I play music through a Bluetooth speaker on the subway. People often look unhappy about this. What interests me is how rarely their behavior supports the intensity of the objection people claim to have on their behalf.
On yesterday's journey, approximately thirty passengers shared the carriage with me for some portion of nine stops. Several looked at the speaker. One woman asked whether I had headphones. Nobody stopped the music. We should be able to discuss what this observation implies without first agreeing that I am a bad person.
Revealed preference concerns the choices people make under actual constraints. A passenger can dislike the music and still prefer its continuation to the available methods of ending it. Remaining seated while I finish a track is one such choice. Approaching me, pressing the power button, or removing the speaker would be others. The first option was overwhelmingly preferred.
The obvious objection is that confrontation has costs. I agree. If D is the disutility of hearing my music and K the expected cost of stopping me, continued tolerance is consistent with D < K. Critics seem to think mentioning K invalidates the observation. It is the reason the observation is informative. Preferences that cease to govern behavior as soon as acting on them becomes unpleasant are weaker than their verbal expression suggests.
There is also a baseline assumption that needs defending. Silence is treated as something the carriage already owns, while my music is treated as an intervention requiring unanimous consent. But a preference for silence is still a preference. My enjoyment is unusually easy to identify: I selected the track, bought the speaker, and brought it with me. The opposing welfare claim is assembled largely from facial expressions interpreted by people who already dislike public music.
I do not infer that every passenger enjoys the track. That would be too strong. My claim is that their dislike has not crossed the threshold at which they choose to do anything effective about it. A request to turn it down supplies information about the direction of a preference. Their response after I decline supplies information about its strength.
Some will invoke a collective-action problem: everyone wants someone else to intervene. Even granting this, it remains unclear why the only participant successfully obtaining his preferred outcome should be required to solve the coordination problem of those who prefer another one. I have not prevented them from organizing.
Public space necessarily accommodates conflicting preferences. We need some practical way to distinguish an objection that constrains conduct from one that merely accompanies it. I use actual intervention. It has the advantage of being observable, and so far it has allowed me to listen at the volume I prefer.