Hart-Celler at the Statue of Liberty and the Country It Remade

Lyndon Johnson liked a setting. On October 3, 1965, he went to Liberty Island in New York harbour, where a signing ceremony had been set up at the foot of the statue. Behind him the copper woman held up her torch; around him were members of Congress who had steered the bill, among them an elderly congressman from Brooklyn named Emanuel Celler, who had been fighting the national origins quotas since he was a freshman in 1924. Johnson signed the Immigration and Nationality Act of 1965 there, and then, in his remarks, he went out of his way to play it down. This was "not a revolutionary bill," he said, and it would not reshape the daily lives of Americans or affect the lives of millions. It was a correction of an old injustice, nothing more. Within a generation that judgment would look like one of the great misreadings in the history of American lawmaking. The law did not just remove a discriminatory system. It set in motion the largest change in the country's makeup in a century, mostly through a provision its authors thought would do the opposite.

In short: The 1965 Hart-Celler Act abolished the national origins quotas and replaced them with a system built mainly on family ties and skills, with per-country limits and, for the first time, a cap on the Western Hemisphere. Its sponsors expected modest change; instead it opened the door to large immigration from Asia, Latin America and Africa. Both its success and its gaps still define the immigration debate.

The quotas that had to go

To see why 1965 mattered, picture the system it replaced. Since the 1920s, immigration from outside the Americas had been governed by national origins quotas that gave most of the available visas to people from Britain, Ireland and Germany and only a few thousand a year to Italy, Greece or Poland. Asian countries received token quotas after the war, around a hundred visas a year for many of them. The design was openly a ranking of peoples by desirability, and Congress had renewed it as recently as 1952 over President Truman's veto.

By the early 1960s, that system had become an embarrassment. The country was fighting a Cold War in which it claimed to stand for freedom and equality, and the Soviet Union and others pointed to American racism at every opportunity. The civil rights movement was forcing the country to face what its laws said about who counted. A quota system that ranked Italians below Swedes was hard to defend in the same years Congress was passing the Civil Rights Act and the Voting Rights Act.

President John Kennedy sent Congress a proposal to abolish the quotas in July 1963. He had written a short book on immigration, and the cause mattered to him; his own family had come from Ireland in the nineteenth century. After his assassination that November, Johnson took up the bill, as he did much of Kennedy's agenda, and pushed it through in the 1965 session.

Two sponsors and a key subcommittee chairman

The act is named for Celler, the chairman of the House Judiciary Committee, and Senator Philip Hart of Michigan, a Democrat known in the Senate as a quiet, principled liberal. In the Senate, Edward Kennedy, then in his first full term, managed the bill on the floor.

But the most consequential person in shaping the final law may have been a less famous one. Representative Michael Feighan of Ohio, a Democrat who chaired the House immigration subcommittee, was sceptical of the reform. He insisted that the new system prioritise family reunification over the skills-based approach the administration had preferred. His reasoning, as historians have reconstructed it, was that if visas went mostly to relatives of people already here, the immigrants who came would resemble the existing population, which was still mostly European in origin. A family-based system, in other words, was meant to be a soft brake.

It became the opposite.

What the law actually did

The 1965 act, which took full effect in mid-1968, made several changes.

It abolished the national origins quotas. In their place it created an annual cap of 170,000 visas for the Eastern Hemisphere, with no single country allowed more than 20,000.

It set up a preference system, with most visas going to relatives of US citizens and permanent residents, a smaller share going to people with needed skills or professional training, and a share for refugees. Immediate relatives of US citizens, meaning spouses, minor children and parents, were not counted against the caps at all.

It placed, for the first time, a numerical cap on immigration from the Western Hemisphere, set at 120,000 a year. That provision was a compromise demanded by members who worried that removing the European quotas would be followed by unlimited immigration from Latin America. Later amendments applied the per-country limits to the Western Hemisphere too.

On the same day he signed the law, Johnson also announced that Cubans seeking refuge from Fidel Castro's government would be welcome in the United States. That announcement led to an airlift, the Freedom Flights, that brought large numbers of Cubans to Florida over the following years. It was a reminder that refugee policy, handled mostly outside the new law, would be a separate and recurring story.

Why everyone got it wrong

The sponsors' predictions were wrong for reasons that seem obvious in hindsight.

First, family reunification depends on who is already here and who wants to come. In 1965, many Europeans no longer wanted to emigrate; Western Europe was booming. The people who did want to come, in large numbers, were in Asia and Latin America. Some arrived first through the occupational preferences, as doctors, nurses, engineers and students, especially from India, the Philippines, Korea and Taiwan. Once they became permanent residents and then citizens, they could sponsor spouses, children, parents and siblings, who could sponsor their own families in turn. Critics later called this chain migration. Its supporters called it what families do.

Second, the Western Hemisphere cap collided with history. The Bracero program, which had brought Mexican farm workers to the United States on temporary contracts for more than two decades, ended at the close of 1964. The demand for that labour did not end. With the legal pathway closed and the new cap in place, much of the movement that had once been legal continued without papers. Unauthorised immigration from Mexico grew in the following decades, a consequence that had less to do with the 1965 law's intentions than with the way it met an economy built on seasonal migrant work.

Third, the sponsors underestimated the pull of a growing economy and the reach of global travel. Jet travel made long migration cheaper, and the American economy kept creating jobs that immigrants filled.

The numbers tell the story. The share of the American population born abroad fell to a low point of about one in twenty around 1970. In the decades since, it has risen to roughly one in seven, near its historic high from the early twentieth century. And the countries of origin changed completely: where Europe once dominated, the largest sources of immigrants became Mexico, other Latin American countries, China, India and the Philippines, with growing numbers from Africa.

There was a fourth factor, and it is the most human one. The people who arrived after 1965 did what immigrants have always done: they built networks. A nurse from Manila who found work in a hospital in New Jersey told her cousins about the job. An engineer in California helped his brother's son apply to a university nearby. Restaurants, churches and community associations made it easier for the next family to settle. None of that appears in a statute, but it is the engine that turned a modest legal change into a demographic one. Lawmakers who wrote the preference system as if families were static had not reckoned with how families actually work.

The laws that followed it

The 1965 act was not the last word, and two later laws filled in what it left out.

The first was the Refugee Act of 1980, signed by President Jimmy Carter. Until then, refugee admissions had been handled through a series of improvised measures, including parole authority used for Hungarians after 1956, Cubans after 1959 and Vietnamese, Cambodians and Laotians after 1975. The 1980 law adopted the international definition of a refugee, a person with a well-founded fear of persecution based on race, religion, nationality, political opinion or membership in a particular social group, and created a regular process for setting admission numbers each year in consultation with Congress. It also gave asylum a statutory basis for people who reach the United States and ask for protection. Much of today's fight over asylum is a fight over how that 1980 framework should work under pressures its authors never imagined.

The second was the Immigration Act of 1990, signed by President George H.W. Bush. It raised the overall level of legal immigration, expanded employment-based visas, and created the diversity visa lottery, which sets aside a number of visas each year for people from countries with low recent immigration to the United States. The lottery grew in part out of an effort to help Irish immigrants, many of whom were in the country without papers in the 1980s, and it has since brought large numbers of immigrants from Africa and Eastern Europe.

Each of these laws was bipartisan in its time. Each was a response to problems the previous law had created or ignored. That is how immigration policy worked for most of the twentieth century: Congress revisited it roughly once a generation. The striking fact of the twenty-first century is that it has not, and that most change now comes from executive action and court orders instead.

The strongest criticism, and an answer

Some critics of the 1965 law argue that it was sold dishonestly, that the public was promised modest change and got a transformation without ever voting for it. They point to the reassurances given by the bill's sponsors, including Edward Kennedy's statement on the Senate floor that the ethnic mix of the country would not be upset. Others argue that the family-based system admits too few people based on skills, compared with countries such as Canada and Australia that rely on points systems.

Those are fair points to raise. The sponsors did underestimate their law, and a reasonable person can argue that the balance between family and skills should be different. Bipartisan reform bills in 2006, 2007 and 2013 proposed shifting that balance; none became law.

But the claim of a betrayal misreads what the law did. It did not set out to change the country's makeup; it removed a rule that had frozen the makeup by law. Any system that treats people of every nation equally will produce results that look different from a system designed to favour some nations. The change that followed is what equality, applied to immigration, looked like in practice. Here is the argument: that is not a flaw in the law but the point of it.

A country that looks like its law

Walk through almost any American city or suburb, and much of what you see traces back to 1965. Hospitals staffed by doctors and nurses from India and the Philippines. Korean churches and Vietnamese restaurants and Indian grocery stores in strip malls. Engineers from Taiwan and China who helped build Silicon Valley. Families from Mexico and Central America who work, pay taxes, raise children who are citizens, and live under a system that has not been comprehensively updated in decades.

That last point is where progressives should be honest about the unfinished work. The 1965 law ended a racist system and replaced it with one that was fairer on paper but has been allowed to calcify. Per-country caps mean that some immigrants from large countries such as India, Mexico and the Philippines wait for many years, even decades, for visas already approved. Unauthorised immigrants with deep roots have no path to legal status. The asylum system is overwhelmed. Every attempt at comprehensive reform since 1986 has failed in Congress. The result is a system that many people on all sides of the debate describe as broken, though they disagree sharply about what fixing it would mean: more enforcement, more legal pathways, a different balance between family and skills, or some combination worked out by people who do not trust each other.

A new Congress sworn in on January 3, 2027, will face the same choices that have defeated its predecessors. Immigration bills that passed with support from both parties are rare but not unknown; you can track them on our common ground page, and follow the year's immigration fights on the immigration topic page.

What you can do

• Ask your members of Congress where they stand on visa backlogs and legal status. The representatives lookup gives you your House member and senators, their phone numbers and their latest votes, and family-based backlogs are a concrete issue to raise.
• Support legal help for immigrants. Immigration cases are complex, and people without lawyers fare far worse. The organizations directory lists immigrant-serving and legal aid nonprofits in every state.
• Read and respond to proposed immigration rules. Agencies propose changes to visa processing, fees and eligibility through formal rulemaking; the public comment page lists open rules, closing soonest first, with tips on writing a comment that will be read.
• Learn your own neighbourhood's immigration story. Libraries, churches, temples and mosques often keep community histories, and the people who lived them are usually glad to tell them.