“Diversity is our strength” has acquired the authority of an axiom without enduring the inconvenience of becoming an argument. Diversity is a condition; whether it becomes a strength depends upon the civilization into which difference is introduced and the institutions capable of turning strangers into a people.
Few political slogans have achieved so much cultural authority while being subjected to so little serious examination. Diversity of what, strength in what sense, and by what mechanism does increasing ethnic, linguistic, religious, or cultural heterogeneity produce a stronger society? Those questions become unavoidable when diversity increases faster than the institutions and cultural bonds capable of integrating it, because a serious proposition must also identify the evidence that would show when difference has become a liability rather than an asset.
The question becomes still more important when simple diversity is expanded into multiculturalism, for the two are not interchangeable. A diverse society contains people of different backgrounds, while a pluralistic society permits private religious, ethnic, familial, and cultural associations within a common political order; multiculturalism, in its stronger political form, goes further by treating the public recognition, accommodation, and preservation of distinct cultural identities as an affirmative social objective.
That distinction matters because a substantial current of progressive and left-liberal thought does more than defend equal citizenship before the law, a proposition that is hardly controversial. It advances a social order in which subgroup distinction receives political and institutional recognition while assimilation into a dominant national culture is regarded with suspicion; at that point the subject is no longer tolerance but a theory of social organization, and theories of social organization can be tested against the evidence of human behavior.
The most useful starting point is not Robert Putnam’s famous study, although his work belongs in the discussion, but the broader review by Peter Thisted Dinesen, Merlin Schaeffer, and Kim Mannemar Sønderskov. Their 2020 analysis in the Annual Review of Political Science examined 1,001 statistical estimates from 87 separate studies concerning ethnic diversity and social trust, pooling an entire research literature rather than asking readers to place extraordinary confidence in the peculiarities of one city, country, dataset, or researcher.
Its central finding was straightforward: across the studies, greater ethnic diversity was associated with lower social trust. The negative relationship became stronger when researchers measured trust in neighbors and when diversity was measured at increasingly local geographic levels, while conventional socioeconomic controls generally altered the association only modestly.
None of this proves that every diverse community will be low-trust or that ethnicity is the only relevant variable. Poverty, residential instability, segregation, migration patterns, crime, institutional quality, economic conditions, cultural distance, and patterns of interpersonal contact all matter; the evidence nevertheless demolishes the simplistic claim that demographic diversity is inherently or automatically social strength.
If heterogeneity reliably generated strength, it should predict greater interpersonal trust or, at minimum, no systematic trust penalty. The average association instead runs in the opposite direction, particularly where people actually live as neighbors, which changes the serious question from whether people of different backgrounds can coexist—they plainly can—to what social machinery makes them willing to trust, cooperate, enforce common norms, sacrifice for one another, and regard themselves as members of the same people.
Trust can sound like a soft psychological concept until its practical work becomes visible. A high-trust society permits enormous amounts of ordinary life to proceed without formal enforcement: a misdelivered package is likely to be returned, a neighbor who sees smoke calls the fire department, an adult intervenes when children vandalize property, contracts mean approximately what they say, and residents can organize around a local problem without assuming that everyone else will free-ride.
Each expectation reduces the verification, surveillance, litigation, bureaucracy, and defensive behavior required for daily cooperation. Trust therefore functions as a form of infrastructure, although it is built from memory, reputation, shared standards, and repeated encounters rather than concrete, steel, or fiber-optic cable.
Sociologists Robert Sampson, Stephen Raudenbush, and Felton Earls captured one crucial dimension through the concept of collective efficacy: social cohesion among neighbors combined with their willingness to intervene for the common good. Their landmark study of 8,782 residents across 343 Chicago neighborhoods found collective efficacy significantly associated with lower violence even after accounting for numerous individual and neighborhood characteristics.
The questions used to measure it were wonderfully mundane. Would neighbors intervene when children skipped school, sprayed graffiti, disrespected an adult, or began fighting; would people help one another, trust one another, and recognize shared values? The answers reveal a layer of civilization situated between the private household and the formal state, where parents, neighbors, churches, employers, shopkeepers, coaches, voluntary associations, extended families, and reputations regulate conduct before public force becomes necessary.
Sampson’s later work found residential stability especially important for reciprocal exchange and intergenerational social relationships because people who remain near one another acquire knowledge about how others behave. When trust collapses, that middle layer thins: neighbors stop intervening, public spaces become somebody else’s problem, parents know fewer adults around their children, and locks, cameras, background checks, police, courts, and agencies assume work that informal relationships once performed.
Robert Putnam’s 2007 E Pluribus Unum study is often reduced to the claim that ethnic diversity makes people distrust other groups, yet his actual finding was more unsettling. Residents of highly diverse communities tended to “hunker down”: trust declined not only across racial and ethnic lines but even toward members of one’s own group, while altruism, community cooperation, friendship networks, and several measures of civic engagement also weakened.
The problem therefore cannot be reduced to conventional racial hostility, because people can become less socially connected without becoming openly hostile. A neighborhood can remain polite while rarely cooperating, and a city can contain many cultures without developing much common life; peaceful occupation of the same physical territory establishes coexistence, but it does not establish cohesion.
Hotels, international airports, and global business conferences are diverse environments in which people coexist easily because their interactions are temporary, structured, and bounded by explicit rules. A nation asks considerably more: citizens who may never meet must pay taxes into the same treasury, obey common laws, educate children through common institutions, accept elections they dislike, cooperate during disasters, bear burdens for strangers, and sometimes die beside one another in war; those obligations require social solidarity, not tolerance alone.
Defenders of multiculturalism can reasonably answer that ethnic diversity and multicultural policy are different phenomena, and the evidence supports that distinction. Akira Igarashi’s analysis of 20 European countries found stronger multicultural policies associated with a smaller national-identification gap between natives and non-European immigrants; other studies have found positive relationships between multicultural orientations and generalized trust, while reviews of the policy literature describe the evidence of direct harm to social cohesion as mixed rather than decisive.
Those findings defeat the crude assertion that recognizing a minority practice automatically destroys national cohesion, but that is not Civerum’s argument. The deeper question concerns what multiculturalism is trying to optimize: if cultural recognition helps immigrants become more attached to America, more trusting of other Americans, and more invested in American institutions, recognition is serving integration, and integration remains the destination.
A country may tell newcomers that ancestry need not be erased before they become American; such confidence can strengthen assimilation by enlarging the national “we.” A country that organizes public life around the continued political salience of distinct cultural communities pursues a different end, because subgroup preservation has become the organizing principle rather than a bridge into common citizenship.
Human cooperation depends heavily upon predictability, which means trust grows when people can make reasonably sound judgments about what others regard as proper conduct. Promises, property, parental discipline, public spaces, corruption, violence, civic duty, truthfulness, and the reach of family obligation cannot be treated as peripheral concerns; they form the behavioral grammar through which strangers decide whether cooperation is prudent.
Cultures answer those questions differently, and the observation requires no racial theory. Two people with identical skin color may possess radically different assumptions about authority, family, religion, contracts, corruption, education, reciprocity, or the relationship between the citizen and the state, which means ethnicity often acts in diversity research as an imperfect proxy for something deeper: social distance and uncertainty about behavioral expectations.
A high-trust society reduces that uncertainty without requiring citizens to agree about everything; it supplies enough common moral and civic grammar for strangers to predict one another. Multiculturalism creates a paradox when a political culture insists that groups need not converge while simultaneously weakening the institutions responsible for generating common expectations, thereby dismantling the bridge while increasing the distance it must span.
A high-trust nation needs a common public language, which does not require suppressing the private languages of families, congregations, and cultural associations. Citizens may speak Spanish with grandparents, Polish at church, Mandarin with relatives, or Yoruba at home while mastering the common language that gives hundreds of millions of strangers access to one another.
The purpose is communicative interoperability, for language carries far more than literal information. Humor, reassurance, correction, negotiation, reputation, moral nuance, friendship, and disagreement depend upon it, so an America capable of incorporating immigrants from everywhere has every reason to promote mastery of English enthusiastically; multiculturalism becomes counterproductive when linguistic separation receives the same public value as the shared language that makes integration possible.
Language permits citizens to communicate expectations, while shared norms tell them what those expectations are. No legal code can specify every act required for social order, because much of civilization exists through internalized obligations: protect the helpless, return what was borrowed, keep promises, respect property, discipline children, preserve common spaces, help a person in obvious danger, resist profitable dishonesty, and tell the truth when others reasonably depend upon it.
A high-trust society does not require universal virtue, which fallen human beings cannot supply; it requires confidence that violations of these expectations are deviations from the norm rather than rival norms with equal standing. Civerum’s anthropology matters here because human beings are self-interested, tribal, capable of deception, and susceptible to corruption, making norms a form of civilizational technology that reduces the situations in which coercive authority must substitute for moral restraint.
Religion belongs in this analysis because civilizations do not obtain their moral premises from nowhere. America never required every citizen to join the same church or profess Christianity, yet its vocabulary of human dignity, moral responsibility, marriage, family, charity, obligation, and the limits of earthly power developed within a civilization deeply shaped by Christianity; for Civerum, that inheritance cannot be dismissed as folklore.
The Reformed understanding of humanity begins with an extraordinarily relevant proposition: men are not angels. Human beings require moral formation, restraint, accountability, institutional limits, and incentives compatible with fallen nature, while government cannot redeem humanity because government is operated by the same morally fallible creatures it claims to improve.
Religion’s empirical relationship with generalized trust is complicated rather than uniformly positive. A major British longitudinal analysis found religious-service attendance positively associated on average with generalized trust, volunteering, and perceived cooperativeness, though effects varied considerably across traditions; the meaningful distinction is therefore between institutions that widen moral obligation and those that produce inward-facing solidarity alone. Healthy religion begins with duties to those immediately entrusted to a person while extending moral concern beyond blood, tribe, race, and class—the architecture generalized trust requires.
Strong families are indispensable because they are the first institutions in which human beings learn authority, reciprocal obligation, self-restraint, sacrifice, responsibility, discipline, and care for someone other than themselves. Family loyalty alone, however, cannot produce a high-trust nation, and a society organized entirely around clans may contain extraordinarily loyal households alongside very little trust among strangers.
Alberto Alesina and Paola Giuliano found that especially strong reliance upon family ties can become a substitute for generalized trust, associated with lower civic engagement and political participation. Civerum’s ideal is therefore neither weakened families nor familism, but strong families that teach people to extend responsibility outward: the family should be the first school of trust, not its final boundary.
Trust takes time because reputation takes time. Repeated encounters reveal who keeps a promise, which adults can be trusted, which customers are known to the shopkeeper, who helps after a storm, and how today’s conduct will affect tomorrow’s relationships; residential instability continually erases that accumulated knowledge.
Research by Sampson and his colleagues found residential stability especially predictive of intergenerational closure and reciprocal neighborhood exchange. The mechanism is not mysterious: continuity permits information, obligation, memory, and reputation to accumulate until a place becomes more than a temporary collection of addresses.
Nations possess an analogous inheritance on a larger scale. Institutions develop reputations, customs become intelligible, moral practices are transmitted, and national stories give later generations a common reference point for what citizens owe the country and what government owes citizens; history is therefore more than what happened before the living arrived, because it creates the expectations within which they operate after arriving.
Trust also depends upon the belief that other people operate under roughly the same rules. That expectation is both cultural and institutional: citizens must believe their neighbors recognize common obligations while formal authority punishes serious defectors without favoritism.
Experimental research by Andrea Martinangeli, Marina Povitkina, Sverker Jagers, and Bo Rothstein provides strong evidence that institutional quality can causally affect generalized trust. Participants exposed to institutions less capable of preventing corrupt behavior subsequently displayed lower trust even when corruption had not necessarily occurred, meaning that the visible possibility of corruption escaping restraint was enough to damage confidence.
People watch how systems treat others and infer the character of the society they inhabit. When rules are selectively enforced, politicians appear above the law, public institutions favor chosen constituencies, criminal conduct receives inconsistent responses, or access seems to depend upon identity and affiliation, distrust becomes rational; a high-trust society thus requires the widespread expectation of common rules impartially administered.
Civil society turns strangers into acquaintances through churches, neighborhood associations, volunteer organizations, sports leagues, schools, charities, veterans’ groups, business and professional associations, parent organizations, and fraternal orders. These institutions create what a government diversity program cannot manufacture by decree: shared experience organized around a common purpose.
Proximity alone is not integration, because people may live beside members of another group while rarely forming meaningful relationships. Repairing the same playground, worshiping together, coaching the same children, belonging to the same volunteer fire department, organizing the same festival, or serving in the same military unit creates structured cooperation through which the category “those people” can become “the person beside me.”
Social psychologists Samuel Gaertner and John Dovidio developed what became known as the Common Ingroup Identity Model, whose central insight is intuitive. Intergroup bias can diminish when people who first understand themselves as members of separate groups come to perceive themselves as members of a larger common group; cooperative interdependence, meaningful contact, common goals, and an inclusive identity can move social categorization from “us and them” toward a broader “we.”
Subgroup identities need not disappear for this to happen. A person can be Sicilian-American, Catholic, Texan, a Marine veteran, a father, and an American at the same time; the danger begins when the subgroup becomes politically superior to the identity capable of ordering all the others.
Research from highly heterogeneous Swedish neighborhoods supplies unusually direct evidence. Across four waves of panel data, stronger local and national identities were associated with increases in interpersonal and institutional trust, with local identity especially related to local trust and national identity to national-level trust; research among immigrants and their descendants in Denmark likewise found national belonging, national pride, and preference for Danish citizenship positively associated with social and institutional trust.
The implication is not that immigrants must forget where their grandparents came from, but that the larger national identity must become consequential enough to outrank ancestral differences in political life. That is assimilation properly understood: neither cultural amnesia, racial conformity, a ban on grandmother’s recipes, nor compulsory religious sameness, but the creation of a people.
The American motto contains a political insight that modern multiculturalism risks forgetting: E pluribus unum, out of many, one. Both halves matter, because America has never been ethnically uniform; English, Scots, Irish, Germans, Italians, Jews, Poles, Scandinavians, Chinese, Africans, Hispanics, Slavs, Arabs, Caribbean peoples, and countless others contributed to the country.
The historical achievement was not that all those differences could be catalogued, but that people with radically different origins could eventually understand themselves as Americans. English became the common public language, the Constitution a shared political inheritance, and American history and symbols common points of reference; schools, workplaces, neighborhoods, churches, military units, associations, and marriages gradually widened the immigrant’s social universe.
Ethnicity did not vanish, but its political importance declined as American identity gained salience. Modern multiculturalism can invert that achievement by asking institutions to preserve, recognize, represent, accommodate, and validate the many rather than asking how people of many backgrounds become one people.
Every subgroup then acquires its own history, grievance narrative, representative organizations, demographic statistics, cultural claims, political advocates, and language of representation or underrepresentation. Politics begins treating citizens less as individuals participating in a common republic and more as demographic constituencies negotiating over power, which produces an incentive structure hostile to generalized trust.
Once political rewards attach to subgroup identity, political entrepreneurs acquire reasons to strengthen it. An individual grievance becomes a claim against “our community,” elections become demographic scorecards, appointments become evidence of representation, policy outcomes become racial distributions, crimes become intergroup symbols, economic disparities become claims between categories, and history becomes a contest among collective memories.
Citizens learn from the system’s incentives. If other groups organize as groups, if public resources flow through demographic claims, and if representation is defined by control of institutions, each person receives a rational reason to interpret politics through the same lens; a society can therefore teach tribalism while officially preaching inclusion.
The problem is not that cultures exist or that ancestry matters privately. The problem is a political order that continually teaches people that their most consequential identity is the one separating them from their neighbors, a movement precisely opposite the mechanism described by superordinate-identity research.
Igarashi’s finding that stronger multicultural policies were associated with a smaller national-identification gap among non-European immigrants suggests that cultural recognition need not compete with national identity. The implication, however, is that recognition becomes valuable insofar as it helps people feel more included in the national community; the national community remains the endpoint.
Telling a citizen that full American membership does not require pretending to have English ancestors can enlarge the American “we.” Describing America primarily as a federation of equally persistent cultural identities whose distinctions must receive permanent institutional recognition can weaken the very category into which integration is supposed to occur.
Multiculturalism consequently contains two trajectories. Respect for subgroup identity may serve as a bridge toward common identity, which can be prudent integration, or subgroup identity may become a permanent organizing principle of public life, which risks institutionalized fragmentation; a coherent society must distinguish the two sharply.
Diversity can sometimes contribute to strength because people bring different skills, knowledge, commercial networks, perspectives, languages, experiences, and cultural traditions into a common civilization. The slogan nevertheless smuggles its conclusion into its premise by assuming that difference itself creates strength while ignoring the architecture required to keep difference from producing distrust, faction, communication failures, weaker norms, or political fragmentation.
The empirical literature supplies no warrant for such confidence. The largest meta-analysis found a negative average relationship between ethnic diversity and social trust, especially at the neighborhood level; Putnam found social retreat rather than spontaneous cosmopolitan solidarity; collective-efficacy research explains why mutual trust and intervention matter; and the evidence on superordinate identity shows how larger common identities can make group boundaries less socially destructive.
Taken together, the research supports a more defensible proposition: diversity is a condition that can be successfully integrated into a strong society when sufficiently powerful trust-producing institutions, norms, and common identities exist. That proposition bears little resemblance to “diversity is our strength,” because the strength resides in the civilization capable of incorporating difference without dissolving itself.
A high-trust society possesses enough shared language for citizens to communicate across social boundaries and enough shared norms for strangers to anticipate one another’s conduct. Its moral and religious inheritance teaches obligations beyond immediate self-interest; its families form responsible people without confining duty to kinship; and its residential and historical continuity permit reputation, memory, and inherited expectations to accumulate.
Such a society also maintains law that citizens expect to be administered impartially, institutions worthy of confidence, and civic associations through which strangers repeatedly cooperate. It shares a national history and symbols that citizens can recognize as partly their own, thereby giving diverse people a political inheritance deeper than the temporary bargain of competing interests.
Above all, it possesses a sufficiently strong superordinate identity that allows people of different backgrounds to meet primarily as fellow citizens rather than as representatives of demographic blocs. For the United States, that identity must be American: not white American, black American, Hispanic American, immigrant American, native-born American, or Christian American as a legal caste, but American; adjectives may describe ancestry, region, denomination, and heritage, while the noun carries the greater civic weight.
A heterogeneous society faces a civilizational choice that a homogeneous society can often avoid: difference must either be subordinated to common citizenship or common citizenship will be subordinated to difference. The multicultural answer too often assumes that subgroup consciousness can intensify indefinitely while national solidarity survives through good intentions, an assumption the trust literature gives every reason to doubt.
Human beings naturally categorize, form in-groups, rely upon familiarity, and cooperate more easily when norms remain legible; none of those tendencies disappears because an institution hangs a diversity banner in its lobby. A successful heterogeneous civilization must therefore work against centrifugal pressures by maintaining centripetal institutions: common language, civic education, law, obligations, public norms, national rituals, integrated institutions, repeated cooperation, residential stability, historical memory, and a moral vocabulary that extends duty beyond tribe.
The proper political objective is integration compatible with individual liberty. People should remain free to preserve food, worship, family customs, music, festivals, languages, and ancestry voluntarily, while the republic remains confident enough to expect citizens to learn the common language, obey the law, understand the constitutional order, participate in common institutions, accept equal citizenship, develop loyalty to the country, and recognize obligations toward Americans who share none of their ancestry.
Children should learn that America is more than the geographic platform upon which an ancestral group happens to reside; it is their country and a real inheritance carrying reciprocal duties. Such an expectation is neither ethnic nationalism nor cultural erasure, but one of the ordinary requirements by which a nation remains a nation.
Multicultural ideology makes its gravest mistake when it assumes that enlightened societies can dispense with a common identity. Human psychology suggests otherwise, because removing one “we” does not create post-tribal individuals floating freely through rational space; smaller identities based on race, ethnicity, religion, ideology, class, sex, party, immigrant community, or region rush into the vacuum.
The question is never whether human beings will organize into groups, but which group identity will regulate the others when interests collide. The nation historically supplied that answer: black and white soldiers, Catholics and Protestants, immigrants and native-born citizens, New Yorkers and Texans could remain different while recognizing an American identity above them that placed demands upon all.
A society that emphasizes every identity below while delegitimizing the identity above should not be surprised when generalized trust becomes difficult to sustain. It is disassembling the psychological architecture that public policy will later be asked, unsuccessfully and at great expense, to reconstruct.
None of this requires hostility toward immigrants, racial uniformity, or the abolition of cultural traditions. It requires taking integration seriously enough to believe that immigrants deserve entrance into something real, recognizing that American identity can cross race, and preserving the freedom through which voluntary cultural life flourishes without turning cultural fragmentation into a public ideal.
Differences can enrich a society, generate creativity, and widen its stock of knowledge; they can also impose coordination costs, increase uncertainty, weaken interpersonal trust, encourage retreat, and make collective action harder. A mature political philosophy acknowledges both realities, whereas a slogan acknowledges only the flattering one.
Civerum’s position is therefore unapologetic: a civilization’s strength does not come from the number of differences it can enumerate, but from its ability to bind different people into a trustworthy social order. That order is neither spontaneous nor indestructible; it depends upon language, law, memory, norms, moral formation, families, faith, civic institutions, reciprocity, loyalty, and a shared understanding that lesser identities exist within one people bound by approximately the same rules.
A highly diverse America can remain strong and may become stronger by integrating talented and industrious people from throughout the world. That possibility depends upon what multicultural rhetoric too often treats as optional: assimilation into an American civic identity powerful enough to supersede subgroup loyalties when the two conflict.
Diversity, multiculturalism, and difference can exist safely and productively on top of a foundation, but none can serve as the foundation itself. The foundation is the shared civilization underneath them—language, law, memory, norms, moral formation, civic institutions, reciprocity, loyalty, identity, and trust—and the old American formulation understood the order of operations better than the modern slogan: E pluribus unum never celebrated the many as the final achievement, because the achievement was what came next: out of many, one.
