For most of recorded history, intense same-sex friendship and intense kin-based partnership occupied a larger share of the emotional and practical center of adult life than they do now. The 19th-century concept of romantic friendship, the medieval European institution of sworn brotherhood (affrèrement in southern France, sometimes ritualized in church), the historical patterns of women living together as life partners that Lillian Faderman documented under the now-loaded phrase "Boston marriage" — these were not deviant arrangements. They were one of several culturally available configurations. What happened over the late 19th and 20th centuries is that the romantic-sexual dyad colonized the partnership space, pushing friendship into secondary status. The current cultural moment, characterized by what Rhaina Cohen calls "the other significant others," is partly a recovery of older possibilities and partly the invention of new ones suited to current conditions. This is Law 5 (Revise) operating on the question of which relationships count as primary.
The friendship-as-primary partnership model means treating one or more close friends as the central organizing intimates of adult life — sharing housing, finances, caregiving, life decisions, and long-term commitment in the ways previously routed almost exclusively through romantic partnership. The model is not a rejection of romance. Some people who build friendship-primary lives also have romantic partners but do not center them. Others are aromantic, asexual, or simply uninterested in romantic centrality. Still others are in conventional romantic relationships but maintain a friendship that holds equivalent or greater weight in practical and emotional terms. The diversity of routes into the configuration is one of the reasons it is hard to count, hard to characterize, and increasingly visible.
The empirical case for the configuration's expansion rests on several converging trends. Marriage rates are falling across the developed world. Median age at first marriage has risen by roughly six years since 1970 in the United States. Cohabitation rates among non-romantic adults — friends sharing housing into their 30s, 40s, and beyond — have risen, partly driven by housing cost but partly by preference. The Survey Center on American Life's 2021 State of American Friendship report documents that Americans report fewer close friends than they did three decades ago, but the friends they do report are weighted with more functional importance. The remaining friendships are doing more work.
The structural drivers are several. Housing cost in major metropolitan areas has made dyadic homeownership unattainable for a substantial fraction of adults; friend co-buying becomes a rational response. The fragility of romantic partnership at current divorce rates means relying on a single dyadic intimate is structurally riskier than it once was; distributing intimacy across multiple deep ties is partly an insurance strategy. Geographic dispersion of family of origin means kin-based caregiving networks are weaker than they were three generations ago; friends step into the role. The rise of asexual and aromantic identity, with their associated queerplatonic vocabulary, has given the configuration explicit articulation that earlier generations had to improvise.
What friendship-primary partnership looks like in practice varies. Two friends buying a house together and raising children, one through donor conception and the other through adoption. A trio of women sharing a multi-generational home and pooling retirement savings. A queerplatonic pair who have legally registered as domestic partners despite no romantic-sexual content. A widowed person and a never-married friend who move in together in their sixties and designate one another as primary kin. The legal scaffolding for these configurations is uneven. Some jurisdictions allow non-romantic domestic partnership; most do not. Some workplaces extend benefits to designated household members regardless of romantic status; most do not. The configurations work, but they have to be built from the ground up rather than inherited from existing institutions.
The collective revision the friendship-primary model implies is significant. It pushes against the late-modern assumption that romantic partnership is the only legitimate site for serious adult commitment. It elevates friendship from supplement to potentially central. It opens space for asexual, aromantic, and divorced or widowed people whose lives are not romantically organized. It also forces the broader culture to confront how thin its scripts for friendship have become — there are no widely recognized rituals for friendship commitment, no legal frameworks for friendship dissolution, no cultural language for the emotional labor of long-term friend partnership. The vocabulary has to be built.
Whether the configuration generalizes or remains a minority configuration is the question. The optimistic case is that it spreads as the housing and partnership-stability pressures that drive it continue to compound, and that the cultural permission unlocked by asexual, aromantic, and queer pioneers spreads into the broader population. The pessimistic case is that without legal, financial, and ritual scaffolding, friendship-primary partnerships remain difficult to sustain, and most people who attempt them eventually default back to either romantic coupling or solo living with weaker friend ties. The realistic case is that the configuration becomes one stable option among several rather than the new default — a meaningful pluralism rather than a wholesale replacement. Law 5 says revise the model when the model fails. The friendship-primary partnership model is one revision the collective is testing.