Discussions of immigration policy often focus on borders: on who should be allowed into a country. I want to focus instead on the people who have already arrived, especially those who are likely to stay—on the men, women, and children who have crossed political boundaries to live in states of which they are not citizens. Public policies addressed to this group are bound to vary, but we must reject the popular idea that anything goes. Political morality sets basic limits on the treatment of immigrants, whether legal or not.

Lying at the heart of this argument is an idea about the moral importance of membership. Whatever their legal status, individuals who live in a society over an extended period of time become members of that society, as their lives intertwine with the lives of others there. These human bonds provide the basic contours of the rights that a state must guarantee; they cannot be regarded as a matter of political discretion.

Whatever their legal status, individuals who live in a society over an extended period of time become members of that society.

The claims I will make—that most immigrants and their children have a moral right to citizenship and that resident non-citizens ought to possess most of the legal rights of citizens—may sound radical in comparison with conventional views of state sovereignty and democratic self-determination. And they would require some significant departures from current practices in Europe and North America. Nevertheless, I will argue that they follow from principles already widely acknowledged and practices already widely adopted in democratic states.


The idea that some individuals living in a state are morally entitled to be citizens may seem on its face undemocratic. In a democracy, it might be said, the people are supposed to rule, and a basic element in their ruling—some might say the most basic element—is the right to decide who belongs.

According to Aristotle’s Politics, this was the Greek view of democratic citizenship. The allocation of citizenship lay entirely within the discretion of the political community and was to be expanded or contracted in accordance with the interests of the regime. For example, the requirement that citizenship be hereditary might be changed in accordance with the needs—primarily the military needs—of a particular city-state. If there were sufficient citizens, Aristotle suggested, it would probably be best to restrict citizenship to those whose parents and grandparents were citizens. Aristotle did not say clearly why he thought such a restriction of citizenship desirable. (Perhaps he assumed that the deeper the citizen heritage, the stronger the loyalty and commitment to the city, although ancient Greece was full of examples of people like Alcibiades and Thucydides who wound up fighting for the enemies of the city in which they were citizens.) If the city needed more citizens, it could relax these requirements, granting hereditary citizenship to those with a non-citizen grandparent or even with a non-citizen parent—and then later tighten them again as needs diminished. So, Aristotle thought that no one had a right to citizenship, and if some people had no citizenship at all (as could easily happen under such practices) that would be a personal misfortune but no injustice.

A democratic society has an obligation to respect the political agency of those who live within it.

No democratic state today could follow Aristotle’s approach. The idea that there are moral limits on the granting of citizenship—that citizenship is not simply a matter of political discretion—is essential to our modern understanding of democracy.

To be sure, modern states have excluded people from citizenship in the name of certain visions of the political community. In 19th-century Europe, the question of whether Jews could be citizens was often answered in the negative. In North America, people of African and Asian descent and indigenous people were excluded from citizenship or denied access to citizenship on the same terms as others. Until the last few decades most states did not give women the same right as men to pass on their citizenship to their children and often revoked their citizenship when they married, sometimes regardless of whether they had acquired their husband’s citizenship. But would anyone now claim that such policies are morally permissible as long as they are made by a government duly elected by a majority?

In a modern state, democratic legitimacy requires the consent of the governed—all those adults permanently subject to the state’s laws—in ways that ancient-democratic legitimacy did not. To exclude people from citizenship is to fail to treat them as free moral agents with a right to participate in the collective determination of the laws to which they are subject, with which they are expected to comply, and which profoundly shape the social world in which they live.

The conflict between democratic legitimacy and the exclusion of some from citizenship is clearest in the case of the children and grandchildren of immigrants, because it is not possible to pretend that their presence is the product of a choice in which they gave up the right to participate in exchange for other opportunities.

When migrants first arrive, it is possible to say that they are only passing through—and, in fact, many do go back. But no one can pretend that the grandchildren of migrants—people who have spent their entire lives in a society and are the children of people who have spent their entire lives in that society—are just passing through. So the third generation must be included in the citizenry; and once one recognizes this, it becomes equally clear that the members of the third generation must be considered citizens from birth, just like the children of citizens, so they can be socialized into democratic norms and practices as they grow up.

But if the third generation deserves citizenship, why not the second generation—those born and raised in the society (or even those only raised there from a young age)? By the time they are adults, it is clear that they are not just passing through, and so they, too, are entitled to be citizens, and from the outset. No plausible contemporary conception of democracy can justify denying rights of democratic participation to people who have spent their entire lives in a society. The possibility that they might leave is insufficient because this is also true of the children of current citizens.

Germany, for example, has only recently extended citizenship at birth to the descendents of immigrants; but for decades leading German politicians both on the right and the left had acknowledged that democratic legitimacy required as much. It was simply not possible to justify excluding an identifiable segment of the population from political participation for generation after generation.

For any non-citizen who becomes entangled with legal authorities, the threat of deportation looms in the background.

What about dual citizenship? The fact that the children of immigrants inherit another citizenship sometimes supplies a pretext for denying them citizenship where they live. But this works only for a while, because it is clear that the strongest moral claim to citizenship is the claim that one must be treated as a full member of the political community in the society in which one lives.

The same basic principle applies to the process of naturalization: a democratic society has an obligation to respect the moral and political agency of those who live within it. So migrants must be included in the political process once it becomes clear that they are not just passing through and once they have had a chance to become familiar with the institutions and practices of the society. This means that naturalization requirements should be clear and limited and that immigrants should be legally entitled to naturalize once these requirements are met. This stands in contrast to what had been the naturalization norm in Germany and many other European states in years past: laws that treated the acquisition of citizenship as the end of the integration process, not a step along the way; that required proof of the immigrant’s having adapted sufficiently to the new society; and that relied upon the discretionary judgment of state officials rather than treating citizenship as a matter of individual entitlement.

Almost all European states now have policies that meet or come close to meeting the norms I have just identified. Children born or raised in a country acquire citizenship, either automatically at birth or upon a simple declaration by their immigrant parents or by the individual herself at an appropriate age. Immigrants themselves can acquire citizenship as a matter of right upon satisfying a modest set of legal requirements.

What sorts of requirements may legitimately be imposed as conditions of naturalization? The normal freedoms of a democratic society—freedom of religion and conscience, freedom of speech and association, the right to privacy and, more generally, the right to live one’s life as one chooses as long as one does not violate the law—set severe limits on what may be demanded as conditions of naturalization. The main thing it is reasonable to expect is the passage of enough time to become acclimated. Three to five years seems a reasonable length of time, and that is what most states require, though some require up to ten. The most common and most plausible additional requirement is a modest knowledge of the language of public life. I would argue that after enough time has passed (ten years at most) even this should be waived because if a person has been able to live and function in the society for ten years, he or she will be able to get enough information to participate in the democratic process.

Are non-citizens entitled to all the rights that citizens enjoy?

Some states (including Canada and the United States) also impose a requirement that the person not have a serious criminal record. This is a reasonable concern, but if the criminal record is not significant enough to warrant deportation, neither should it warrant permanent exclusion from the democratic process. Some states (including the United States) also have a good-character requirement. This is an invitation to discretionary abuse, though it is rarely invoked. Some European states (including France, Germany, and Italy) require proof of a certain level of income or the absence of reliance upon social assistance, but this seems to me a form of discrimination against the poor. If a person is entitled to some form of social assistance, she should not then be penalised politically for taking advantage of it.

Only a few states (five of the 20 European and North American states, according to a recent survey carried out by Patrick Weil) require renunciation of previous citizenships, and some of those (including the United States) do not try to enforce the requirement. The arguments in favor of accepting dual citizens outlined above apply here as well, if perhaps less strongly. The main point is that migrants often have very good reasons for wanting to maintain their original citizenship, and this often implies nothing about their attachment or lack of attachment to the state to which they have moved. Nothing would do more to foster the political integration of immigrants than the elimination of prohibitions on dual citizenship, something that is particularly important in the EU, where other important rights are often attached to EU citizenship.

A few states require that those seeking citizenship pass a test demonstrating their knowledge of the history or institutions of the new country. (Only four states, including Canada and the United States, required such tests at the time of Weil’s survey, but a couple have added such a requirement since, and the United States is presently seeking to make this a more significant requirement.) If the pass level for these tests is set too high, it becomes a barrier to integration, and if set appropriately low, it is possible to pass without knowing very much. The test itself, then, is rarely of much use, but studying for it can be a valuable socialization experience for some. The key is to construct enough alternatives that it does not become a barrier. Something similar might be said about the naturalization ceremony. If this involves some sort of welcome by the state and affirmation of commitment, respect for the constitution, and so on, by the new citizen, it can be a valuable opportunity to build identification with the political community. If it requires some sort of loyalty oath that entails a renunciation of prior commitments, it may well be counterproductive—and in any event, it would be an unreasonable demand.


Citizenship means full membership in a political community, and, according to any modern conception of democracy, this citizenship must be equal: generally speaking, all citizens must enjoy equal rights. But not everyone who is physically present in a society becomes a citizen, at least not right away. What rights are the non-citizens entitled to?

Short-term visitors. Even tourists or short-term visitors are morally entitled to some legal rights, such as security of the person and security of property. If they are accused of a crime, they are entitled to a fair trial. Standards of fairness may legitimately vary from one country to another to some degree, but it would not be morally permissible to use one set of standards for citizens and a different one for non-citizens. For example, if hearsay evidence cannot be used in a criminal case against citizens, it must not be used in a criminal case against non-citizens.

How do things work in practice? The legal systems of Europe and North America all establish many important legal rights that are to be enjoyed by persons regardless of citizenship status. This formal commitment matters, even if it does not settle everything, because it recognizes the principle that there are universal legal rights. In practice, of course, non-citizens are vulnerable to state authorities in ways that citizens are not, because their right to be present is much less secure and this affects their ability to enjoy the legal rights to which they are entitled. This is clearest in the case of non-citizens who are present without the authorization of the state. They fear (with good reason) that any involvement with state authorities may invite deportation. So, for example, if they are the victims of a crime, they are unlikely to report it.

Often, those who are deported have no ties to the country to which they are sent.

Even those non-citizens authorized to be present may find that their effective enjoyment of legal rights is reduced because they are vulnerable to investigations of their compliance with immigration law. Authorities are sometimes able to take actions against non-citizens (e.g., extended detention, questioning without lawyers) under the cover of enforcing immigration law that would be prohibited otherwise. (A particularly vivid illustration of the phenomenon was the roundup and extended detention of young Arab and Muslim males in the United States in the aftermath of the September 11 attacks and the subsequent deportation of many of them for technical violations of immigration law that would normally have been overlooked. But for any non-citizen who becomes entangled with legal authorities, the threat of deportation looms in the background and affects her capacity to enjoy and exercise her rights.) These concerns are important but should not be overstated. The rights of non-citizens are not always secure, but neither are the rights of citizens.

Non-citizens are, then, entitled to some rights. But are they entitled to all the rights that citizens enjoy? I want here to assume the conventional moral view that states are entitled to restrict and regulate entry, at least in some ways. In particular, I assume that states are entitled to admit people as tourists and visitors on the understanding that they will leave within a reasonably limited period and will not try to seek employment or gain access to the society’s social programs while they are present. If all states enjoyed roughly comparable political, social, and economic conditions, this would be a plausible moral claim, and many think it is plausible even under contemporary conditions. On this assumption, then, non-citizens admitted for these limited periods and purposes are not morally entitled to all of the same rights as citizens. Again, this corresponds well to contemporary practice: all democratic states admit temporary visitors in vastly higher numbers than they do migrants who plan to stay.

Permanent residents. Now consider the other extreme: people whom the state permits to reside on an ongoing basis. May the state distinguish their legal rights from the rights of citizens?

Not if membership is what matters morally. As people stay longer, their moral claims grow stronger, and after a while they pass a threshold that entitles them to virtually the same legal status as citizens. Once people have been settled for an extended period—say, five years or so—they are morally entitled to the same legal rights (and ought to be subject to the same legal obligations) as citizens, except perhaps for the right to vote and the right to hold high public office. It is permissible to restrict these rights, if access to citizenship is readily available as I have argued it should be. During the early stages of settlement it is permissible (though not required, and, in my view, often not desirable) to limit some legal rights (e.g., to redistributive benefits or protection against deportation) but not most others.

Someone may object that if naturalization is relatively easy, then people who choose not to naturalize forfeit their claim to such a strong legal position. But people have many good reasons for choosing not to naturalize. For example, they may want to be able to return to their country of origin later to care for aging parents. Inheritance laws in the country of origin may require heirs to be citizens. It is not fair to insist that they sacrifice such interests and concerns in order to receive equal treatment as members of civil society here, an equal treatment to which they are entitled by the fact of their membership.

Over the past century, policy has moved significantly in the direction that I am advocating. In most liberal-democratic states in Europe and North America, there has been a significant transformation of the legal status of permanent residents. Once there were many significant legal distinctions between citizens and permanent residents; now there are few. The 1996 American welfare-reform legislation that reduced the rights of permanent residents to some social-welfare programs is only a minor retreat from this general trend of the last century and is likely to affect only a small proportion of the permanent resident population. In Europe, EU nationals enjoy an increasing number of rights that are not available to non-EU nationals who have permanent residence status, and that also marks a modest reversal of the dominant trend. Yet many permanent residents spend their entire lives in states in Europe or North America without becoming legal citizens and without that fact affecting their lives in any significant way apart from their not being able to vote (which many citizens, especially in North America, choose not to do anyway) or get an EU or North American passport.

The unique restrictions that temporary workers face render them much more vulnerable to mistreatment by their employers than ordinary workers.

Some have characterized these expansions of non-citizen rights as a devaluation of citizenship, which raises the question of whether we should try to make citizenship more valuable by reserving more rights for citizens. On the contrary, I think that we should celebrate this historical development and extend it into the three key areas where significant legal differences between residents and citizens persist, or have even grown.

While voting and holding public office can both properly be reserved for citizens (assuming easy naturalization), there are two main areas where citizens almost universally enjoy more rights than residents—public employment and security of residence—and a third, access to social programs, where some people think the differentiation should be much sharper.

Let me take the last idea first. Residents normally enjoy the same rights as citizens to public pension programs, workers’ compensation, unemployment compensation, and other social programs that either tie individual benefits to the level of individual contributions or function as collective insurance schemes. This makes sense because it would hardly be fair to require people to contribute to such programs and then deny them the benefits. Residents also normally enjoy equal access to social programs funded through general taxpayer revenues and aimed at the general public. Access to public libraries, public recreational facilities, and public universities is sometimes tied to residence, rarely to citizenship. And that makes sense, too, because non-citizen residents are taxpayers and members of the general public.

With regard to redistributive social programs, however, people sometimes claim that the citizen–resident distinction matters and ought to count for more. There has been a move lately (as in the 1996 U.S. legislation) to restrict the access of recent immigrants to various forms of social assistance. This is unwelcoming and unwise, but not unjust so long as it applies only to recent arrivals. But long-term residents are entitled to be treated as full members of society, and proposals to ban immigrants forever from social-assistance programs would clearly be unjust.

Let us turn now to the issue of public employment. National-security or major policy-making responsibilities may permissibly be confined to citizens, but restricting all civil-service jobs is unjust. If one accepts the principle that non-citizen residents are normally entitled to be treated equally in the occupational sphere, the reasons are obvious. In Europe, the opening up of public-sector jobs within the EU to citizens of other EU countries has undermined many of the traditional justifications for restrictions (loyalty, acting as an agent of the state, etc.). The ongoing refusal of some EU states to open public employment to their own permanent residents is blatantly discriminatory.

Finally, and perhaps most controversially, it seems to me deeply unjust to deport long-term residents, even if they have been convicted of crimes; this is a violation of their human rights and unfair to the receiving country. We regard it as morally wrong for states to deport their own citizens, even if they can find another state willing to accept them, because we recognize that the right to remain in a society of which one is a member is a fundamental human right. Long-term residents, as members of the societies in which they live, have the same moral right—and should enjoy the same legal right—not to be deported. This should be particularly clear in the case of people who have lived in a society since early childhood.

The New York Times of November 16, 2003, told one story that indicates what is wrong with deporting long-term residents, even if they have been convicted of crimes. Loeun Lun arrived in the United States with his parents at the age of six as a Cambodian refugee. As a teenager, he committed a crime of assault, for which he was convicted and served 11 months in prison. This was the only crime he ever committed, apart from traffic violations. He found work, married, and had two children. Nine years after his conviction, he was arrested and deported (having come to the attention of authorities in the course of applying for citizenship), leaving his wife and two young children behind. Or take the case of Rafael Ramirez, reported in the New York Times of April 4, 1999. Ramirez was then a 35-year-old Army veteran facing deportation for a single conviction for marijuana possession in 1990, shortly after his discharge from the army. Ramirez came to the United States at the age of seven and has a family (including four children) and a small business.

What legal rights should be given to those who reside and work without official authorization?

There are thousands of stories like this about people who came to the United States or Canada as young children, spent all or almost all of their lives there, and were then deported because of a criminal conviction. Often they have no social ties in the country to which they are deported. Sometimes they don’t even speak the language. Nonetheless, many will feel that these are criminals who deserve whatever they get (although it is important legally for the authorities to insist that these deportations are not punishments for the crimes). The counterargument is that we do not do this to our own citizen criminals, and it would be a violation of widely accepted international human-rights norms if we did.

Temporary residents. Sometimes people are not admitted on a permanent basis but rather for a limited time and subject to certain conditions. How may the rights of temporary workers differ from those of citizens and permanent residents?

The first and most important difference in practice is, of course, that temporary workers are entitled to stay only for a limited period, whereas citizens and permanent residents are able to stay indefinitely. This restriction is permissible provided that the workers are actually present only for a short time. It is not acceptable to keep people in a temporary status that is regularly renewed, so that they actually stay for a long time but remain in a vulnerable and restricted situation. The general principle applies here as well: the longer the stay, the stronger the claim to be regarded as a member of society and thus entitled to the rights of membership. If someone comes to work in a place for a year or perhaps even two, she does not have particularly strong claims to remain. After five or six years, however, the moral case is very different. This is so even if people are admitted with explicit limitations on their rights to stay. This is surely the lesson to be drawn from the European experience with “guest workers” (temporary workers) in the 1960s and 1970s. The guest workers had been told from the outset that they could not stay permanently and had agreed to come under those terms. But the European states were unable to send them back when economic conditions changed and they were no longer needed. In the face of their obvious deep connections to the places to which they had moved, the original terms of admission had become irrelevant. As time passes then, temporary residents are entitled to become permanent ones, and with the right to stay comes the right to be treated as a full member.

In a similar vein, seasonal workers who come back year after year have stronger claims than ones who come only for a year or two. And the longer the season, the stronger the claim. It is not acceptable to pretend that permanent workers are merely seasonal ones.

The second important way in which the rights of temporary workers often differ in practice from those of citizens and residents is that temporary workers are usually much more restricted in the work they are permitted to do—the kind of work, the geographic areas in which they’re permitted to work, and the employers they may work for.

For example, Canada has special programs providing temporary visas for seasonal agricultural workers and for live-in caregivers. After two years, the latter (but not the former) are entitled to convert their status to that of a permanent resident without restrictions on work.

The unique restrictions that temporary workers face inevitably render them much more vulnerable to mistreatment by their employers than ordinary workers, whatever the formal similarity of their legal rights. The risk is probably greatest when their permission to enter and work is limited to one particular employer rather than to a type of work, but even in the latter case their status renders them vulnerable. It is possible to take some steps to reduce this vulnerability (e.g., permitting changes of employer, monitoring work conditions), but it cannot be eliminated altogether without eliminating this form of immigration.

Some people think that is exactly what we should do: we should abolish all temporary-worker programs because the restrictions they impose on the kind of work people can do and who they can work for are inherently exploitative and unjust by the standards we have set for the working conditions of citizens and permanent residents. I feel the force of this argument. On the other hand, the women who enter Canada under the live-in-caregiver program would not get in at all, or would have to wait much longer, if they went through the normal immigration process. And the seasonal workers, who are often able to provide comparatively good incomes for their families with such work, would find themselves without it. Some in both categories would doubtless enter without authorization and do the same work under worse conditions. So I find myself perplexed about what to do with such programs.

Unauthorized immigrants. Finally, what legal rights should be given to those who reside and work without official authorization? (For simplicity, I am going to assume that the state genuinely wants to prevent unauthorized immigration; if it did not, “unauthorized” immigrants would in fact be authorized, and therefore entitled to all the rights accompanying that status.) At first blush, it may appear anomalous to speak of the rights of unauthorized immigrants. Since they are violating the law by settling in the receiving state without its permission, why are they entitled to any legal rights?

But unauthorized immigrants do have legal rights that ought to be respected. Individuals are entitled to a wide range of legal rights simply by virtue of being within the jurisdiction of the state—for example, the right to be secure in their person and their possessions. If they are accused of a crime, they have the same rights as any other criminal defendant. If they are shot or struck by a car, they have the right to receive life-saving medical treatment. So far as I know, even the harshest critics of illegal immigration do not challenge these principles.

There are other legal rights that unauthorized immigrants possess that are more controversial but still generally acknowledged. First, they should and do have a legal right to the pay that they earn, even though they are working without authorization. This is simply an extension of the principle that they have a right to be secure in their possessions. They can be deported, but (in principle) they cannot be robbed.

Second, they are entitled to the same legal protections regarding wages and working conditions as other workers, because these protections are designed to establish minimum standards for economic activities. The fact that they lack authorization to work cannot relieve employers of the duty to meet these standards. Again, this principle is generally (though not universally) acknowledged in law.

The legal rights of unauthorized immigrants exist only in name because they fear any entanglement with the authorities.

Third, children who are present without authorization are entitled to a free public education. In the United States, this principle was established by the Supreme Court in 1982 in Plyler v. Doe. The same practice is followed in Canada and in most European states. This is a particularly controversial idea because it requires a direct expenditure of public funds on behalf of people who are present without authorization, and it creates an incentive for unauthorized immigrants to bring their families to join them. The alternative, however, is to permit the children who come without authorization to grow up without an education, which is a disaster for them and the society in which they live.

The biggest problem with the legal rights of the unauthorized is that they exist in name only because those present without authorization fear that any attempt to assert their rights will entangle them with the authorities and lead to their expulsion. A partial solution to this problem would be to build a firewall between immigration law enforcement on the one hand and the rest of the legal system on the other. In principle we ought to establish as a firm legal principle that no information gathered by non-immigration officials can be used for immigration-enforcement purposes. The objection to this is obvious. By increasing the security of the position of the unauthorized, one increases the incentives for others to come without authorization. Although this proposal is hardly realistic, it seems to me an essential step if we are to take seriously the rights that these people ought to enjoy.

As the list of rights grows, one might ask whether there are any rights that authorized immigrants have that unauthorized immigrants do not. The most obvious candidate is the right to stay. If one accepts the right of the state to control immigration, then one must accept as a corollary the right of the state to deport people who are staying without permission. But this right does not persist indefinitely. The general principle with which I began—the longer the stay, the stronger the claim—applies even in the case of those who have settled without authorization, and for the same reasons. When people settle in a country they form connections and attachments that over time make them members of the society. After a while, the conditions of admission become irrelevant.

This recognition of the moral importance of the length of stay, even if unauthorized, is reflected in the practices of many states, both in the granting of general amnesties to unauthorized residents and in the common practice of granting exemptions from the normal rules of deportation, both of which are almost always limited to those who have been in the country for an extended period. Not everyone accepts this principle, but in my view, long-term settlement carries significant moral weight that eventually forms the basis of a moral right to stay—one that ought to be recognized in law.