A friend and I used to speculate about the role of proximity, layout of tables, etc, in influencing the chances of meeting new people. The pub conversations (minus the anecdotally useful bit about table topologies*—it’s a very slow running process) were partly satirised in Fugard (2006). But I wasn’t really joking.
Anyway, was pleased to see Back, Schmukle, and Egloff (2008) have performed a Real Experiment on related issues. I’ll let you read the paper and judge for yourself if you believe it, but the punchline is:
“coincidentally being near another person or being in the same group with him or her during an initial encounter may promote the development of a friendship with that person. In a nutshell, people may become friends simply because they drew the right random number. Thus, becoming friends may indeed be due to chance.”
Fugard, A. (2006). A theory of hubs, ruins, and blockers. Bluebook note 1555, Mathematical Reasoning Group, The University of Edinburgh.
Mitja D. Back, Stefan C. Schmukle, Boris Egloff (2008). Becoming Friends by Chance. Psychological Science, 19 (5) , 439–440.
* If you want to experience the phenomena yourself, and live in Edinburgh, try going upstairs in Opium, to the GRV, or to Ecco Vino. All very different places, but note the table layouts. In each of these places I have found the likelihood of speaking to some random person/people much higher than elsewhere.