Relationships5 min read
Every Relationship Has Its Own Radius
Why some friendships break the moment you get closer. Dunbar's layers, the hours it takes to make a friend, and the theory that explains why oversharing early goes so badly.
Schopenhauer told a story about porcupines. On a cold night they huddle for warmth, prick each other, move apart, get cold, huddle again — until they find a tolerable middle distance. Freud later quoted it in a footnote, and it became the standard image for human closeness.
The fable leaves one thing out. That middle distance isn’t a number. It’s a table. The right distance for a best friend, a colleague, and the person who runs your corner shop are not the same measurement — and treating them as if they were is how relationships go wrong.
(The distance between lovers is its own problem, handled here. This piece is about everyone else.)
Closeness comes in layers
Edward Hall measured the physical version in 1966: intimate distance (0–0.45 m) for embraces and whispers, personal distance (0.45–1.2 m) for friends, social distance (1.2–3.7 m) for colleagues and acquaintances. A 2017 study of 8,943 people across 42 countries found the dial is set differently everywhere — Argentinians are comfortable about 76 cm from a stranger; Romanians want 140.
The psychological version has a map too, drawn by Oxford anthropologist Robin Dunbar: roughly 5 people in the support clique (the ones you can call at 3 a.m.), 15 in the sympathy group, 50 friends, and 150 as the ceiling on stable, meaningful relationships. Beyond that, acquaintances.
Dunbar's number applies to quality relationships, not to acquaintances.
邓巴数说的是有质量的关系,而不是点头之交。
Dunbar’s own 2021 essay stresses what the layers really are: a budget constraint. Time and emotional bandwidth are finite. Someone entering an inner layer means someone else drifting out.
Moving inward costs hours — and work hours don’t count
Jeffrey Hall’s 2019 research put a price on it: about 50 hours together to go from acquaintance to casual friend, roughly 90 to friend, 200+ to close friend.
The best detail in the study is what doesn’t count. Time spent working together doesn’t move the needle. What does is leisure — meals, jokes, conversation about anything other than the task. Which explains a familiar puzzle: three years beside the same colleague and still not friends. You accumulated shift hours, not friendship hours.
Why premature intimacy backfires
The most common accident on this map is disclosing deeper than the relationship is.
Social Penetration Theory (Altman & Taylor, 1973) models personality as an onion: an outer layer anyone may see, an inner core of values, fears, and wounds. Relationships deepen when both people move inward gradually and reciprocally. It fails in two recognisable ways — disclosure that outruns the relationship (the second date that opens with childhood trauma), and disclosure that isn’t mutual (a relationship where only one person opens up isn’t intimacy, it’s a workload).
The fix isn’t to talk about the weather forever. It’s to let depth track the layer: they answer one level down, you go one level down.
Weak ties are not failed strong ties
A warning against reading the layers as a hierarchy. Granovetter’s 1973 classic showed that new information, opportunities and cross-cutting perspectives arrive mostly through weak ties — your close circle’s world overlaps too much with your own to bring news. Sandstrom and Dunn added the emotional half in 2014: brief exchanges with peripheral people — the regular barista, someone in the next seat — measurably raise happiness and belonging that day.
The current results highlight the power of weak ties.
研究结果凸显了弱联系的力量。
An acquaintance’s job is to be an excellent acquaintance. Staying weak is where the value comes from. Not every relationship needs promoting, the way not every houseplant needs to become a tree.
Work: where the radius is most often misjudged
Colleagues deserve their own note, because the setting lies to you: physical distance sits in the intimate zone (you see them daily) while the relationship sits at layer 50 or 150. Lisa Rosh’s advice in Harvard Business Review compresses to a sentence — be yourself, but be it very carefully. Disclosure at work should be small, gradual, situation-appropriate, and in service of the work rather than your own need for comfort; performed authenticity gets detected and does damage that’s hard to repair.
Using the map
- Sort your contacts into layers. Then calibrate expectations by layer. “He didn’t even check on me” — he’s at 50; that was never his post.
- To promote someone, spend the hours that count. Not overtime. The lunch, the game after work.
- Let disclosure follow response. Before going a level deeper, check whether the last level was caught. If it wasn’t, stay — that’s not guardedness, it’s respect for the relationship’s actual radius.
The porcupines never found a distance. They found a whole set of them: nearer to this one, further from that one, and for most, a civil nod.
Which is also what this site is named after. An echo needs the right distance — too close and it’s noise, too far and nothing returns.
Sources
- Robin Dunbar, Dunbar’s number: why my theory has withstood 30 years of scrutiny, The Conversation, 2021; The Anatomy of Friendship, Trends in Cognitive Sciences, 2018
- Jeffrey Hall, How many hours does it take to make a friend?, JSPR, 2019 (University of Kansas release)
- Social Penetration Theory (Altman & Taylor, 1973)
- Mark Granovetter, The Strength of Weak Ties, AJS, 1973
- Sandstrom & Dunn, The Surprising Power of Weak Ties, PSPB, 2014
- Lisa Rosh, Sharing Personal Information Can Build Trust on Your Team, HBR, 2024
- Sorokowska et al., Preferred Interpersonal Distances, JCCP, 2017