In 1900, the average life expectancy in wealthy nations was roughly 47 years. By 2100, demographers working with current projections anticipate average life expectancy approaching 100 in those same nations, with a non-trivial cohort living to 120 or beyond. This is not merely a health story. It is, with a directness that the public conversation has not yet absorbed, a friendship story. Friendship is a time-structured relationship. The assumptions built into it — how long it will last, how many of them a person can have, what constitutes abandonment versus simply distance, what you owe a friend over forty years versus over eighty — are calibrated to a particular temporal horizon. Longevity rewrites that horizon, and friendship has to be revised accordingly.

The first and most basic revision concerns duration. A friendship formed at age 25 in a world of 80-year lifespans could, if it survives, last 55 years. That same friendship formed in a world of 110-year lifespans could last 85 years. The arc of a single friendship will routinely exceed what previously required a multigenerational family saga. The pressure this places on friendship — its capacity to weather change, conflict, the transformation of the persons involved, the accumulation of debt and forgiveness and reinvention — is without historical precedent. Marriages of 80 years will not be unusual. Friendships of 80 years will not be unusual. Both will require frameworks for understanding and sustaining them that do not yet exist at scale.

The second revision concerns multiplicity. If a person lives to 110 and can form and maintain meaningful friendships through most of that lifespan, the total number of significant friendships over a lifetime increases substantially. But friendship is not merely additive. Research consistently shows that the human capacity for genuine social maintenance — the attentive presence, the reciprocal investment, the ongoing contact that distinguishes friendship from mere acquaintance — is constrained, probably by cognitive and temporal limits. Robin Dunbar's findings on social group sizes, which point to a ceiling of roughly 150 for stable social networks and 5-15 for intimate relationships, are themselves calibrated to a particular lifespan. Whether they hold in a world where people accumulate social history over 80 or 90 adult years is not yet known. Longevity will test those limits in ways no previous generation has experienced.

The third revision concerns the lifecycle of friendship itself. Friendships currently have a rough correspondence to life stages: childhood friendships, school friendships, early adulthood friendships, parenting-era friendships, late-life friendships. In a longevity world, the number of distinct life stages multiplies. A person living to 110 may experience not two or three major life transitions but five or six: early adulthood, career establishment, midlife, post-career active years, a long later-life period, a final elder period. Each stage creates distinct social conditions, distinct needs and capacities for friendship, distinct pressures on existing friendships and distinct opportunities for new ones. The friendship literature has barely begun to theorize a five-stage adult life, let alone a life with seven or eight distinct chapters.

The fourth revision concerns death and loss. In a shorter-lifespan world, the death of a friend is statistically concentrated in the last decade or two of life. In a longevity world, with people living into their hundreds, the accumulation of friend-deaths occurs over a longer span but the total count is higher. A person who lives to 110 will, if they have been a person of genuine friendship, have attended a very large number of funerals — will have outlived not one or two cohorts of friends but potentially three or four. The psychological and social weight of this is not a detail. The grief carried by the very long-lived, the capacity to keep forming friendships in the knowledge that they will most likely be the one who survives again, requires an entirely different orientation to friendship than the one that has sufficed in shorter-lived societies.

The fifth revision is structural. Institutions — neighborhoods, workplaces, religious communities, civic organizations — are among the primary infrastructure through which friendships form. Longevity will put pressure on all of these. If retirement at 65 is replaced by more varied patterns of work withdrawal — partial retirement, career reinvention, encore careers — the institutional basis of middle-age and late-life friendship formation changes with it. If more people live in active health into their 80s, the institutional spaces designed for older adults (retirement communities, senior centers) will need to be conceived differently, or new spaces will need to exist. The question is not just whether individuals will have friends in a longevity world, but whether the structural conditions for friendship — the contexts in which it forms, the institutions that sustain it — will be redesigned to match the new temporal reality.

Law 5 names this moment correctly: what is needed is not hope but revision. The friendship forms, norms, and institutions inherited from a world of 70-year lifespans will not survive 110-year lifespans intact. The revision required is honest, structural, and urgent.