More than two decades ago, Owen Fiss made a major contribution to constitutional theory with the unveiling of his antisubjugation approach to equal protection analysis. Fiss's focus on status harms to specially disadvantaged groups appeared to capture a deep truth about constitutional equality. The antisubjugation approach also had the virtue-ignored by a current majority of the Supreme Court-of offering an explanation for why race-conscious policies adopted on behalf of subordinated groups ought to be judged by different standards than race-conscious policies that harm subordinated groups.

Fiss's critics have argued that his theory has a difficult time identifying exactly which groups deserve special protection and which do not: the elderly? the disabled? gays and lesbians? the poor? Read too broadly, the antisubjugation principle threatens an equality revolution, one that conflates disadvantage with caste, political powerlessness with pariahhood.

The nation is worse off for having largely rejected the application of the antisubjugation principle to African-Americans. By applying his theory to legal and undocumented immigrants, however, Fiss plays into his critics' hands.

Fiss is surely correct that the federal statutes denying social safety-net benefits to lawful long-term immigrants are immoral and possibly unconstitutional. The Republican congressional majority has recognized how impolitic these policies were by repealing welfare cuts as applied to those immigrants in the country at the time of the law's enactment. (This reversal, by the way, arguably does significant damage to Fiss's assertion of the political powerlessness of immigrants.)

Fiss finds constitutional fault with the welfare policies because of their "subordinating effect" on immigrants; such laws, he says, run the risk of "transform[ing] immigrants into pariahs." Is he correct? The welfare statutes certainly produce hardship; but, as Fiss recognizes, hardship alone does not establish a constitutional violation: that requires the demonstration of a status-harm- evidence that a historically subordinated group is being kept in its place.

Immigrants as a class (even poor immigrants) are not obvious candidates for Fiss's special protection. Immigrants, unlike African-Americans, can remove themselves from the disadvantaged class; naturalization after five years of lawful residence is fairly straightforward. Furthermore, American immigration policies remain rather generous. As Fiss notes, the United States admits nearly one million legal immigrants a year, and in recent years about the same number of immigrants have naturalized; US immigration and naturalization laws have been shorn of their explicit racial bars; the children of immigrants born in the United States are automatically citizens. These are not the kinds of policies usually adopted toward a pariah class. The new welfare law is ugly, targeting those with little political power to object. But these objections apply much more readily to the poor as a class-citizens and immigrants alike-than to immigrants.

Additional problems arise with Fiss's description of the scope of his principle. He would not apply his analysis to regulations of immigration-admission and deportation policies-or to the denial of political rights to immigrants. For the former, it is hard to see how the imposition of social disabilities is more "pariah-creating" than laws that threaten removal. History provides terrible examples of expulsion as a key element in programs of subordination and exploitation.

Fiss would permit the withholding of political rights to immigrants because nations, as voluntary organizations, "require rules of membership" and members are vested with the power to write those rules. He asserts that the imposition of social disabilities, however, is destructive of "the conception of community embodied in the Constitution."

But this distinction seems to assume what it needs to demonstrate. Why are immigrants part of the constitutional community? Why are full members not entitled to provide for the welfare of other full members before looking to the needs of non-members-particularly if the acquisition of full membership is made relatively easy? The answer cannot be that the disfavoring of non-members makes them pariahs. Fiss must say why they are entitled to equal treatment in the first place.

The answer, I believe, lies in the antidiscrimination principle that Fiss casts aside. In his hands, that principle does little work, requiring only that the government provide some rational reason for its discrimination-a standard that, as Fiss states, is easily met. But as he notes later on, violations of the antidiscrimination principle may trigger heightened protection when the classification is one that is viewed as pernicious, likely to be the product of ill-will or thoughtlessness, or arbitrary.

It is upon this analysis that I believe the welfare laws denying benefits to immigrants are vulnerable. In the 1971 case barring most forms of state discrimination against aliens, Justice Blackmun initially argued that special protection was necessary because immigrants constitute a "discrete and insular minority"-a view that Fiss seems to share. But towards the end of the opinion Blackmun offered another critique of the state law denying welfare. Quoting a lower court opinion, he noted that

"aliens like citizens pay taxes and may be called into the armed services. . . . [T]hey may live within a state for many years, work in the state and contribute to the economic growth of the state." There can be no "special public interest" [permitting the state to limit benefits to citizens] in tax revenues to which aliens have contributed on an equal basis with the residents of the State.

Blackmun recognized that in the United States immigrants are part of our community and function as such. Their children born here are citizens. They are not pariahs, but rather "citizens-in-training." To deny immigrants welfare is to single out for punishment a class that is not appropriately distinguishable from others entitled to welfare.

This reasoning supports the existing constitutional doctrine regarding state discrimination against aliens. Discriminatory federal statutes raise an additional complication. Arguably, the denial of social benefits is no different than an immigration regulation denying admission to persons likely "to become a public charge" (as the immigration laws put it) or requiring that those admitted have sponsors who agree to take care of them in times of need. But a court needn't worry about whether or not these distinctions hold in order to strike down the welfare law. The current disabilities imposed on immigrants were not exercises of the immigration power in any meaningful sense. Those writing the welfare legislation sought ways to show that welfare "reform" would save money. Terminating benefits to immigrants constituted a significant share of those savings. A court would be justified in seeing the statute for what it is: not a regulation of immigration, but rather a regulation of immigrants. Such mistreatment, under the antidiscrimination principle, would require a better governmental justification than merely the desire to cut federal outlays.

So Fiss is correct to be worried about immigrants as easy targets of state and federal law-makers, particularly when they go looking for ways to save money. But the "pariah" claim seems overstated and unsustainable. It is the class structure of our post-industrial society, far more than our immigration rules, that may be responsible for creating and sustaining a caste of subordinated persons isolated from dominant groups in society.