Experimenting with the Intentional Stance

Untitled 3

Why do people leave a tip on the table even if they leave a restaurant unnoticed, in a place they will never visit again with no further interactions with the restaurant and its staff?

One reason lies in social preferences, such as an altruistic desire to reward a waiter who gave good service.  Economists tend to distinguish pure altruism (acting only out of concern for others) and impure altruism (acting out of the “warm glow” feeling none gets out of being charitable). In practice both motivations are at work together when, for example, one gives to charity. You may desire to help others but also enjoy feeling good about yourself for having done so. Social preferences have been widely studied by economists.

A different possible motivation behind tipping in the example above (and many similar cases) is “perceived intentionality”, our awareness of the waiter being a purposeful, intentional human being rather than a robot, irrespective of our motivation to reward him. When the other party has intentions we tend to treat it differently from unintentional forces, such as nature or machines. The restaurant guest who leaves without tipping feels shame when contemplating the waiter’s later disapproving reaction. Jean-Paul Sartre described this as “the gaze”: hell indeed is other people.

I have just glued my ear to the door and looked through a keyhole […] But all of a sudden I hear footsteps in the hall. Someone is looking at me […] I see myself because someone sees me […] Shame is by nature recognition. I recognise that I am as the Other sees me […] thus shame is shame of oneself before Others.
Jean-Paul Satre, Being and Nothingness

In our recently published experiment we wanted to tell apart social preferences and perceived intentionality in situations where our participants made decisions that affected others. In the experiment our participants could act either kindly or nastily in strategic games against three kinds of “others”: (1) human others who received money when participants played kindly, (2) human others who did not receive any money either way and (3) computer programs. The difference in the way participants treat (1) and (2) is down to social preferences, the difference between (2) and (3) is due to perceived intentionality. We used this latter difference to show that people act to avoid the disapproval of others irrespective of social preferences. They treat humans in (2) more kindly than computers in (3) presumably to avoid raising the humans’ ire (computers do not receive much sympathy from anyone although I have seen otherwise rational people argue and fight with them). We thus find evidence for disapproval avoidance: Our conclusion is that “a guest may tip the waiter not (only) out of concern for his income but to avoid the psychological cost of contemplating his resentment.” Philosopher Daniel Dennett calls this  the Intentional Stance.

This idea also explains findings in other experiments where simply presenting an image of eyes makes participants more pro-social, for instance in putting money in a donation box:


(There are apparently people immune to the gaze. Oxford don (and later Vice Chancellor) Maurice Bowra covered only his face when spotted sunbathing in the nude by a group of female students saying, “in Oxford I am known by my face”.)