited, however, signals the presence of migration systems rather than mere individual choices. More recent labor migrations still show very high levels of geographic concentration. Turks make up the largest single immigrant group in any of Europe's labor-receiving countries today, with 1.5 million in Germany.

It seems that once an area becomes a significant exporter of labor it does not easily catch up in development with labor-importing areas precisely because the latter have high, or relatively high, growth. Advantage appears to accumulate. History suggests that labor-sending areas cannot attain or are structurally unable to participate in such advantage because the spatial pattern of growth is uneven. Although generalizations are dangerous, it is clear that Italy and Ireland, which now receive immigrants, have for two centuries been labor exporters and that this has not been a macroeconomic advantage, even though individuals and localities may have benefited.

The case of Japan is interesting here. It reveals the intersection of economic internationalization and the beginning of immigration in a country with a radically different history, culture, and, to a lesser extent, economic organization from those of other advanced economies. 18 One evident difference is Japan's lack of an immigration history. However, Japan now has a growing unauthorized immigrant workforce filling low-wage, unskilled jobs that Japanese youth reject. Processes long since established in the United States and Western Europe are becoming apparent there.

Why is this happening now rather than during the 1950s and 1960s, the period of rapid economic growth when Japan experienced sharp labor shortages? Japan is a major presence in a regional Asian economic system where it is the leading investor, foreign-aid donor, and exporter of consumer goods (including cultural products). And while Japan is not as open to foreign firms as the United States is, such businesses are increasingly common.

Is the new immigration to Japan related to these processes of internationalization? I have argued before that this trend is part of the globalization of Japan's economy. 19 It is easy to recognize in the case of high-level foreign manpower in Tokyo's financial industry. It is less clearly manifested in the new, mostly unauthorized immigration of manual workers, who are employed in construction, manufacturing, and low-wage service jobs. There, internationalization is the context in which bridges are built to the homelands of potential emigrants and contributes to making the Japanese economy more porous, particularly in Tokyo, Yokohama, and Osaka, where most immigrant workers, including the undocumented, are concentrated.

Discussions of immigration policy customarily treat the flow of labor as the result of individual actions, particularly the individual's decision to migrate in search of better opportunities. Such a view puts all the responsibility for immigration on immigrants. Commentators who speak of an immigrant "influx" or "invasion" treat the receiving country as passive: immigration is unconnected to the past or current actions of receiving countries, and immigration policy is portrayed as more or less benevolent toward immigrants. Absent is any awareness that the international activities of the governments or firms of countries receiving immigrants may have contributed to the formation of economic links with emigration countries, links that may invite the movement of people as well as capital.

Similar migration processes are now forming in all major advanced economies at the intersection of economic internationalization with labor markets. Locating the origin of immigration flows and their continuation here points up important parallels in advanced economies whose history and culture differ significantly. The parallels result from their transnational economic influence and from the economic restructuring evident in all advanced economies in the 1980s. The differences stem from the specifics of each country's culture and history.

Implications for Immigration Policy

In all the highly developed countries (and in many of the developing countries), the state has participated in constructing a global economic system and furthering a consensus to pursue this objective. This participation has affected the power of different agencies within the state and advanced the internationalization of the interstate system. Thus, although the state continues to play the most important role in immigration policy making and implementing, it is no longer sufficient simply to examine its formal role in this arena; it is also necessary to examine the transformations of both the state itself and the interstate system and what these changes can entail for migration policy and the regulation of migration flows and settlement.

There is some consensus among policymakers and scholars that a growing gap exists between immigration policy intent and immigration reality in the major developed receiving countries. 20 One possible explanation for this gap is that the limited effectiveness of immigration policy today is partly due to its neglect of these transformations in the larger context of international migration and the institutional apparatus for its regulation. Immigration policy continues to be characterized by its formal isolation from other major processes, as if it were possible to handle migration as a bounded, closed event. States' internationally recognized powers to police their frontiers and to control admission to and exclusion from their territories do not necessarily guarantee state insularity. 21

Current immigration policy in the highly developed countries is increasingly at odds with other major policy frameworks in the international system. There is a combination of drives to create border-free economic spaces yet intensify border control to keep immigrants and refugees out. The juxtaposition between these two dynamics provides one of the principal contexts in which today's efforts to stop immigration assume their significance. There are, in effect, two major epistemic communities: one dealing with the flow of capital and information; the other with immigration. Both are international, and both enjoy widespread consensus in the community of states.

The coexistence of such different regimes for capital and for immigrants has not been seen as an issue in the United States. For this reason, the case of the eu is of more interest here: it represents an advanced stage of formalization, and European states are discovering the difficulties if not impossibility of maintaining two such diverse regimes. The European Community and the national governments of the member states struggled to handle the juxtaposition of the divergent regimes for immigration flows, on the one hand, and all other types of flows, on the other. The discussion, design, and implementation of policy aimed at forming a European Union make it evident that immigration policy has to accommodate the facts of rapid economic internationalization. The other major regional systems in the world are far from facing this contradiction in designing their formal policy frameworks, and they may never do so, yet even they must reconcile the conflicting requirements of border-free economies and border controls to keep immigrants out. NAFTA is one such instance, as are, in a more diffuse way, various other initiatives for greater economic integration in the Western Hemisphere. There are also the regional systems constituted partly as zones of influence of major economic or geopolitical powers, such as the long-term dominance of the United States over the Caribbean Basin. The quasi-transnational economic integration characterizing such systems produces its own contradictions between drives for border-free economic spaces and border control to keep immigrants and refugees out.

There are strategic sites where it becomes clear that the existence of two very different regimes for the circulation of capital and the circulation of immigrants poses problems that cannot be solved through the old rules of the game, where the facts of transnationalization weigh in on the state's decisions regarding immigration. One example relates to the need to create special systems to govern the circulation of service workers within both the GATT and NAFTA as part of the further internationalization of trade and investment in services. Much of what we call international trade and investment in services actually involves specialized service workers; they embody the service and need to cross borders to deliver it. Hence provisions must be made for their international circulation. These labor circulation systems have been uncoupled from any notion of migration, even though they involve a version of temporary labor migration, and they are in good part under the oversight of entities that are quite autonomous from the government. Another instance can be seen in Japan's new immigration law, passed in 1990 (actually an amendment on an earlier law on the entry and exit of aliens), which opened the country to several categories of highly specialized professionals with Western backgrounds (e.g., experts in international finance, Western-style accounting, Western medicine, etc.) and made the entry of what is referred to as "simple labor" illegal. 22 In other words, the law permits the importation of the commodity "Western human capital" but closes the borders to traditional immigrants.

Human Rights and State Sovereignty

Against this web of obligations and in view of the multiplying actors involved in the immigration debate and policy world, one development emerges as singularly important for understanding the impact of immigration on questions of sovereignty and territoriality. This is the emergent international human rights regime. Human rights are not dependent on nationality, unlike political, social, and civil rights, which are predicated on the distinction between national and alien. Human rights override such distinctions and hence can be seen as potentially contesting state sovereignty and devaluing citizenship.

International human rights, while partly rooted in the founding documents of certain nation-states, are today a force that can undermine the exclusive authority of the state over its nationals and thereby contribute to transforming the interstate system and international legal order. Membership in territorially exclusive nation-states ceases to be the only ground for the realization of rights. All residents, whether citizens or not, can claim their human rights. 23 Human rights begin to impinge on the principle of nation-based citizenship and the boundaries of the nation. How does immigration fit into this, particularly undocumented immigration, which carries with it a de facto erosion of the state's sovereign power to control ingress? This is a complex question that I can only begin to address here. First, let me examine some of the instruments available and the actual possibilities for individuals and nonstate actors to use these instruments.

The concept of inalienable rights of the individual as a universal principle was already present in the French and American revolutions. Central to its interpretation was the notion that such rights could be safeguarded and realized only within the context of a nation-state. References to inalienable cross-national rights in the founding documents of nation-states were highly abstract. Already in the early twentieth century, several legal instruments promoted human rights and made the individual an object of international law. But elaboration and formalization of these rights did not occur until after World War II. The covenants and conventions that guarantee human rights today are derived from the Universal Declaration of Human Rights adopted by the un in 1948. The formulation of human rights is partly drawn from the Declaration of Independence, the U.S. Constitution, and the Declaration of the Rights of Man and Citizen. 24

To what extent can individuals and nonstate groups make claims on the state, particularly in the United States and Western Europe, where the human rights regime is most developed? The Universal Declaration of Human Rights is not an international treaty and thus is not legally binding. But because it is so often cited many consider it to have the status of customary international law, an international and general practice that is accepted and observed as law. The International Covenant on Civil and Political Rights and the International Covenant on Economic, Social, and Cultural Rights legislated much of what the Universal Declaration called for. Ratification procedures started in 1966, but it was ten years before thirty-five states ratified the covenants, the number required to make them legally binding. In 1976 a Protocol to the Covenant on Civil and Political Rights was opened for ratification; it enables private parties to file complaints to the un Human Rights committee if a state that has ratified the protocol has violated its dictates. Other un human rights treaties include the Convention on the Prevention and Punishment of Genocide (1948), the International Convention on the Elimination of All Forms of Racial Discrimination (1965), and the Convention on the Political Rights of Women (1981). These instruments are only binding on states that have ratified them, although some instruments have taken on the character of international customary law and are binding on all states.

There are also regional conventions. The European Convention on Human Rights, adopted in 1950, was established by the member states of the Council of Europe to fulfill the goals of the Universal Declaration. The inter-American system for the protection of human rights, the Inter-American Commission on Human Rights, is grounded on two distinct legal documents: the Charter of the Organization of American States and the American Convention on Human Rights, which were both adopted in 1969 and entered into force in 1978. The human rights regime of the oas was markedly strengthened in a 1967 protocol that came into force in 1970.

In the earlier postwar phase of the elaboration of human rights instruments, the state was still the object and subject of international law. It was up to a state to lodge complaints about human rights violations to the un or pertinent regional organs. In many ways these instruments lacked weight and influence. By the 1970s a marked transformation became evident. More instruments were available, and they started to be used more frequently; human rights activists also mention the strong support of high-ranking politicians, such as presidents Carter (United States), Arias (Costa Rica), and Alfonsin (Argentina), as well as the growing integration of communications media across the globe, which facilitates instantaneous reporting of violations. Increasingly, individuals and nonstate actors petitioned governments on the basis of international human rights codes or regional instruments, and these instruments became partly independent of the states themselves even though they were grounded in interstate agreements. This growing authority of human rights law is particularly evident in Europe. It was not until the 1980s that such law began to exert significant influence in the United States, where it still does not carry the weight it has in Europe.

The European Convention on Human Rights has taken on considerable importance. As with other human rights instruments, it bestows rights on "persons" rather than citizens, and declares that the enjoyment of rights and freedoms it contains do not depend on, among other criteria, national origin. However, the convention does not prohibit distinctions between nationals or citizens and aliens. Provisions in the convention and in the Court of Human Rights rules authorize individuals and nonstate actors to petition. Such cases increased rapidly in the 1970s and 1980s. 25 Several states have incorporated many of the convention's provisions into their domestic law, among them, Germany, Netherlands, France, Spain, Switzerland, and Turkey. In these cases, decisions by the court have a direct effect on domestic judiciaries, which emerge as key organs for the implementation of human rights provisions. This pattern has grown markedly since the early 1980s with the growth of court case law.

In the United States, this process has been much slower and less marked. This is partly due to U.S. definitions of nationhood, which have led courts in some cases to address the matter of undocumented immigrants within U.S. constitutionalism, notably the idea of inalienable and natural rights of people and persons, without territorial confines. The emphasis on persons permits interpretations about undocumented immigrants that would not be possible if the emphasis were on citizens. It was not until the mid-1970s and early 1980s that domestic courts began to consider human rights codes as normative instruments in their own right. The rapid growth of undocumented immigration and the sense that the state was incapable of controlling the flow and regulating the various categories of immigrants was a factor leading courts to consider the international human rights regime: it allows courts to rule on basic protections of individuals not formally covered in the national territory and legal system, notably undocumented aliens and unauthorized refugees. 26

Several court cases show how undocumented immigration creates legal voids that are increasingly filled by invoking human rights covenants. 27 In many of these cases, individual or nonstate actors bring claims based on the notion of international human rights codes as expanding international law. The state, in this case the judiciary, "mediates between these agents and the international legal order." 28 Courts have emerged as central institutions for a whole series of changes. 29

Perhaps one of the most important documents seeking to protect the rights of migrants is the 1990 International Convention on the Protection of the Rights of All Migrant Workers and Members of Their Families adopted by the General Assembly of the un in 1990. 30 It took a decade to elaborate and approve this convention. It also contains a broad range of explicit human rights protections for undocumented migrants and members of their families to an extent not seen before in other instruments, with the exception of an International Labor Organization convention that guaranteed equality of treatment for undocumented workers in various employment-related matters; this agreement, however, covered only past employment. 31

Immigration, Human Rights, and State Sovereignty

The growing accountability of the state to international human rights codes and institutions, together with the fact that individuals and nonstate actors can make claims on states based on those codes, signals a development that goes beyond the expansion of human rights within the framework of nation-states. It contributes to a redefinition of the bases of the legitimacy of states under the rule of law and the notion of nationality.

From an exclusive emphasis on the sovereignty of the people and right to self-determination, a shift to rights of individuals regardless of nationality has occurred. Human rights codes can thus erode the legitimacy of the state if states fail to respect such human rights. It is no longer just a question of self-determination but of respect for international human rights codes. The extent to which the organizations and instruments that privilege these individual rights are likely to be implemented is uncertain. One interpretation views these developments as indicating that international law today makes the individual and nonstate groups subsidiary to the laws between states.

In accumulating social, civic, and even some political rights in countries of residence, immigrants have diluted the meaning of citizenship and the specialness of the claims citizens can make on the state. When it comes to social services (education, health insurance, welfare, unemployment benefits) citizenship status is of minor importance in the United States and Western Europe. What matters is residence and legal alien status. Most of these countries will pay retirement benefits even if recipients no longer reside there. Some countries--for instance, Sweden and the Netherlands--have also granted local voting rights. EU immigrants have the right to vote in European parliamentary elections, but non-eu immigrants do not. Aliens are guaranteed full civil rights either constitutionally or by statute. Given the little difference between the claims that citizens and immigrants can make, immigrants have little incentive to seek naturalization.

Even unauthorized immigrants can make some of these claims. Peter Schuck has noted that new social contracts between undocumented aliens and society-at-large are being negotiated in the United States every day, contracts that cannot be nullified through claims concerning nationality and sovereignty. Courts have had to accept the existence of undocumented aliens and to extend to these aliens some form of legal recognition and guarantees of basic rights. Various decisions have conferred important benefits of citizenship upon undocumented aliens. This clearly undermines older notions of sovereignty. At the same time, California's Proposition 187 signals the possibility of reversals in this arena. Although Proposition 187 will be challenged in the courts, its success at the polls signals the political/legal potential for anti-immigrant sentiment.

For scholars who see in the new international human rights regime a major political development, 32 the erosion in the distinction between citizen and alien also devalues the institution of citizenship. 33 This devaluing may well contribute in turn to the burgeoning importance of international human rights codes. The growing ability of ngos and individuals to make claims on the basis of international human rights instruments has implications beyond the boundaries of individual states. It affects the configuration of the international order.

The concept of nationality is being partly displaced from a principle that reinforces state sovereignty and self-determination (through the state's right/power to define its nationals) to a concept emphasizing that the state is accountable to all its residents on the basis of international human rights law. The individual emerges as the object of international law and institutions. International law still protects sovereignty and has in the state its main object, but the state is no longer the only subject of international law. In addition to all its other functions, it becomes an institutional apparatus of a transnational order based on human rights. In this process, the relationships among nongovernment entities, the state, and international institutions is reconfigured. Individuals and NGOs can make claims on the state--the one where they reside or another--and they participate in the debates and actions concerning the international legal order. That the state's jurisdictional and judicial role and its relation to the individual and nonstate actors are, in some regards, becoming organized by the international human rights legal order is clearly not an irreversible trend, as current events in the former Yugoslavia indicate, but this change does create new conditions that any international legal order must accommodate. Even strong nationalist or ethnic resistance must confront the undeniable existence of the international human rights regime.

These and other developments point to an institutional reshuffling of some of the components of sovereign power over entry and can be seen as an extension of the general processes whereby state sovereignty is being partly decentered onto non- or quasi-governmental entities for the governance of the global economy and international political order. Put simply, these developments reduce the autonomy of the state in immigration policy making and multiply the sectors within the state that are addressing immigration policy, thereby multiplying the opportunities for conflicts within the state. Immigration policy is now shaped by forces ranging from economic globalization to international agreements on human rights, and it is made and implemented within settings ranging from national and local legislatures and judiciaries to supranational organizations.

Policy making regarding international issues has always engaged different parts of the government. Each state is constituted through multiple agencies and social forces. Indeed, it could be said that although the state has central control over immigration policy, exercising that power often begins with a limited contest between the state and interested social forces: agribusiness, manufacturing, humanitarian groups, unions, ethnic organizations, and zero-population-growth efforts, to name a few. 34 Today the old hierarchies of power and influence within the state are being reconfigured by increasing economic globalization and the ascendance of an international human rights regime. 35


Note 1: A vast and rich scholarly literature documents and interprets the specificity and distinctiveness of immigration policy in highly developed countries; see, e.g., Weil 1991; Cornelius, Martin, and Hollifield 1994; Weiner 1995; Soysal 1994; Thranhardt 1992; Bade 1992 to mention just a few. As a body, this literature reveals the many differences among these countries. Back.

Note 2: Refugee policy in some countries does lift the burden of immigration from the immigrant's shoulders. U.S. policy, for example, particularly for Indochinese refugees, acknowledges partial responsibility on the part of the government. Clearly, in the case of economic migration, such responsibility is far more difficult to establish and by its nature is far more indirect. Back.

Note 3: I should note that the right to entry is part of the 1969 Convention on Refugee Problems in Africa adopted by the Organization of African States. Back.

Note 4: Plender 1988, 159-91. Back.

Note 5: For instance, major human rights codes place procedural restrictions on lawful expulsions; bilateral and regional accords for employment migration tend to contain provisions through which member states relinquish some control over the entry and exit of foreigners. See Plender 1988; Goodwin-Gill 1988. Back.

Note 6: See, e.g., Hollifield 1992; Baubock 1994; Sassen forthcoming. Back.

Note 7: The efforts that mix the conventions on universal human rights and national judiciaries assume many different forms. Some examples in the United States are the sanctuary movement in the 1980s, which sought to establish protected areas, typically in churches, for refugees from Central America; judicial battles such as those around the status of Salvadorans granted indefinite stays though formally defined as illegal; and the fight for the rights of detained Haitians in an earlier wave of boat lifts. It is clear in these cases that notwithstanding the lack of an enforcement apparatus, human rights limit the discretion of states in how they treat non-nationals on their territory. It is also worth noting in this regard that the United Nations High Commissioner for Refugees (unhcr) is the only un agency with the authority to intercede worldwide on behalf of those in danger. Back.

Note 8: While these developments are well known in Europe and North America, there is not much general awareness that incipient forms are showing up in Japan as well (see, e.g., Sassen 1991, chap. 9; Shank 1994). For instance, in Japan today a strong group of human rights advocates is working on behalf of immigrants; nonofficial unions are trying to organize undocumented immigrant workers; and organizations working on behalf of immigrants are receiving funding from individuals or government institutions in sending countries (e.g., the Thai ambassador to Japan announced in October 1995 that his government will give a total of 2.5 million baht, about 100,000 U.S. dollars, to five civic groups that assist Thai migrant workers, especially undocumented ones; see Japan Times, October 18, 1995; see also Sassen 1993). Back.

Note 9: Further, the growth of immigration, refugee flows, ethnicity, and regionalism challenges the accepted notion of citizenship in contemporary nation-states and hence the formal structures for accountability. My research on the international circulation of capital and labor caused me to review the meaning of such concepts as national economy and national workforce under conditions of growing internationalization of capital in both developed and developing countries and the burgeoning presence of immigrant workers in major industrial countries. Furthermore, the rise of ethnicity in the United States and Europe among a mobile workforce raises questions about the content of the concept of nation-based citizenship. The portability of national identity gives rise to concerns about bonds with other countries, or localities within them, and the resurgence of ethnic regionalism creates barriers to the political incorporation of new immigrants. See, e.g., Soysal 1994; Baubock 1994; Sassen forthcoming). Back.

Note 10: There is a large and rich literature on the development of immigration policy at the European level; please refer to n. 1, above, for a few citations. Longer bibliographies and analyses of the particular angle under discussion here limitations on the autonomy of the state in making immigration policy can also be found in Sassen forthcoming. Back.

Note 11: Mitchell 1989. Back.

Note 12: Diverse social forces shape the role of the state depending on the matter at hand. Thus in the early 1980s bank crisis, for instance, the players were few and well coordinated; the state basically relinquished the organizing capacity to the banks, the imf, and a few other actors. It was all very discreet indeed, so discreet that the government was hardly a player in the crisis. This is quite a contrast with the deliberations around the passing of the 1986 Immigration and Reform Control Act, which was a national brawl. In trade liberalization discussions, too, there are often multiple players, and the executive may or may not relinquish powers to congress. Back.

Note 13: Aman (1995) has noted that although political and constitutional arguments for reallocating federal power to the states are not new, the recent reemergence of the Tenth Amendment as a politically viable and popular guideline is a major political shift in the relations between the federal government and the states since the New Deal. Back.

Note 14: Reimers 1983; Briggs 1992. Back.

Note 15: Sassen 1988, 1993; Journal fur Entwicklungspolitik 1995; Bonacich et al. 1994. Back.

Note 16: Mahler 1995. Back.

Note 17: Sassen forthcoming. Back.

Note 18: For a more detailed account, see Sassen 1993; Shank 1994. Back.

Note 19: See Sassen 1993. Back.

Note 20: E.g., Cornelius, Martin, and Hollifield 1994. Back.

Note 21: Bosniak 1992, 745. Back.

Note 22: Sassen 1993. Back.

Note 23: Jacobson 1996; Reisman 1990. Back.

Note 24: Henkin 1990, 144. Back.

Note 25: It is interesting to note that in the International Court of Justice it is very rare for a judge to vote against the position of his or her government in disputes before the court. This is not at all uncommon, however, in the European Court of Human Rights; indeed, it is becoming more common. This is significant because the European Court has become the main organ for the interpretation of the European human rights convention's provisions. Back.

Note 26: For instance, the Universal Declaration was cited in 76 federal cases from 1948 through 1994; over 90 percent of those cases took place after 1980, and, of those, 49 percent involved immigration issues, a proportion that rises to 54 percent if cases involving refugees are included (Jacobson 1996, 97). Jacobson also found that the term human rights was cited in 19 federal cases before the twentieth century, in 34 from 1900 to 1944, in 191 from 1945 to 1969, in 803 in the 1970s, and in more than 2,000 in the 1980s. He estimates that such references will have been made in more than 4,000 cases by the end of the 1990s. Back.

Note 27: See, e.g., Hassan 1983. Back.

Note 28: Jacobson 1996, 100. Back.

Note 29: See Shapiro 1991. Back.

Note 30: See Bosniak 1992 for an in-depth examination of this convention and its potentials and limitations for the protection of undocumented migrants. Back.

Note 31: It is perhaps worth noting that after much debate, undocumented migrants were also included under the protections provided by the un General Assembly's 1985 Declaration on the Human Rights of Individuals Who Are Not Citizens of the Countries in Which They Live. But being a nonbinding instrument it is probably of little use to undocumented migrants. Back.

Note 32: See, e.g., Jacobson 1996; Soysal 1994. Back.

Note 33: Jacobson 1996. Back.

Note 34: Cf. Mitchell 1989. Back.

Note 35: An example is the ascendance of so-called soft security issues: according to some observers, recent government reorganization in the State Department, the Department of Defense, and the cia reflects an implicit redefinition of national security. Back.