These matter-of-fact instructions on how to dis-relate send a chill down the spine of those raised on the traditional notion that friendships are for the long haul. But before accusing the business world of undue ruthlessness, it is important to recognize that precisely this pattern is employed, often beneath a veil of hypocritical regrets, in other strata of society as well. The professor who is promoted to dean, the military officer, the engineer who becomes a project leader, frequently play the same social game. Moreover, it is predictable that something like this pattern will soon extend far beyond the world of work and formal organization. For if friendship is based on shared interests or aptitudes, friendship relationships are bound to change when interests change – even when distinctions of social class are not involved. And in a society caught in the throes of the most rapid change in history, it would be astonishing if the interests of individuals did not also change kaleidoscopically.

Indeed, much of the social activity of individuals today can be described as search behavior – a relentless process of social discovery in which one seeks out new friends to replace those who are either no longer present or who no longer share the same interests. This turnover impels people, and especially educated people, toward cities and into temporary employment patterns. For the identification of people who share the same interests and aptitudes on the basis of which friendship may blossom is no simple procedure in a society in which specialization grows apace. The increase in specialization is present not merely in professional and work spheres, but even in leisure time pursuits. Seldom has any society offered so wide a range of acceptable and readily available leisure time activities. The greater the diversity available in both work and leisure, the greater the specialization, and the more difficult it is to find just the right friends.

Thus it has been estimated by Professor Sargant Florence in Britain that a minimum population of 1,000,000 is needed to provide a professional worker today with twenty interesting friends. The woman who sought temporary work as a strategy for finding friends was highly intelligent. By increasing the number of different people with whom she was thrown into work contact, she increased the mathematical probability of finding a few who share her interests and aptitudes.

We select our friends out of a very large pool of acquaintanceships. A study by Michael Gurevitch at the Massachusetts Institute of Technology asked a varied group to keep track of all the different people with whom they came in contact in a one hundred-day period. On average, each one listed some 500 names. Social psychologist Stanley Milgram, who has conducted a number of fascinating experiments dealing with communication through acquaintanceship networks, speaks of each American having a pool of acquaintanceships ranging from 500 to 2,500.

Actually, however, most people have far fewer friends than the twenty suggested by Professor Florence, and perhaps his definition was less restrictive than that employed in everyday use. A study of thirty-nine married middle-class couples in Lincoln, Nebraska, asked them to list their friends. The purpose was to determine whether husbands or wives are more influential in selecting friends for the family. The study showed that the average couple listed approximately seven "friendship units" – such a unit being either an individual or a married couple. This suggests that the number of individuals listed as friends by the average couple ranged from seven to fourteen. Of these, a considerable number were non-local, and the fact that wives seemed to list more non-local friends than their husbands suggests that they are less willing than their husbands to slough off a friendship after a move. Men, in short, seem to be more skilled at breaking off relationships than women.