One of the biggest long-term demographic challenges faced by the U.S. is lack of an enforcement mechanism for the assimilation of new arrivals (i.e., legal immigrants who have been “naturalized” as U.S. citizens). There are some groups that refuse to assimilate, including Islamists and some racial groups (e.g., Somalis recruited into the al-Shabaab terrorist network).
Read more It’s Time to Recognize Iran’s Complicity in the 9/11 Attacks
Persistent failure of large immigrant groups to assimilate linguistically, civically, and economically can produce enduring ethnic enclaves, lower social trust, weaker shared civic norms, and political and cultural polarization that strains public institutions and social cohesion. Over the long term, these issues place the U.S. constitutional republic at dire risk.
The current naturalization process has no mechanism, at any point before or after the oath taken by new citizens, to assess or incentivize civic, cultural, or structural assimilation. One solution is to develop an “assimilation contract” for every new citizen that can be measured over time.
The purpose of this two-part article is to analyze the existing immigration and naturalization process using sociologist Milton Gordon’s seven-dimension framework in terms of how the system evolved, the contrasts among the waves of immigrants over time, the legal and administrative shortfalls within the system, and a potential framework that could help fix the problems.
THE SEVEN-DIMENSION FRAMEWORK
In 1964, Milton Gordon developed a framework that distinguished seven stages of assimilation:
- Cultural/behavioral assimilation: language, everyday norms, civic knowledge, usually the fastest elements to change
- Structural assimilation: entry into the same schools, workplaces, neighborhoods, and social networks as the native-born, which Gordon treated as the pivotal stage
- Marital assimilation: intermarriage rates across generations (upward trending is the key indicator)
- Identificational assimilation: primary national loyalty and self-identification (expressly, a shift away from the country of origin)
- Attitude-receptional assimilation: the near-absence of prejudice against the group among the native-born
- Behavior-receptional assimilation: the near-absence of actual discrimination against the group
- Civic assimilation: no persistent value or power conflict with the core constitutional order, plus the fulfillment of civic duties.
This is perhaps the best breakdown of what assimilation really means and how it is achieved. The seven stages represent laudable goals for the U.S. system that are rarely all achieved by naturalized citizens (certainly very few in the first two generations) but which should nevertheless be facilitated by an America First immigration and naturalization system.
THE PROBLEM: TWO WAVES, TWO SYSTEMS
The year 1924 was a key inflection point in the immigration waves to America. The Immigration Act of 1924 imposed national-origin quotas keyed to the 1890 census, which by design favored Northern and Western Europe and sharply curtailed Southern/Eastern European immigration and virtually all Asian immigration. This differentiates who was let in (pre-1924 and afterward).
The other inflection point differentiates how people were let in. Before 1980, people were admitted under open or quota-based entry (as reset periodically by Congress), while after 1980, people were admitted to the U.S. via refugee, asylum, and parole channels that did not previously exist.
These two axes are analytically separate; the ethnic-selection pathway and the humanitarian-based channel raise different policy issues and call for different fixes.
The Ellis Island Wave (~1880-1924)
With the preceding in mind, three structural facts distinguished the Ellis Island wave from today’s immigration:
There was no refugee or asylum infrastructure. The Refugee Act that created the modern refugee resettlement and asylum system didn’t exist until 1980. Almost everyone who arrived between 1880 and 1924 came as a voluntary economic immigrant, not as a person fleeing persecution and processed through a humanitarian channel.
There was no meaningful welfare state. Pre-New Deal America offered no unemployment insurance, no federal cash assistance, and minimal public relief. Economic survival required work, and work required enough functional English and local knowledge to get and keep a job.
For some, assimilation was slow and contested, not instant. For example, Little Italy, the Lower East Side, and Chinatown in New York City were real, multi-decade ethnic enclaves that also existed in other large American cities. Historians generally find that full civic and structural assimilation of the Ellis Island generation took two to three generations, not one.
The Modern Wave (post-1980)
The relevant legal categories among today’s immigrants are distinct from one another for assimilation policy analysis purposes:
Refugees. These people are admitted from abroad under an annual presidential ceiling set through consultation with Congress. Actual admissions by era: the ceiling stood at 110,000 at the end of the Obama administration (FY2017); it was cut sharply under the first Trump administration (down to roughly 15,000 by FY2021); the Biden administration raised the ceiling to 125,000 (FY2022–FY2025) and reached 100,034 actual admissions in FY2024 (the highest since FY1994). The second Trump administration set the FY2026 ceiling at 7,500, the lowest in the 45-year history of the program.
Asylum Seekers. These people apply from inside the U.S. or at the border and must individually prove a well-founded fear of persecution. FY2023 saw about 54,400 asylum grants, the most in at least a decade (and consistent with Biden administration open-borders policies in general).
Read more First-Ever Midterm GOP Convention Launches–Live Updates
Parole. This is a separate, discretionary executive authority that lets DHS admit people temporarily for urgent humanitarian or public-benefit reasons without going through either process. The Biden administration used it at a scale without recent precedent: the CHNV program (Cubans, Haitians, Nicaraguans, Venezuelans) paroled roughly 532,000 people. For other groups, as cited here: Uniting for Ukraine admitted about 240,000, CBP One scheduled roughly 936,500 appointments for parole consideration at the border, and about 76,000 Afghans were paroled after the 2021 withdrawal (again, all consistent with Biden administration open-borders policies in general).
DACA. Begun during the Obama administration, Deferred Action for Childhood Arrivals is a prosecutorial-discretion deferred action category complete with work authorization that confers no path to a green card or citizenship on its own. However, the Democrat Party has pushed for naturalization of DACA individuals from the moment the policy went into place. The Institute for Taxation and Economic Policy estimates that there were approximately 455,000 DACA recipients in the U.S. as of March 2026.
Note: parole is the one lever in this list that Congress never wrote to be used for the purposes of gaining U.S. citizenship that a single administration can turn on and off at a scale of hundreds of thousands of people with no ceiling, no congressional vote, and minimal individualized vetting compared to the refugee or asylum tracks. The Biden administration opened the parole aperture wide while the Trump administration closed it almost completely.
CURRENT LEGAL AND ADMINISTRATIVE SHORTFALLS IN THE SYSTEM
A number of problems exist in the U.S. immigration and naturalization system that relate to assimilation.
The oath is a one-time speech act with no follow-up. The Oath of Allegiance, which is administered at the end of the process (either administratively by USCIS or judicially before a federal judge), requires the applicant to renounce foreign allegiances, support and defend the Constitution, and bear arms or perform noncombatant service if required by law.
Once taken, nothing in the system checks whether the naturalized citizen’s subsequent conduct is consistent with it. There is no reaffirmation, no periodic review, and, most critically, very limited legal capacity to act even if a citizen’s conduct plainly contradicts the oath because of the constitutional constraints implemented by Congress.
The civics/English test measures memorization, not integration or understanding. The test was expanded in 2025 (from 10 to 100 possible questions, 6 correct to pass, to 20 of 128 possible questions, 12 correct to pass). These and previous changes did not change the fact that the test has remained a one-time, multiple-recall knowledge check administered in a single interview. It tests whether an applicant can recite facts about American government; it does not test functional civic participation, ongoing English use, or whether the applicant’s stated commitment to U.S. constitutional supremacy is more than a memorized answer.
There is a genuine tension in this test design, as making it harder raises a real barrier for elderly or low-literacy applicants (who already have partial exemptions), while leaving it as a one-shot recall test does very little to predict or produce real assimilation.
“Good moral character” is a backward-looking, individualized screen, not an assimilation screen. The good moral character requirement looks at the applicant’s conduct during the statutory period before naturalization (criminal record, tax compliance, etc.). It was never designed to, and doesn’t assess, whether the applicant intends to integrate into the language, civic life, or legal norms of the receiving society going forward. It is a character-and-legal-compliance test, not an assimilation test.
Denaturalization (the only post-oath enforcement tool) is both constitutionally narrow and historically almost unused. Denaturalization requires proving, in a civil action under a “clear, convincing, and unequivocal evidence” standard (or in a criminal case, beyond a reasonable doubt), that citizenship was obtained illegally or by concealment/misrepresentation of a material fact at the time of naturalization. It cannot be used against a citizen for conduct or beliefs that only becomes apparent after a lawful naturalization; it is a fraud remedy, not a “citizenship on probation” remedy.
Actual usage confirms how narrow this tool is in practice. After the Supreme Court’s Afroyim decision (which will be discussed in Part Two of this article), fewer than 150 people total were denaturalized between 1967 and 2017. This was a roughly 99% decline from the pre-1967 average of about 367 cases per year.
Recent administrations have varied widely in enforcement intensity (Obama-era: ~15.8 case filings/year; Biden-era: ~4/year; the Trump administration has averaged ~6/month up to this point). But even at the high end, this is a fraud-detection docket processing dozens of cases a year against roughly 25 million naturalized citizens, not a mechanism that could plausibly enforce an assimilation standard at any scale, nor is it designed to.
No structural incentive and no data infrastructure. There is no federal tracking of post-naturalization English retention, civic participation, residential integration, or intermarriage. The practical prerequisites and structural markers in Gordon’s framework are simply not measured after the oath is administered. And no denaturalization enforcement is possible without real data, as the government measures almost nothing about immigrants after the day they become citizens.
CONCLUDING THOUGHTS
This ends Part One of this two-part article. Part Two will cover a proposed assimilation framework that attacks some of the problems identified in this part, some roadblocks to its implementation, and a wrap-up summary.
Read more China Is Actively Preparing a Humanoid Robot Army for Future Battlegrounds
