Immigration is an issue of concern to all the nations of the world, but it has a special meaning for the United States because we are a nation of immigrants. In the eighteenth century, most of the continent belonged to Indian tribes and Mexicans living in the western states, but of the 270 million persons that now constitute the United States, only a very small number can trace their roots to these groups. The country is now largely populated by persons whose ancestors immigrated to it. Some came in chains (the African slaves), but most came out of choice, to avoid famines and wars or simply in search of economic and social betterment.

To this very day, the doors of the United States have remained open. Last year almost one million immigrants were admitted for permanent residence. This number is vastly smaller than the number of those wanting to immigrate, and arguably fewer people than the country might support economically or socially, but I take a measure of pride in the fact that we admit as many as we do and that since the 1960s our immigration policy has been purged of an overt racial and ethnic bias. The national origins quota has been abolished, and today we have specific immigration programs designed to promote diversity among those admitted. These features of the law pay tribute to our history as a nation of immigrants. What is new, and at odds with this history, is that we have increasingly imposed disabilities on immigrants, mostly on illegal immigrants, but even on those who are lawfully present.

Political vs. Social Disabilities

One form of disability–let’s call it political–is deeply rooted in our culture and strikes me as acceptable. Here I am thinking principally of laws restricting to citizens the right to vote. At its core, becoming a naturalized citizen involves affirming one’s loyalty to the Constitution, and such an affirmation seems an unobjectionable requirement for participating in those activities such as voting that are responsible for sustaining or implementing the political principles that define the nation.

At one point in our history, some states allowed aliens to vote. It is also true that many immigrants who reside in the country a long time may actually feel a similar loyalty to the Constitution or political principles under which they happen to live. Yet as long as naturalization remains a viable and fairly economical option, as it indeed is in the United States–last year more than one million immigrants became citizens–it seems sensible for the law to require completion of the formal process of affirmation.

The more troublesome form of disability is social as opposed to political. In recent years the social disabilities have increased, and have taken three different forms: (a) bars on employment; (b) exclusions from public schools; and (c) denials of statutory entitlements, such as food stamps or medical services, routinely provided to the poor by the welfare state.

Many of these disabilities were imposed by the states and could be seen as a fiscal adjustment for federalism. In America, both the formulation and implementation of immigration policy are in the hands of the federal government. Social services are generally provided by the states, and a number of states, principally California and Texas, imposed social disabilities on illegal aliens as a way of financially protecting themselves from the failure of the federal government adequately to police the borders. The number of illegal immigrants is estimated to be about five million and almost three million reside in the states bordering Mexico.

Within the last few months a federal district judge in Los Angeles struck down a California law denying almost all state services, including medical care, to illegal immigrants–the notorious Proposition 187, adopted by popular referendum in 1994. The Los Angeles court found this law intrudes upon the exclusive jurisdiction over immigration reserved to the federal government by the Constitution. This ruling may not survive an appeal and in any event provided no protection against analogous federal enactments, some of which followed closely on the heels of the California law.

Federal measures already barred illegal immigrants from working, but in 1996 Congress enacted a far more severe measure, outstripping even Proposition 187. As part of its welfare reform agenda, Congress denied certain welfare benefits, including food stamps and financial support for the aged and disabled, to all immigrants, legal or illegal. Indeed, the law even cut off the benefits of those immigrants lawfully admitted to the United States before the date of enactment, August 22, 1996.

This unusual retroactive feature proved to be too much for Congress, and it soon had second thoughts. In August 1997 Congress enacted a measure to restore certain benefits to the infirm and aged who had been admitted to the United States before August 22, 1996, and in June 1998 it enacted a measure providing food stamps for select categories of immigrants–children, the elderly, and the disabled–lawfully admitted to the United States before that same date.

These amendments took some of the edge off the 1996 welfare act, but the basic regime introduced by that law–no welfare benefits for immigrants, even those lawfully admitted– remains intact and brought to the fore many issues of law and policy relating to immigration. It raised for me, with special urgency and clarity, the question whether enactments imposing social disabilities on immigrants can be squared with the Constitution, particularly the provision that guarantees to all persons–not all citizens, but all persons–equal protection of the laws.

The Constitutional Rule

In addressing this question, lawyers and commentators have turned to the now familiar rule that bars arbitrary discrimination. This rule has emerged over the last century as a gloss on equal protection and in fact been used as a powerful aid for the protection of minorities and women. The temptation is therefore great to use this rule to protect immigrants, yet such an effort seems misplaced.

The laws imposing social disabilities on immigrants do indeed seem at odds with the Constitution, but only because of the social stratification they tend to produce, not because they violate the antidiscrimination principle. The constitutional guarantee of equality bars not just discrimination, but also laws that create or perpetuate caste-like social structures, and for that reason calls into question the 1996 welfare act and similar measures.

Antidiscrimination is essentially individualistic. It guarantees individual fairness in the award of a scarce opportunity, say a job, and demands that individual applicants be selected on the basis of criteria that are functionally related to some legitimate purpose of the institution. The principle is given practical effect in the law through the construction of a short list of criteria–the so-called suspect criteria–that presumptively are not related to any such purpose. Race is the paradigm because we assume that an individual’s race or color is never, or almost never, functionally related to the pursuit of any legitimate purpose. This assumption about the irrelevance of race undergirds and gives continuing vitality to Justice Harlan’s now famous aphorism, “Our Constitution is color-blind.”

In recent years, many have come to question the general presumption about the irrelevance of race. We have discovered that a certain measure of color consciousness, as exemplified by affirmative action and school desegregation remedies, is necessary to eradicate the legacy of slavery and Jim Crow. As a result, we have grown uneasy with antidiscrimination as the sole strategy for dealing with race. In the case of immigrants, however, the inadequacy of antidiscrimination seems even more glaring because we cannot engage any general presumption about the irrelevance of alienage, which is commonly employed in a wide variety of legal processes, many of which seem of unquestionable validity. Indeed, drawing a distinction between aliens and citizens, or illegal and legal immigrants, is integral to the effort of any sovereign nation to determine its own membership. It is essential to the very existence of the nation as a community.

Although no blanket judgment can be entertained as to the irrelevance of alienage, there may be particular cases when it is in fact irrelevant. In these cases, a court might conclude that alienage serves no legitimate end, but rather is an expression of a hatred or dislike of immigrants, or in the words of contemporary legal doctrine, an expression of an “animus.” This highly contextualized version of the antidiscrimination principle might be similar to the one now applied to handicapped persons, but would be qualitatively different from that which over the years has protected blacks or even women. For them, antidiscrimination has had a near universalistic reach. Even more, in contrast to the rule used to protect blacks or women, the particularized version of antidiscrimination now proposed for immigrants would be devoid of any real bite. It would rarely invalidate a law.

There is no denying that, as a purely psychological matter, anti-immigrant sentiment is pervasive in America, but that is not the issue. A statute should never be judged on the basis of actual motivation, for then the validity of the law would depend on an inquiry into the minds of the legislators–never an easy or welcomed endeavor for a coordinate branch of government. Such an inquiry would also require some method for aggregating the good and bad motivations of the multitude of legislators, and might yield anomalous results. A statute would be struck down if it were the product of a political process in which animus entered, but the same statute would be allowed to stand if the legislators had a different, more benign, motivation. For these reasons, laws should be judged on the basis of social purpose objectively conceived, not actual motivation, and from that perspective it would be hard to challenge a law imposing disabilities on immigrants as an arbitrary discrimination. A legitimate social purpose can almost always be imagined.

A Court Decision

To demonstrate why antidiscrimination is inadequate in this context, and to lay the foundation for the shift to a more structural approach, let me take up a case decided by the Supreme Court in 1983, Plyler v. Doe. This case involved a Texas regulation denying children who were illegal immigrants free admission to elementary and secondary schools. All children born in the United States, even those of parents who are illegally present, are by virtue of the Fourteenth Amendment citizens of the United States. Yet if children who were born abroad immigrate to the United States illegally, either alone or, more plausibly, with their parents, they would themselves be regarded as illegal aliens. It was these who the Texas law sought to exclude from local schools.

No one would deny that as a purely historical matter, there was a lot of anti-immigrant animus in Texas when the state legislature enacted the law challenged in Plyler. It also seems fair to say that this hatred or fear played a role in the causal dynamic that produced the regulation. Yet the Texas law could not be invalidated under the antidiscrimination principle because it could be rationally justified as means to serve a legitimate social purpose, specifically to discourage the influx of illegal immigrants. Arguably, there may have been other ways of achieving this end; for example, increasing the barriers on the border or increasing the sanctions of the criminal law for illegal entry, though none of these measures were available to Texas as opposed to federal government. But the antidiscrimination principle itself does not require the state to select the best means available for pursuit of a legitimate end. Antidiscrimination promises individual fairness, not state efficiency, and is satisfied if the criterion for distinguishing among individuals is functionally related, even weakly related, to a legitimate end.

To regulate the choice of means by the state another theory was needed, and the Plyler Court found it in the rule prohibiting the state from creating, in Justice Brennan’s terms, a “subclass of illiterates.” On this reading, the Equal Protection Clause prohibits not only discrimination, but also the creation of a near-caste structure. It prohibits creating socially and economically disadvantaged groups that are forced to live at the margin of society, isolated from the mainstream, always at risk, seen in their own eyes and those of the dominant group as inferior.

Over the last fifteen years, Plyler has remained good law, but there has been some debate as to the scope of the ruling and its underlying principle. Some of this controversy has been fueled by certain passages of Brennan’s opinion for the Court, which are more keyed to antidiscrimination than antisubjugation. A question has also been raised as to whether the case is limited to the protection of children. Brennan spoke movingly of the children, and one of the Justices whose vote was necessary for the judgment–Justice Powell–saw the school-age children as the principal victims of the Texas law and made their alleged innocence the basis for intervention. Admittedly, these children were illegally present in the United States, but they were brought here by their parents and thus, Powell reasoned, not responsible for any wrongdoing.

This concern with “innocent victims” is a constant and pervasive theme in Justice Powell’s jurisprudence. It surfaced most dramatically in his effort to place limits on affirmative action programs in the name of antidiscrimination and no wonder–it partakes of the individualism that gives life to the antidiscrimination principle. In the family context, however, this individualism seems especially misplaced, since it is so hard to allocate responsibility within that social unit. True, parents may have made the decision to cross the border, but they may have done that to improve the life chances of their children. Thus, although the children did not choose to enter, they may have been the reason for the illegal entry and thus responsible for it. Choice is not the only basis for allocating responsibility, especially within the family.

I also find it difficult to understand why the constitutional validity of the Texas regulation should turn on the alleged innocence of the victims. Under the antidiscrimination principle the fault of the excluded group may be of relevance, on the (questionable) assumption that it is permissible for a government to allocate benefits only to the “morally worthy” or those not guilty of wrongdoing. Yet fault is of no relevance under the antisubjugation principle, for what that principle tests for is not the fairness of an allocative rule but the social structure that such a rule creates. No one, innocent or not, should be transformed by the state into a pariah in the way that exclusion from schools would.

To give substance to this intuition, imagine a public education system in Texas that ran an adult literacy program for all the community, but which barred illegal aliens. The targeted group chose to enter illegally, and thus are responsible for their status. They are not, in Powell’s terms, “innocent,” yet I would insist that the same result of Plyler should follow. Excluding illegal aliens from this program severely disadvantages them economically and socially and creates the very same danger that moved the Court in Plyler–creating a subclass of illiterates. The fact that they have violated the law in coming here is irrelevant to this conclusion.

Others who have sought to limit Plyler have emphasized that the case involved elementary and secondary education. They draw their inspiration from the concurrences of Justices Blackmun and Marshall, yet fail to articulate any principled basis for limiting the case in this way. In some state constitutions, elementary and secondary education is singled out for special protection. The federal constitution does not, however, have any special provision for education, and it has long been held–in a decision that Blackmun joined but from which Marshall dissented–that no such right is implied. The judgment in Plyler did not presuppose that there was a federal constitutional right to education. Like Brown v. Board of Education, the emphasis was on the functional as opposed to the formal significance of public education, and what it might mean for a group of people to be excluded from it.

A Structural Approach

Once we take this functional perspective, we can well understand why the antisubjugation principle should reach beyond education and guard against exclusion from all manner of state programs, food stamps, public housing, or medical treatment. Illiteracy is a severe disability in modern society, but perhaps no more so than being malnourished or homeless or sick and in need of medical attention. The state routinely responds to these needs of its citizens, and against this background, the exclusion of immigrants has a severely subordinating effect upon them.

So too do laws that bar illegal immigrants from working. These laws typically operate by imposing criminal sanctions on employers who hire illegal aliens. Once again, the border states were the prime movers in this domain, and in 1976 the Supreme Court upheld a California law penalizing those who hire illegal aliens. The Court reasoned that such a law is within the authority of the states because it relates to matters of only local concern. By 1986, the federal government moved in and imposed its own general ban. Today, laws prohibiting the employment of illegal immigrants are a familiar part of the legal landscape, and assumed to be of unquestionable validity. Yet they should be seen as having the same subordinating effect as the law condemned in Plyler, which, recall, was also aimed at illegal aliens. These laws force illegal immigrants to survive by begging or stealing and thus to live at the margins of society and to prey upon it–no education, no welfare, no work.

In calling into question laws excluding immigrants from welfare and educational programs and barring them from working, I am not surreptitiously questioning the validity of laws regulating admission. For present purposes, I am prepared to assume that these laws are just, and further that the availability of jobs, welfare benefits, and public education may provide incentives for illegal entry. My point is not to subvert the admission process or otherwise open the borders, but rather to insist that laws regarding admission cannot be enforced or implemented in ways that would transform immigrants into pariahs. Admission laws can be enforced by fences at the borders, deportation proceedings, or criminal sanctions, not, I maintain, by imposing social disabilities.

From a purely individualistic perspective this may seem an oddity. If put to the choice between deportation or a life in the United States without education or welfare or work the individual immigrant might well choose the latter. He or she may well prefer to live a marginal existence in the United States than to return to the country from which he or she came, and there may be good and sufficient reasons for that choice. A pariah in San Diego may live a lot better life than a citizen of Guatemala.

This fact–if it is a fact–may have a great deal of relevance for the theory of rational choice, but not for the proper construction of the Constitution, for the Constitution is not a set of rules for maximizing individual welfare on some global scale. Rather, it is a statement about how a society wishes to organize itself, and prohibits subjugation, even voluntary subjugation, because such a practice would disfigure society. In that respect, the antisubjugation principle of the Fourteenth Amendment is analogous to the Thirteenth Amendment’s ban on slavery, which precludes not only forced slavery, but also selling oneself into slavery or choosing to become a slave. Both amendments seek to prevent the emergence of social practices that are at odds with the egalitarianism to which the Constitution aspires. We ought not to subjugate immigrants, not because we owe them anything, but to preserve our society as a community of equals.

The Constitution does not condemn all distinctions between people, say between the rich and the poor or between the smart and the dumb. The subjugation produced by the laws imposing social disabilities on immigrants is, however, of another character, for what is entailed is not simply drawing a distinction among people, but rather creating a different kind of social structure altogether: it entails a further stratification or degradation of the very poor, or in Justice Brennan’s words, “raises the specter of a permanent caste.” Not all immigrants are poor; not all illegal immigrants are poor; but the laws in question take their toll on the poor and compound the disadvantages that come from poverty. These laws further isolate poor immigrants from the dominant groups in society and make them vulnerable in a way that is not true of laws that privilege intelligence or wealth. Those immigrants who ordinarily count on working, or using the public schools, or who sometimes need to fall back on welfare, stand in danger of being turned into pariahs.

In some cases, the process of degradation is facilitated by other marks, such as race. Although the Plyler Court did not dwell on this fact, it well knew that the illegal immigrants targeted by the Texas regulation were from Mexico and thus suffered from the degradation that all people of color suffer in the United States. But race seems to be only an aggravating factor, not essential to the harm. Even if the illegal immigrants were not from Mexico, or Asia, or Africa but were from Canada, or Ireland, or even a number of Northern European countries, the danger would exist that these people–not defined by race or even national origin, but by their legislative or cultural designation as immigrants–would be isolated from the dominant groups in society and treated as inferior. They would become the new underclass.

Political Vulnerability

A rule protecting immigrants against social disabilities may seem inconsistent with my willingness to accept political disabilities, for denying the right to vote to aliens may also isolate them and make them vulnerable to the dominant group. I acknowledge that, as a purely moral matter, the political and social disabilities might be indistinguishable, yet from the perspective of the Constitution there is a crucial difference (comparable to the one that necessitated two separate constitutional amendments to set the record straight on race–the Thirteenth abolished slavery, the Fifteenth conferred the right to vote on blacks.) The social disabilities are unconstitutional because they create a social structure that is inconsistent with the conception of community embodied in the Constitution. The political disabilities, in contrast, are not inconsistent with that conception of community; indeed, as I said at the outset, they are an expression of it. The political disabilities rest on the idea that the nation is a community, not just a geographic bounded territory, and like any such voluntary organization requires rules of membership and vests the power of governance in those lawfully admitted to membership.

For these reasons, I would draw a distinction in the constitutional realm between social and political disabilities. One can recognize the distinction between citizen and noncitizen in the political realm without turning the latter into a pariah. The legitimacy of political disabilities does not legitimate social disabilities. In fact, the very allowance of political disablement–confining the right to vote to citizens–is a good and sufficient reason, constitutionally or even as a matter of democratic theory, for the judiciary to scrutinize with a healthy measure of skepticism the work of the elected bodies insofar as it affects immigrants.

Over the last thirty or forty years certain groups, above all blacks, have received a special measure of protection from the judiciary. Not all laws disadvantaging them have been invalidated–far from it–but the courts have not let these laws stand without undertaking an exacting inquiry into whether they can be reconciled with the relevant constitutional principle (commonly antidiscrimination). To some extent, this special attentiveness can be seen as a measure of corrective justice, as an implicit atonement for the wrongs admittedly done to African Americans in the past. Immigrants cannot easily appeal to considerations of corrective justice. As an abstract category immigrants have long been subject to abuse, but as a concrete people they are mostly newcomers. On the other hand, they are very much entitled to claim the benefit of the more instrumental rationale for judicial solicitude–political powerlessness.

Immigrants who are not citizens are a numerical minority. Some can count on other nationals who are citizens to voice their concern, but that is exceptional. For the most part, they encounter great difficulties in forming coalitions that are the mainstay of politics, since all the others can gain by shunning them. In that sense, aliens, like blacks, are a discrete and insular minority, yet their powerlessness has another and more dramatic dimension–they are denied the vote. They are excluded, openly and formally, from the electoral process and for that reason make a very special claim to the protection of the judiciary.

Rather, therefore, than justifying social disabilities, the political disabilities immigrants suffer stand as a powerful reason why the one institution that stands above the political fray–the courts–should make doubly certain that the life of an immigrant, never an easy matter, is not more difficult than it has to be. The Constitution denies the majority the power to turn any group into a pariah, and the burden properly falls on the judiciary to make this rule, to borrow a phrase Brennan coined in still another context, a living truth.