"Today we've had a national tragedy," announced President George W. Bush, addressing the nation for the first time on September 11, 2001. "Two airplanes have crashed into the World Trade Center in an apparent terrorist attack on our country." Bush then promised "to hunt down and to find those folks who committed this act." These remarks, made from Emma T. Booker Elementary School in Sarasota, Fla., may not seem extraordinary, but in subtle ways they exemplify Bush's skill as a leader. When viewed through the lens of a radical new theory of leadership, Bush's 9/11 address contains important clues to how the president solidified his political power in his early months and years in office.

In the past, leadership scholars considered charisma, intelligence and other personality traits to be the key to effective leadership. Accordingly, these academics thought that good leaders use their inborn talents to dominate followers and tell them what to do, with the goal either of injecting them with enthusiasm and willpower that they would otherwise lack or of enforcing compliance. Such theories suggest that leaders with sufficient character and will can triumph over whatever reality they confront.

In recent years, however, a new picture of leadership has emerged, one that better accounts for leadership performance. In this alternative view, effective leaders must work to understand the values and opinions of their followers--rather than assuming absolute authority--to enable a productive dialogue with followers about what the group embodies and stands for and thus how it should act. By leadership, we mean the ability to shape what followers actually want to do, not the act of enforcing compliance using rewards and punishments.

Given that good leadership depends on constituent cooperation and support, this new psychology of leadership negates the notion that leadership is exclusively a top-down process. In fact, it suggests that to gain credibility among followers, leaders must try to position themselves among the group rather than above it. In his use of everyday language--such as "hunt down" and "those folks"--Bush portrayed himself on 9/11 as a typical American able to speak for America.

According to this new approach, no fixed set of personality traits can assure good leadership because the most desirable traits depend on the nature of the group being led. Leaders can even select the traits they want to project to followers. It is no accident, then, that Bush has often come across to Americans as a regular guy rather than as the scion of an elite East Coast Yale University dynasty.

But far from simply adopting a group's identity, influential presidents or chief executives who employ this approach work to shape that identity for their own ends. Thus, Bush helped to resolve the mass confusion on 9/11 in a way that promoted and helped to forge a new national unity. Among other things, people wondered: Who or what was the target? New York? Washington? Capitalism? The Western world? Bush's answer: America is under attack. By establishing this fact, he invoked a sense of a united nation that required his leadership.

From Charisma to Consensus
Nearly 100 years ago the renowned German political and social theorist Max Weber introduced the notion of "charismatic leadership" as an antidote to his grim prognosis for industrial society. Without such leadership, he forecast, "not summer's bloom lies ahead of us, but rather a polar night of icy darkness and hardness." Since then, the notion of charisma has endured, alternatively attracting and repelling us as a function of events in the world at large. In the chaos following World War I, many scholars continued to see strong leaders as saviors. But in the aftermath of fascism, Nazism and World War II, many turned against the notion that character determines the effectiveness of leaders.

Instead scholars began to favor "contingency models," which focus on the context in which leaders operate. Work in the 1960s and 1970s by the influential social psychologist Fred Fiedler of the University of Washington, for example, suggested that the secret of good leadership lies in discovering the "perfect match" between the individual and the leadership challenge he or she confronts. For every would-be leader, there is an optimal leadership context; for every leadership challenge, there is a perfect candidate. This idea has proved to be a big moneymaker; it underlies a multitude of best-selling business books and the tactics of corporate headhunters who promote themselves as matchmakers extraordinaire.

[break]

In fact, such models have delivered mixed results, contributing to a partial resurgence of charismatic models of leadership in recent decades. In particular, James MacGregor Burns's work on transformational leadership in the late 1970s rekindled the view that only a figure with a specific and rare set of attributes is able to bring about necessary transformations in the structure of organizations and society.

How, then, do we get beyond this frustrating flip-flop between those who argue that a leader can overcome circumstances and those who retort that circumstances define the leader? In our view, strong leadership arises out of a symbiotic relationship between leaders and followers within a given social group--and hence requires an intimate understanding of group psychology.

In the 1970s Henri Tajfel and John C. Turner, then at the University of Bristol in England, performed seminal studies on how groups can restructure individual psychology. Tajfel coined the term "social identity" to refer to the part of a person's sense of self that is defined by a group. As Turner pointed out, social identity also allows people to identify and act together as group members--for example, as Catholics, Americans or Dodgers fans. Social identities thus make group behavior possible: they enable us to reach consensus on what matters to us, to coordinate our actions with others and to strive for shared goals.

Tajfel and Turner's original social identity framework does not refer to leadership explicitly, but it helps to clarify why leadership requires a common "us" to represent. Leadership theorist Bernard Bass of Binghamton University has shown, for example, that leaders are most effective when they can induce followers to see themselves as group members and to see the group's interest as their own interest.

The emergence of social identity helps to explain the transformation in the strategies of rulers associated with the birth of modern nation states in the 19th century. According to historian Tim Blanning of the University of Cambridge, before national identities emerged European monarchs could only rule as autocrats, using power (rather than true leadership) to control people. But once people identified with nations, effective monarchs needed to rule as patriots who were able to lead the people because they embodied a shared national identity. Monarchs such as Louis XVI of France who misunderstood or ignored this shift literally lost their heads.

More recently, we affirmed the importance of social identities for leadership in an experiment we called the BBC Prison Study, an investigation of social behavior conducted within a simulated prison environment [see "The Psychology of Tyranny," by S. Alexander Haslam and Stephen D. Reicher; Scientific American  Mind, October 2005]. We randomly assigned volunteers to two groups: prisoners and guards. Surprisingly, we found that meaningful and effective leadership emerged among the prisoners but not among the guards, because only the prisoners developed a strong sense of shared social identity based on a common desire to resist the guards' authority. The guards, on the other hand, lacked a group identity, in part because some of them were not comfortable being in a position of authority; accordingly, they did not develop effective leadership and ultimately collapsed as a group.

One of the Gang
When a shared social identity exists, individuals who can best represent that identity will have the most influence over the group's members and be the most effective leaders. That is, the best leaders are prototypical of the group--they not only seem to belong to it but also exemplify what makes the group distinct from and superior to rival groups. For example, Bush was connecting with Middle America--intentionally or otherwise--when he littered his speeches with verbal gaffes, something that columnist Kevin Drum suggested in the Washington Monthly worked in Bush's favor in the 2004 election. Indeed, those who scoffed at Bush's awkward utterances suffered, because their criticism cast them as an alien elite out of touch with most ordinary Americans.

[break]

Even the way leaders dress can help them appear representative of the groups they lead. Bush's leather jackets and cowboy clothes round out the image of him as a regular guy. In the same vein, the late Palestinian leader Yasser Arafat adopted the headscarf of the peasantry to identify himself with his people. The founder of Pakistan, Muhammad Ali Jinnah, wore a dress made of distinctive items from the various regions of the new country, suggesting a newly unified national identity and establishing himself as its figurehead.

Such examples counter the notion that leadership requires a particular set of personality traits or that leaders should behave in a fixed way. The most desirable traits and actions have to fit with the culture of the group being led and thus vary from group to group. Even some of the most oft-touted leadership traits, such as intelligence, can be called into question in some settings. Some people consider being down-to-earth or trustworthy as more important than being brilliant, for instance. Where this is the case, being seen as too clever may actually undermine one's credibility as a leader, as Bush's tactics suggest.

Followers may also shun an otherwise desirable trait such as intelligence if doing so helps the group differentiate itself from competitors. In a study published in 2000 by Turner, now at the Australian National University, and one of us (Haslam), we asked business students to choose the ideal characteristics for a business leader. When the students were confronted with a rival group that had an intelligent leader (who was also inconsiderate and uncommitted), the students wanted their leader to be unintelligent (but considerate and dedicated). But when the rival leader was unintelligent, virtually nobody wanted an unintelligent leader.

If fitting in is important for gaining influence and control, then anything that sets leaders apart from the group can compromise their effectiveness. Acting superior or failing to treat followers respectfully or listen to them will undermine a leader's credibility and influence. Similar problems can emerge if a leader and followers are separated by a wide compensation gap. Financier J. P. Morgan once observed that the only feature shared by the failing companies he worked with was a tendency to overpay those at the top.

Another experiment of ours, which we reported in 2004, confirms Morgan's wisdom. We created work teams in which leaders' remuneration was either equal to, double or triple that of followers. Although varying the remuneration structure did not affect the leaders' efforts, team members' efforts diminished markedly under conditions of inequality. As the late Peter F. Drucker, then professor of management at Claremont Graduate University, wrote in his book The Frontiers of Management (Dutton, 1986), "Very high salaries at the top... disrupt the team. They make... people in the company see their own top management as adversaries rather than as colleagues.... And that quenches any willingness to say 'we' and to exert oneself except in one's own immediate self-interest."

Favoring Fairness
Another reason not to lavishly compensate those at the top is that followers are likely to perceive such financial inequity as unfair. Followers generally respect fairness in leaders, although what fairness means can depend on the followers. Ways to be fair as a leader include refraining from helping yourself and making sacrifices for the group. Gandhi won people over by adopting an Indian villager's dress, which symbolized his refusal of luxuries; Aung San Suu Kyi similarly attracted supporters with her willingness to endure ongoing house arrest to promote collective resistance to military rule in Myanmar (Burma).

Effective leaders can also display fairness in the way they resolve disputes among group members. Favoritism, or even the appearance of it, is the royal road to civil war in organizations, political parties and countries alike. In some cases, however, leaders should favor those who support their own group (the in-group) over those who support another group (the out-group).

[break]

In a 1997 study conducted by one of us (Platow) in New Zealand, people endorsed the leadership of a health board CEO who allocated time on a kidney dialysis machine equally between two fellow New Zealanders. Yet when the CEO had to split the time between a New Zealander and a foreigner, people liked the leader who gave more time to the in-group member. And in a 2001 study we asked Australian undergraduates about their support for a student leader, Chris, who had distributed rewards between student council members who were known to either support or oppose core student positions (regarding cuts to university funding, for example). Chris was more popular to the extent that he showed a preference for the council members who supported the in-group position. And when Chris showed such partiality, the undergraduates were more likely to back him and devise ways to make his proposed projects succeed [see box below].

People do not always prefer leaders who are biased against the out-group, however. A leader who represents a group that holds a strong belief in equality must treat in- and out-group members equally. Thus, when a member of the British Parliament recently put British families before migrants in allocating public housing for those in need, charitable groups, religious groups and socialist groups all protested strenuously. Good leadership does not mean applying universal rules of behavior but rather understanding the group to be led and the types of actions it esteems and considers legitimate.

Wielding Words
But, of course, leadership is not simply a matter of conforming to group norms. Anyone who is in the business of mobilizing people--whether to get them to the polls, to the office or to protest an injustice--must also work to shape and define those norms. Presidents and other leaders most often mold social identities through words, as Bush did in his 9/11 address.

The most effective leaders define their group's social identity to fit with the policies they plan to promote, enabling them to position those policies as expressions of what their constituents already believe. In the Gettysburg Address, which begins, "Fourscore and seven years ago our fathers brought forth upon this continent a new nation, conceived in Liberty, and dedicated to the proposition that all men are created equal," Abraham Lincoln strongly emphasized the principle of equality to rally people around his key policy objectives: unification of the states and emancipation of the slaves.

In fact, the Constitution contains many principles, and no one stands above all others, according to historian Garry Wills in his Pulitzer Prizewinning book, Lincoln at Gettysburg: The Words That Remade America (Simon & Schuster, 1992). Nevertheless, Lincoln elevated equality to a position of supreme importance and made it the touchstone of American identity. After Lincoln's address, Americans interpreted the Constitution in a new way. As Wills writes of the Gettysburg audience: "Everyone in that vast throng of thousands was having his or her intellectual pocket picked. The crowd departed with a new thing in its ideological luggage, that new constitution."

This reshaping of American identity as centered on equality allowed Lincoln to unite and mobilize Americans around freeing the slaves--a previously divisive issue. Through his skills as a wordsmith, this supreme entrepreneur of identity secured one of the greatest achievements in American history.

Identities and Realities
If Lincoln's definition of American identity moved people to create a more equal society, then the realities of emancipation served to reinforce equality as the core of American identity. That is, there is a reciprocal relation between social identity and social reality: identity influences the type of society people create and that society in turn affects the identities people adopt.

[break]

An identity that is out of kilter with reality and that has no prospect of being realized, on the other hand, will soon be discarded in favor of more viable alternatives. Our BBC Prison study provided a stark warning as to what happens if a leader's vision is not accompanied by a strategy for turning that vision into reality. In this study the collapse of the guard system led participants to set up a commune whose members believed passionately in equality. But the commune's leaders failed to establish structures that either promoted equality or controlled those who challenged the system. In the end, the commune also tottered, and the enduring inequality led even the most committed to lose faith. They began to believe in a hierarchical world and turned to a tyrannical model of leadership that would bring their vision into being.

The wise leader is not simply attuned to making identities real but also helps followers experience identities as real. In this vein, rituals and symbols provide perspective by reproducing a dramatized representation of the world in miniature. In her book Festivals and the French Revolution (reprinted by Harvard University Press in 1991), Mona Ozouf, director of research at the French National Center for Scientific Research, writes that the revolutionaries fashioned a whole new set of festivals to symbolize a France based on "liberty, equality, fraternity." In the past, people had paraded according to social rank, but now rich and poor paraded together, organized by age instead. In contrast, Adolf Hitler choreographed his Nuremberg rallies to portray an authoritarian society. He started among the masses, but at a strategic moment he would ascend a podium from where he could talk down to the serried and orderly ranks.

No matter how skilled a person might be, however, a leader's effectiveness does not lie entirely in his or her own hands. As we have seen, leaders are highly dependent on followers. Do followers see their leader as one of them? Do followers find their leader's visions of identity compelling? Do followers learn the intended lessons from rituals and ceremonies? Our new psychological analysis tells us that for leadership to function well, leaders and followers must be bound by a shared identity and by the quest to use that identity as a blueprint for action.

The division of responsibility in this quest can vary. In more authoritarian cases, leaders can claim sole jurisdiction over identity and punish anyone who dissents. In more democratic cases, leaders can engage the population in a dialogue over their shared identity and goals. Either way, the development of a shared social identity is the basis of influential and creative leadership. If you control the definition of identity, you can change the world.