You can only really know about 150 people
A famous limit on human friendship has survived three decades of attacks — but the newest fight over it is about which kind of statistics you trust.
In 1992, anthropologist Robin Dunbar plotted primate group sizes against brain size and asked what number the same relationship would predict for humans. The answer was 150 — a figure that then turned up, oddly consistently, in hunter-gatherer bands, Christmas card lists, military companies, and Hutterite farming communities, all landing near the same ceiling of people someone can maintain as actual relationships, as opposed to names they recognize.
The number has taken hits for thirty years, most recently from a 2021 Stockholm University study claiming the real figure is closer to 70. Dunbar's counter isn't "trust me" — it's a statistics argument. The Stockholm team used a standard regression method built for cases where you can fix one variable precisely; brain size and group size across species don't work that way, and the more appropriate method pushes the estimate right back up near 150.
Either number would tell you the same uncomfortable thing: your phone's contact list vastly overstates how many people you actually, functionally know.
The argument about the exact ceiling keeps changing. The existence of a ceiling hasn't budged in thirty years.