The social identity approach to leadership and why it matters
In June 1954, two groups of a dozen 11-year-old boys alighted from separate buses in isolated Robbers Cave State Park in Oklahoma. For the next three weeks these young men would participate in what later became known as the Robbers Cave experiment. For the first week they would live in separation in different parts of the park, as the two groups separately bonded. In this week, one group would kill a rattlesnake and would proudly name themselves the Rattlers. The other group would name themselves the Eagles. In the next week, the groups were brought together to play competitive games. At this point all hell broke loose as the Eagles and the Rattlers competed and fought with each other. Then, in the study’s final week, the researchers set cooperative tasks for the boys. This involved them working towards shared goals rather than conflicting ones. This repaired the damage of the previous week and the boys went home on the same bus, with Eagles and Rattlers in some cases even riding together as friends.
Some years later, Henri Tajfel, a University of Bristol professor of social psychology, wondered what would be the minimal intervention that could get boys of approximately this age to divide themselves into separate groups— like the boys from Oklahoma. In this and in many subsequent experiments with different co-authors, he found that even the most minimal interventions would cause in-group favoritism and out-group discrimination. In the most famous of these experiments, the subjects were divided into a Klee group and a Kandinsky group, supposedly on the basis of their liking for paintings by these two abstract artists. Although in fact the division was random, the Klees subsequently preferred their fellow Klees and discriminated against those awful Kandinskys, while the Kandinskys symmetrically preferred fellow Kandinskys and discriminated against those awful Klees.
These experiments with schoolboys would hardly seem to be the origins for a serious book on the psychology of leadership, that most adult of subjects, traditionally concerned with the behavior of CEOs, generals, and presidents. But the behavior of the schoolboys in Oklahoma and Bristol brought into question assumptions that underpinned huge areas of psychology, and also huge areas of economics. The boys’ behavior also points to the theoretical underpinning for The New Psychology of Leadership. Why? Because in these experiments the schoolboys demonstrated that their motivation was different from the standard motivation described in economics and also from the standard behavior examined in psychology. More specifically, in the context of the experiments, the boys showed that they made a distinction between we and they. The we of the Rattlers, the they of the Eagles, or vice versa. The we of the Klees, the they of the Kandinskys, or vice versa.
Of course, to make such distinctions is a basic human propensity. The experimenters should not have been surprised that this occurred in Oklahoma, nor that it occurred in Bristol. It is seen in kids’ games of ball, where friends divide themselves into groups, often chosen with some randomness, and in more serious fights which can arise regardless of whether or not the other group is playing fair. Much more seriously, such we–they distinctions are seen in wars, where patriotic young men, and now women, put their lives on the line, to protect us against them. At the same time, in other contexts, individuals seek to establish a distinct identity for their in-group through acts of kindness and generosity towards out-groups. However, in every case the importance of us is paramount.
The division of we and they is therefore one of the most important features of human psychology. It is no coincidence that it should lie at the heart of the psychology of leadership, because understanding and engaging with such distinctions is basic to what leadership is all about.
Leadership has been perhaps one of the most written-about topics in all of history. As Haslam, Reicher, and Platow indicate, we can find discussions of the topic going as far back as Plato. But it is a major theme of yet older literature as well, since much of The Odyssey and The Iliad, the Vedas, and the Old Testament concern what leaders did and the outcomes of their decisions and actions, for good or ill. In modern times, more prosaically, leadership books, and biographies of leaders, take prime shelf space in airport bookstores. To give just one example, John C. Maxwell, a consultant who has made a list of the 21 “indispensible qualities” of a leader, claims to have sold more than 13 million copies of his many books.
But, as Haslam, Reicher, and Platow point out, there is something missing in the previous works on leadership. For when, like Maxwell, people consider a person as a potential leader, they typically consider the traits or qualities of the individual in question. Haslam, Reicher, and Platow show us how the psychology of leadership has been largely concerned with such individual attributes. But whatever truth there may be to this approach, it ignores the other side of the equation: it ignores the motivation of those who are to follow. It fails to recognize that the major role of the leader is to get those followers to identify themselves with a we whose goals are aligned with those of the leader. That, for the most part is what leadership is all about: it is about the interaction between the motivation and actions of the followers and the leader—and that motivation is mediated by how those followers think of themselves, and, correspondingly, how they define their goals.
I do not know of a literature in economics that explicitly claims to be about leadership, but economics’ handling of the theory of organizations tells us what such a theory of leadership would be. Traditional economics makes a different error from that of failing to consider the motivation of the followers. It considers their motivation, but too narrowly. The standard economics of organizations derives from the so-called “principal-agent model,” where there is a manager, who is called the “principal,” and there is a worker, who is called the “agent.” This agent must decide whether to follow the leader, and to what extent. In standard economics the agent only cares about his or her own self-interest. Agents do not care at all about doing what the leader would want them to do, or about fulfilling the goals of the organization, or even about doing well in the job to which they have been assigned. A typical first-year problem for economics graduate students is thus to derive the monetary incentives that the principal should give to the agent in the interest of the organization.
There are two reasons why this description of the relation between the principal and the agent is bad economics and also a bad description of the role of the leader. First, there is a yet more advanced literature in economics that shows that there are many ways in which the agent will game the system, rather than do what is in the principal’s interest; and, empirically, economists have verified that people are very smart at gaming those incentives. (This should be no surprise to dog owners; dogs are also smart in responding to incentives.) Thus organizations that rely only on their members’ personal self-interest and the provision of monetary incentives are likely to operate very badly.
But there is also a much more fundamental problem with this economics: it has left out the lessons of Robbers Cave and of the minimal group experiments. It has overlooked the fact that agents may also form a we, and that identification will be associated with goals that align or conflict with the goals of the organization. Insofar as the agents identify themselves with a we whose goals accord with those of their organization, that organization will make the best of its environment. But insofar as the agents identify with a we whose goals are counter to those of their organization, the organization will fall short of its potential; I think, in most cases, disastrously so.
Leadership is thus only partially about individual personality traits (the elementary psychology approach—although these traits may be of some importance). Leadership is also only partially about setting the right incentives (the elementary economics approach—although these incentives are also of some importance). This is where Haslam, Reicher, and Platow and their New Psychology of Leadership come in. They say something new and fundamental about leadership. It is not just about what leaders say and do; it is about what they say and do in the context of their followers’ willingness to identify as a we, who accordingly accept or reject what the leader wants them to do.
There is also a very special role for a leader in this process. When followers identify with a we, they almost invariably take on a notion of what we should or should not do. It is natural for followers, or potential followers, to define this notion of what they should or should not do in personal terms. For them, the leader serves as the role model—someone who sets the standards, who is the ideal, who is the focus of attention and the topic of gossip. Sometimes, the leader is even the protagonist in the creation myth of the group of we, as in the stories told in most firms about their founding. This can be seen in documents as disparate as the placemat menus of restaurants such as Legal Seafood or Hart’s Turkey Farm, a family restaurant in Meredith, New Hampshire. It is also seen in the annual reports of the great corporations, such as Goldman Sachs, IBM, and Microsoft.
People take stock in their group’s leader; the leader’s actions symbolize for them what they should or should not do. The leader is the archetypal “one of us.” In some cases leaders are so great that we cannot even aspire to be like them, but nevertheless their actions still indicate what we are supposed to do. To give but one example, consider Jesus Christ, who many consider the world’s greatest leader to date. For his followers, we are the Christians and our goal is to be like Him.
As Haslam, Reicher, and Platow set it out, a simple but profound theory underlies their New Psychology of Leadership. And that theory seems so very right that it may come as a surprise that this is not already the concept of leadership everywhere—from psychology and economics textbooks to the airport bookstores. But it is new because it runs counter to the major trends in both economics and psychology. In the case of economics it expands motivation to take into account our identification as a we, and the associated notion of how we should behave. That is new to economics.
In psychology, social identity theory, as the school of thought following Tajfel is called, is outside of the mainstream. A prominent psychologist once explained to me why. He said that the goal of the mainstream of psychology is to deduce how people think. As expressed by Nisbett and Ross, people are amateur scientists, who have “models” of how the world operates. The role of the psychologist is to deduce what those cognitive processes are, and how they differ from the thinking of real scientists. But this view of psychology rules out the possibility that people may have exactly the right model of how the world works, but want to do things that are peculiar to their group. Because it explores the nature of the we’s that people ascribe to, and the way in which these group memberships affect how they want to behave, social identity theorizing thus takes a very different perspective from mainstream psychology.
But it is precisely because The New Psychology of Leadership begins with such a novel perspective that it can give us such an original view. This captures the true structure of what leadership is all about. Accordingly, on almost every page of the text that follows there is a new subtlety about what leadership means and about how it works. It takes a subject older than Plato and as current as Barack Obama in a new and correct way. I am very much honored to have been asked to write the Foreword to this book. I hope that you, the reader, will appreciate it as much as I do.
George A. Akerlof
Berkeley, California
December 24, 2009