In Chy Lung v. Freeman, for example, the Court considered a California statute that permitted state officials to examine new immigrants arriving at its ports, and to impose a significant bond if the state examiner concluded that the immigrant was within any one of numerous classes of undesirable persons. In 1875, the Supreme Court struck down the California statute, explaining: "The passage of laws which concern the admission of citizens and subjects of foreign nations to our shores belongs to Congress, and not to the States."

Similar language appears in dozens of judicial opinions in the ensuing century. With the expansion of the federal immigration apparatus in the late nineteenth and early twentieth centuries, backed by Supreme Court decisions such as Chy Lung v. Freeman, the role of individual states in developing or implementing immigration policy rapidly diminished.

The full history of federal immigration law, from its late nineteenth century origins to the modern twenty-first century conflicts, is beyond the scope of this essay. That history, however, has reflected periods of high legal immigration, including in the "Ellis Island" era at the start of the twentieth century; admission of large numbers of temporary workers, such as during the "bracero" programs of the 1940s and 1950s; and large-scale unauthorized migration, as in the 1990s.

It has also included periods of intense nativism; ideological persecution of labor activists and political radicals in the 1920s and during the McCarthy era; explicit racial exclusions, including Chinese exclusion, the long era of national origins quotas, and ineligibility to citizenship for many Asian immigrants; and discriminatory enforcement against disfavored immigrant groups, from Japanese internment to the mass deportation of bracero workers and the express targeting of Arab, Muslim, and South Asian immigrants after the September 11 terrorist attacks.

These developments in immigration law have all occurred at the federal level. Throughout the decades, while state or local officials have occasionally been conscripted to assist in federal implementation, there has been little direct role for states in immigration policy-making.

Today's immigration regime is codified primarily in the Immigration and Nationality Act (INA), adopted by Congress in 1952 and frequently amended since, especially in 1965, 1986, and 1996.

In the course of crafting the INA, Congress has reflected the post-Civil War understanding that immigration regulation is an exclusively federal function, and has occasionally legislated narrow and explicit derogations from the otherwise muscular preemption of state or local laws.

None of these exceptions, however, authorize the sort of laws and practices that have cropped up in places like Arizona.

For instance, Congress has expressly authorized direct enforcement of two criminal immigration provisions. One provision of the INA prohibits the smuggling, transporting, or harboring of illegal immigrants; another, entitled "Authority to Arrest," empowers INS agents "and all other officers whose duty it is to enforce criminal laws" to make arrests for those violations.

In other words, Congress expressly authorized all those empowered to enforce criminal laws—including state and local police—to make arrests for smuggling, transporting, and harboring offenses.

The legislative history of the provision confirms its plain meaning. When first drafted, the provision allowed officers "of the United States" to arrest alleged smugglers. Congress's elimination of the limiting phrase "of the United States" was intended to expand enforcement authority by allowing all criminal law enforcement officers—federal, state, or local—to make arrests for smuggling, transporting or harboring illegal immigrants.

It makes sense specifically to authorize local police to enforce those specific provisions only in a world in which police are otherwise prohibited from such enforcement. And as the Supreme Court has repeatedly reminded litigants, Congress does not intend any of its statutes to be superfluous, and thus each provision must be read to have new and definite meaning.

In 1996, Congress enacted amendments authorizing state and local police to arrest a second set of immigration violators, those who have committed the criminal offense of illegal reentry following a prior deportation.

The 1996 measure expressly provides that "state and local law enforcement officials are authorized to arrest and detain an individual who (1) is an alien illegally present in the United States, and (2) has previously been convicted of a felony" and ordered deported.

Representative John Doolittle offered a floor amendment that became this provision. He explained that "the Federal Government has tied the hands of our State and local law enforcement officials," because "current Federal law prohibits State and local law enforcement officials from arresting and detaining criminal aliens whom they encounter through their routine duties." His amendment, he argued, would "untie the hands of those we ask to protect us," at least with respect to previously deported felons who have illegally reentered the country.

Congress has also long understood that local police enforcement of civil immigration law is broadly preempted, unless expressly authorized by Congress. One section of the INA, first enacted in the 1950s, confers emergency powers on the Attorney General to authorize "any State or local law enforcement officer" to enforce federal immigration laws in the event the Attorney General certifies that there exists "an actual or imminent mass influx of aliens arriving off the coast of the United States, or near a land border."

And in 1996, Congress added non-emergency procedures for state and local jurisdictions to enforce federal immigration laws, in INA section 287(g). This provision requires training of state or local police, execution of a written agreement, and close oversight by federal immigration authorities.

As of 2010, over 70 jurisdictions have entered into "287(g) agreements" with federal authorities. More importantly, these twin procedures—the emergency "mass influx" procedures and the ordinary "287(g)" process—suggest that immigration laws may be enforced by state and local police only as legislated by Congress and subject to federal training, supervision, and oversight—and not at the unilateral initiative of local jurisdictions.

In other words, in the second half of the 20th century, Congress has approved state and local enforcement of specified immigration crimes, and broad civil enforcement if done pursuant to detailed statutory procedures. Each of these provisions was enacted on the understanding that only the federal government can make or enforce immigration laws, and each represent a narrow but explicit departure from that principle.

The Contemporary Period

While the twentieth century was largely a time of ascendant federal authority over immigration, and diminishing state and local participation, the past decade looks different.

The federal regime, creaky and out-of-date, has endured. The Department of Homeland Security has become today the largest federal law enforcement agency, with skyrocketing budgets and new records for arrests, detentions, and removals set virtually every year.

At the same time, however, there has been an undeniable explosion of immigration-related policymaking at the state and local level. Punitive measures such as those adopted in Hazelton, Pennsylvania and in Arizona have received much of the media attention.

Some local integrationist strategies, such as municipal confidentiality or non-cooperation orders regarding immigration enforcement, or New Haven's Elm City Resident Card (an optional municipal identification card), have provoked local controversy. Still other inclusionary approaches, many modest or incremental (such as a town hall offering some of its materials in languages other than English), have become common but do not always attract the same public scrutiny.

What explains the growth of state and local measures, and the wildly divergent approaches adopted by communities across the country? And given the longstanding history of federal supremacy in immigration policy and enforcement, are they likely to survive and flourish? Four primary factors explain the current turmoil.

First is a general trend toward devolution, visible across government programs affecting immigrants and predating the most recent events.

For instance, in 1996, Congress enacted a major overhaul of welfare programs. [Please read this Origins article for more on Welfare Reform in the 1990s] When Congress could not resolve a disagreement about immigrant eligibility for the major cooperative federal-state programs, such as Medicaid and Temporary Assistance to Needy Families (TANF), and subject to a few federal limitations, it largely punted to the states, authorizing each to determine for itself whether large classes of foreign nationals would be eligible for those programs.

Thus, in the late 1990s, when all state legislatures had to rewrite their own welfare laws in response to vast federal changes, they also were forced to legislate in historic detail regarding a long list of particular immigration statuses, mapping those onto state, local, and cooperative federal-state programs.

Similarly, state criminal justice systems have been increasingly forced to confront the consequences of immigration status at arraignments, plea hearings, sentencing, and during probation. This trend dramatically intensified in the aftermath of the Supreme Court's 2010 decision in Padilla v. Kentucky that a criminal defense lawyer's failure to advise her client of the immigration consequences of a plea may constitute ineffective assistance of counsel. This ruling is likely to result not only in numerous vacated convictions of non-citizens, but also to reforms in all fifty states regarding the role of judges in state criminal courts when accepting plea agreements (by which the overwhelming majority of criminal prosecutions are resolved).

The 2002 Supreme Court decision, Hoffman Plastic Compounds, Inc. v. NLRB, concerning the rights of undocumented workers under federal labor laws, has also forced many state labor and worker compensation agencies to engage more deeply than previously in the intersection of labor and employment law with immigration status.

Second is demography. While not reaching historic peaks as a percentage of the population, legal and unauthorized migration in the last twenty years has been high. More importantly, in the 1990s, new immigrants moved beyond the traditional "receiving" states such as California, Texas, Florida, and New York, to other states. And within all states, they moved beyond the largest cities to suburban and rural communities, many of which had not experienced significant new immigration in nearly a century.

The arrival of new, largely Spanish-speaking Latino immigrants caused local frictions in communities where, at least initially, mediating religious, business, and other civic institutions often did not yet exist.

At worst, members of some such communities reacted with racism and nativism; at a minimum, the reality of new immigrant populations in such towns compelled public institutions such as police departments, public schools, hospitals, and libraries to adapt their practices to address the reality of these new residents.

And it is these sort of local adaptations, from Hazelton's effort to penalize landlords who rent to immigrant tenants, to the Houston Police Department's non-cooperation policy and New Haven's municipal ID, that have often sparked local controversies, protests, and lawsuits.

Third is September 11. Following the terrorist attacks, the U.S. Department of Justice rescinded a prior Office of Legal Counsel memo concluding that local police lacked authority to engage in civil immigration enforcement. Despite enactment in 1996, not one local jurisdiction had executed a 287(g) agreement by 2001. After the attacks, the Justice Department encouraged jurisdictions to carry out this type of agreement.

It began entering tens of thousands of administrative immigration warrants into the FBI's National Crime Information Center database, through which local police conduct records checks on millions of motorists, arrestees, and others they encounter every day. And it began an aggressive, public campaign to persuade local law enforcement officials to make civil immigration enforcement part of their routine duties, even without a formal agreement.

The Bush Administration argued that civil immigration enforcement was a national security imperative, and sponsored programs that increased state and local participation in immigration enforcement after September 11—a trend with a few exceptions that the Obama Administration has largely continued and even intensified.

Fourth is the absence of meaningful federal reform to our immigration laws, which remain rooted in a post-WWII model that is out of step with our current economic and security needs, and inconsistent with our moral values.

Polls indicate that a significant majority of the American public support immigration reform built on the three pillars of 1) legalization for some but not all of the current undocumented population; 2) expanded but not unlimited opportunities for future, lawful immigration, both to reunify families and through temporary worker programs; and 3) more effective enforcement of the resulting immigration order.

These three principles were also the foundation of bipartisan discussions in the Senate in 2006-07. In the absence of federal legislative action, however, states and localities have had no choice but to undertake their own diverse efforts to adapt local rules to the demographic realities of their communities.

Where will this lead?

I expect that many of the most punitive local measures, like Arizona's S.B. 1070 and the Hazelton ordinance, will be struck down by the courts as preempted by federal law. By this time next year, the Supreme Court will have decided the "corporate death penalty" case from Arizona, and this opinion will no doubt shape the legal and policy landscape, especially in other states considering their own S.B. 1070s.

More inclusionary local measures have rarely been challenged in court, by contrast, and the few suits initiated have generally failed. Thus I expect that communities inclined in this direction will continue to pursue integrationist policies.

More broadly, the state and local debate about appropriate policies towards new residents that is as old as the Acadian refugee crisis of the 1750s will surely endure. Yet this very local friction and debate will likely continue to contribute to the pressure on Congress to modernize our antiquated immigration statutes, in ways that reflect the center of public opinion: some legalization, some enhanced "future flows," and meaningful enforcement of laws capable of being carried out.

Years from now, I expect it will be difficult to explain to our children and grandchildren why for decades this nation countenanced the de jure subjugation of millions of immigrants, benefiting from their labor, while forcing them to live a life in the shadows, members of a reviled caste, denied their full human flourishing.

And in that context, even in the centuries-long debate about the content of our immigration laws and the role of state and local governments in carrying them out, the conflicts of the Summer of '10 may well turn out to look a lot like those of the Summer of '57 for civil rights.