An Outline of Friendship

Den neie.

I should like to eat your intestines.

In the highlands of Papua New Guinea, a Wandeki man shouts this phrase as an old friend comes to visit. At first glance, the expression is startling, invoking gory images of cannibalism. Even in islands not far from New Guinea, the promise of eating someone's body parts is a sign of anger and aggression. However, in the presence of a Wandeki friend, the phrase means something quite the opposite—unbridled affection and happiness at seeing a companion after a long separation. The greeting continues as the two men wrap their arms around each other and the visitor responds in kind, "A! Ene den neie!"—"Yes, I too should like to eat your intestines."

From the perspective of a European or American, the appropriate behaviors among friends in other cultures may appear bizarre and indeed unfriendly. Consider, for example, the obligation among Dogon farmers of Mali not only to attend a close friend's funeral, but also to dress in rags, overturn jars of millet porridge, and insult the generosity of the family. Among Bozo fishermen in the same region, friends demonstrate their love by making lewd comments about the genitals of one another's parents. Given these diverse and frequently counterintuitive behaviors between friends, how can we hope to define the relationship? As the Wandeki example suggests, focusing exclusively on overt behavior is not enough. The intentions, feelings, and thoughts behind those behaviors also matter in differentiating a hostile act from a gesture of friendship.

Accordingly, I examine friendship as an integrated social and psychological system defined not only by behaviors, but also by underlying feelings and motivations. Figure 3 illustrates this multidimensional view of friendship. Behaviors between friends are the most visible parts of the system—both to participants in a relationship and to researchers who attempt to observe it. In the course of daily life, behaviors such as gift giving and kind acts and words are about the only observables for grasping the workings of friendship. Below this visible surface of behavior, psychological processes, such as perceptions, feelings, and motivations, play a role in steering actions among friends. For the last half-century, psychologists have tried to get a handle on this submerged system through individual self-reports and behavioral experiments. Deeper still are physiological mechanisms, including the activity of neurons, neurotransmitters, and hormones. Researchers have only recently begun to investigate activity at this physiological level by measuring the concentration of specific chemicals in the body and taking pictures of blood flow in the brain. Although each of these analytical levels is equally important, the relative weight I give to each of them is also a function of how much time and effort researchers have devoted to their study. Therefore, the fact that there are so few observations about the physiological underpinnings of friendship says more about the relatively short time period in which they have been studied than about their relative importance in the functioning of friendship.

The chapter is organized into three sections that focus on key aspects of the friendship system, as shown in figure 3. The first section focuses on behavior and describes how two important activities among self-described close friends—helping and sharing—are not observed to the same extent among strangers and acquaintances. Moreover, it describes why three standard explanations for friends' increased generosity—a norm of reciprocity, an urge to balance accounts, and a concern about the shadow of the future—do not fit empirical findings from observation and experiment. The second section focuses on psychological constructs commonly used to describe feelings among friends—including closeness, love, and trust—and how these relate to behavior. The last section brings the discussion full circle by examining how people display and communicate these internal psychological states through behavior, and why the mutual communication and recognition of these feelings and intentions is an important part of maintaining a friendship.

A caveat is due here. Comparing claims about why friends help one another often requires carefully controlled experiments that can parse the precise relationships between variables such as subjective closeness and helping (see box 3). Therefore, I devote considerable time to describing such experiments. However, while such experiments serve an important purpose in the scientific process, they also come with limitations. First, the more tightly controlled an experiment, the more artificial and oversimplified it becomes, raising questions about how much it can tell us about behavior in the real world. This is a necessary evil of experimental research, and one to consider when interpreting its results. Second, researchers have traditionally found it easier to conduct such experiments in the United States and Europe (and most frequently on college campuses). Therefore, there is a substantial Western (and collegiate) bias in this chapter. I will attempt to remedy this in chapter 2 by looking to descriptions of friendship, like that of the Wandeki men, found in a wider range of world cultures.

HELPING AND SHARING AMONG FRIENDS, AND WHY THREE COMMON MECHANISMS CAN'T EXPLAIN THEM

Among Trobriand sea voyagers off the coast of Papua New Guinea, friends who trade with one another also provide support and lodging for one another when one of them is traveling. Among Baka Pygmy foragers and their farming neighbors in Central Africa, close friends (or loti) openly share and exchange material and social goods; male friends may even exchange wives. And in U.S. high schools, friends stand up for one another against verbal backstabbing, they keep important secrets, and they help talk one another through problems and conflicts. These examples illustrate a recurring expectation about friendship in its diverse manifestations around the world. Whether one asks a Wandeki gardener in Papua New Guinea or a Turkana cattle herder in East Africa, the reply will be the same: Close friends should share what they have and help one another in times of need.

Despite the ubiquity of an ideal of mutual sharing and support among friends, little quantitative data exists in cross-cultural settings to determine whether this ideal reflects real acts. In the U.S., at least, carefully designed experiments comparing self-described friends with...