Communication Assignment

JOURNAL OF LEADERSHIP STUDIES, Volume 4, Number 4, 2011 ©2011 University of Phoenix

View this article online at wileyonlinelibrary.com • DOI:10.1002/jls.20190 29

Groupthink, as traditionally conceived, is a failure in group decision making that occurs in highly co-

hesive groups. In the current case study, we propose an alternative potential form of groupthink in

which the group’s cohesiveness results from the shared pursuit of a collective goal rather than from

strong interpersonal bonds between members. Our model, recognizing the multifaceted nature of

cohesion, assumes that a group whose members are united in pursuit of a valued collective goal

while guided by a directive leader may experience breakdowns in the decision process. Specifically,

drawing on reviews of personal accounts, media descriptions, online interviews, and past empirical

papers, we propose that the May 1996 Mount Everest disaster can be understood in part from a

groupthink perspective. Applications for the proposed model are discussed, along with implications

for leaders seeking to improve organizational decision-making practices.

L E A D E R S H I P I N E X T R E M E

C O N T E X T S : A G R O U P T H I N K

A N A L Y S I S O F T H E M A Y 1 9 9 6

M O U N T E V E R E S T D I S A S T E R

JENI L. BURNETTE, JEFFREY M. POLLACK, AND DONELSON R. FORSYTH

On May 10, 1996, two teams of climbers set off from their camp high on Mount Everest in an attempt to reach the summit. New Zealander Rob Hall, recognized as one of the world’s most reputable Everest expedition guides, was the leader of the Adventure Consultants Guided Expedition. The other team, Mountain Mad- ness Guided Expedition, was led by esteemed and re- spected American guide Scott Fischer. Although members of both teams reached the summit, they met

with disaster. Many of the climbers lost their way in a storm and five of them—including Hall and Fischer— perished. Death is not uncommon for climbers on Ever- est. However, when a disaster of this magnitude occurs, involving well-designed teams guided by skilled and ex- periences leaders, we must ask, “Why?”

Experts, in both mountaineering and in human be- havior, have attributed the May 1996 Everest tragedy to a host of factors, including changes in the climbing

30 JOURNAL OF LEADERSHIP STUDIES • Volume 4 • Number 4 • DOI:10.1002/jls

unity of the mountaineering expeditions was not based on friendship, esprit de corps, or camaraderie but on members’ commitment to their quest. We suggest groupthink can also occur in groups of this type, brought together to reach a common goal. In this groupthink model, task cohesion takes the place of in- terpersonal cohesion as the necessary precondition for decisional dysfunction when coupled with directive leadership and a provocative context.

Keeping with the tradition of groupthink theory de- velopment, we use a case study analysis to identify the qualities of groups that make them prone to suffer from groupthink. Thus we first describe the events surround- ing the tragedy of the attempted ascent of the summit of Mount Everest in 1996, drawing on archival materials that present a description of the events, including the synthesis of first-person accounts written by Krakauer (1997), Boukreev (with De Walt, 1997), Gammelgaard (1996), and Weathers (Weathers & Michaud, 2001). In light of Elmes and Frame’s (2008) argument that re- countings of the May 1996 episode are filled with myth, we also considered the viewpoints of less-prominent team members. After describing the context of the dis- aster, we examine the two expeditions to determine if they meet the conditions of the proposed groupthink theoretical model. We review scholarly work and histor- ical perspectives regarding the multidimensional nature of cohesion before elaborating on the contribution of directive leadership style and provocative contexts. We conclude with implications for leaders operating under extreme contexts and applications for organizational de- cision-making practices more generally.

The Everest Disaster of May 1996 A number of expeditions attempted to reach the sum- mit of Mount Everest in the summer of 1996, but two of these teams seemed destined for success. Rob Hall, leader of Adventure Consultants Guided Expedition, had reached the summit of Mount Everest four times and in the process had led thirty-nine climbers safely to the top. Scott Fischer, who had summited Everest once before, was the head guide on the Mountain Madness Guided Expedition and had a reputation as an impec- cable climber and talented guide (Dowling, 1996). Both of these groups were commercial expeditions led by

culture, narcissistic personality characteristics, and over- crowding. Sir Edmund Hillary, one of the first climbers to reach the summit of Everest, blamed commercializa- tion and the “erosion of mountaineering values” (quoted in Dowling, 1996, p. 41). Medical experts noted the advanced age of some of the climbers; research shows that Everest favors the young, particularly on the de- scent (Huey, Salisbury, Wang, & Mao, 2007). Psychol- ogists Elmes and Berry (1999) used a blend of psychodynamics and structural theory to argue that narcis- sism and regressive dynamics led to the disaster. Mangione and Nelson (2003) suggested that scapegoating may have undermined the groups’ effectiveness. In his anal- ysis of the tragedy, Kayes (2004, 2006) stressed ineffec- tive team learning with precursors such as narrowly defined purpose and ill-defined problems. Roberto (2002) argued that system complexity, team structure, and cognitive limitations offer a holistic approach for understanding high-stakes decision making such as the Everest disaster. In a recent analysis, Elmes and Frame (2008) suggested that a more critical perspective on the Everest 1996 events is needed.

Extending groupthink theory, the current work ex- amines the May 1996 disaster and offers an explana- tion for the incident by drawing on one of the venerable theories of group decision making. Groupthink, as Janis (1972, 1982) explained, is a distorted style of thinking that renders group members incapable of making sound decisions. After examining a series of famously bad de- cisions, Janis identified three key antecedents of group- think: high interpersonal cohesion, a provocative context (e.g., situational pressures, isolation from other groups), and strong leaders who stated their preferences clearly (i.e., directive leadership style).

For Janis, the critical condition for groupthink was a sense of strong solidarity, friendship, and respect within the group. The Everest climbing teams were not cohe- sive, at least in this interpersonal sense. As one mem- ber wrote, we “were a team in name only” (Krakauer, 1997, p. 163). The Everest groups were, however, ex- tremely cohesive in a different sense of the word: they were united in a shared commitment to their task. Al- though interpersonal cohesion is often indexed by the strength of group members’ bonds to one another and to the group, task cohesion depends on members’ shared drive to accomplish their goals (Guzzo, 1995). The

JOURNAL OF LEADERSHIP STUDIES • Volume 4 • Number 4 • DOI:10.1002/jls 31

professional high-altitude climbers who guided the clients. Hall’s team consisted of 15 members and profes- sional guides Mike Groom and Andy Harris, and local guides, known as sherpas, who play an integral role on expeditions. Fischer’s team consisted of 12 members, guides Neal Beidleman and Anatoli Boukreev, and sher- pas. After nearly six weeks on the mountain acclimatiz- ing and preparing, the expeditions set out for the summit in the early morning of May 10, 1996, from Camp IV, which was established at 26,100 feet.

The climb to the summit is a choreographed under- taking. Midnight departures are common on Everest since they improve the chances of an early summit and a safe return to Camp IV before late afternoon storms arrive and darkness falls. Expeditions also typically es- tablish and adhere to deadlines for reaching the sum- mit. This turn-around time is established since descending in the dark or in poor weather adds unnec- essary risk to the most difficult part of the climb: the descent, where “fatigue and inattention” are most prevalent (Graydon & Hanson, 1997, p. 83). Addi- tionally, the turn-around time is especially important on Everest because most climbers use supplemental oxygen at higher altitudes. When teams set off from Camp IV, they carry enough oxygen canisters so that at a conservative flow rate they have 16–17 hours for the summit push and return. Thus, by 4:00 or 5:00 p.m. climbers run out of oxygen, making hypothermia, frostbite, high-altitude cerebral edema (HACE), and high-altitude pulmonary edema (HAPE) more likely (Krakauer, 1997). Teams adhere to a set turn-around time to minimize excessive risk of lack of oxygen, storms, and descending in the dark. On Everest, a noon turnaround is cautious, and a 2:00 p.m. time is consid- ered risky.

During the May 10, 1996, attempt to reach the sum- mit, time started to become more of a factor as climbers waited to ascend the “Balcony” section of the climb, which requires fixed ropes—climbing lines that are an- chored to the mountain and used as an extra safety pre- caution in especially steep or exposed areas. Numerous climbers all attempting the summit on the same day made the more technical parts of the climb slower, and climbers began to worry about the fast- approaching designated turn-around time. By 2:00 p.m. only Neal Beidleman, Andy Harris, Anatoli Boukreev,

Jon Krakauer, Klev Schoening, and Martin Adams had summited. The remaining expedition members should have turned around at that time. Climbers in both teams, however, continued to the summit until about 3:30 in the afternoon. This decision to ignore the turn- around time, combined with a (relatively common) late afternoon storm that brought snow, winds, and colder temperatures, resulted in clients running out of supple- mental oxygen during the descent. Three groups of clients and guides became stranded and lost sight of their camp. Beidleman, who had started to lead a phys- ically deteriorating group of clients that included Pittman, Gammelgaard, Fox, and Madsen toward safety, had trouble finding the camp during the descent. As the conditions continued to worsen, the group was un- able to continue. Hall and a client, Doug Hansen, were stranded just below the summit, and Fischer and one of the sherpas, Lopsang Jangbu, were in trouble at about 27,200 feet.

Hall, Harris, Yasuko Namba, Hansen, and Fischer all died that evening. The clients in the group guided by Beidleman could have died, as well, had it not been for the skill of Beidleman, Boukreev, and the sherpas. Beidleman, during a break in the weather, used the stars to identify the direction of Camp IV. The clients who could walk on their own (Schoening and Gammelgaard) set out with Beidleman, Groom, and the two sherpas to get help and managed to reach the safety of Camp IV. By 4:30 a.m.—nearly dawn—Boukreev had managed to rescue Pittman, Fox, and Madsen. However, Weathers and Namba were left for dead. Weathers miraculously managed to stumble back to Camp IV despite severe frostbite and hypothermia. Of the six climbers on Hall’s team to reach the summit, only Mike Groom, a guide, Krakauer, and Weathers made it down alive.

Although such a disaster is without a doubt a com- plex issue with multiple contributors, we propose that the climbers proceeded, under the direction of the leader, to continue beyond the turn-around time, and this decision was triggered by groupthink.

Groupthink Analysis Not all groups that fail suffer from groupthink. Corpo- rate boards make faulty decisions not because of a

32 JOURNAL OF LEADERSHIP STUDIES • Volume 4 • Number 4 • DOI:10.1002/jls

Building on previous research offering alternatives to interpersonal cohesion as a key antecedent, we too argue for another potential contributor. Namely, in the cur- rent analysis we use the events of the May 1996 Everest disaster to propose that task cohesion can replace inter- personal cohesion as a catalyst for concurrence seeking. Additionally, we focus on the role of directive leader- ship as a key component. We answer three fundamen- tal questions suggested by Janis, before concluding that the decision on Mount Everest was due, at least in part, to groupthink. First, was the outcome the result of the collaborative actions of individuals? Second, did the group make not just a mistake but instead a blun- der that could have been avoided? Third, were the an- tecedent conditions (i.e., cohesion, directive leadership, and provocative situational context) and groupthink symptoms (e.g., concurrence seeking) present? In an- swering the question regarding antecedent conditions, we review scholarly work on the history of cohesion as a multifaceted construct and emphasize the critical im- pact of the leader. Initially, however, we focus on our reasoning for suggesting the group as the decision maker and the faulty nature of the decision.

THE GROUP AS DECISION MAKER

Groupthink, by definition, is a group-level process. In- dividuals can make bad decisions, and they do so with remarkable regularity, but only a group experiences groupthink. However, because groups make decisions in a variety of ways, the question “Was this a group de- cision?” is not easily answered. In some of the cases Janis (1972) analyzed, the leader of the group made final de- cisions for the group, but he did so with input from his advisors. The leader took responsibility for the outcome, but his advisors shared in the decision process (Abbasi & Hollman, 1991; Raven, 1998). Similarly, on Mount Everest the head guide structures the expedition and is responsible for the outcome, but the other guides and expedition members take part in the process. The shared nature of the group decision is indicated, indirectly, by the degree of guilt expressed regarding disregard for the turn-around time. For example, one guide, Beidleman, remarked “now I kick myself for it,” where it refers to the decision to continue climbing (Krakauer, 1997). We suggest that regarding the turn-around time this was a group decision, albeit one made primarily by the

“deterioration of mental efficiency, reality testing, and moral judgment” (Janis, 1972, p. 9) but because their members lack requisite skills and experiences (LeBlanc & Gillies, 2005). As Kowert (2002) explains, errors by U.S. President Ronald Reagan and his advisors, such as their handling of the budget crisis and the Iran-Contra situation, resulted from Reagan’s decision-making style and the advisors’ inability to reach agreement. In many cases, business failures, such as the Millennium Dome in London and cost overruns at Denver International Airport, can be blamed on basic foibles of human deci- sion makers, such as the tendency to overestimate one’s capabilities, and the strong desire to recoup sunk costs (Nutt, 2002).

In the current analysis, we propose that groupthink remains a viable explanation for some of the fiascos and blunders that continue to plague groups (Baron, 2005; Moorhead, Neck, & West, 1998). Namely, we suggest that Hall’s and Fischer’s teams did not make faulty judgments due to lack of necessary skill, nor did they consciously make a risky choice. Rather, we propose that the groups succumbed to groupthink, a style of thinking that rendered them unable to consider all necessary components and consequences. Groupthink occurs apart from the usual sources of human error, as groups fall prey to concurrence-seeking tendencies (Whyte, 1998). Although Janis’s original model ar- gued that the key cause of consensus seeking resulted from interpersonal cohesion, more recent analyses in- dicate that additional variables may help explain the tendency to seek agreement (Street & Anthony, 1997). For example, a groupcentrism model suggests that groups tend to rush to make judgments on the basis of in- sufficient information (Kruglanski, Pierro, Mannetti, & De Grada, 2006). In a re-analysis of many of Janis’s original case-studies, Whyte (1998) suggested that groupthink-style decision making stems primarily from collective efficacy rather than interpersonal cohesion. A self-regulatory model of groupthink argues that goals and feedback on goal pursuits are relevant antecedents (Flippen, 1999). Baron (2005), after reviewing much of the existing research on Janis’s theory, agreed with Janis that members of groups often strive for consen- sus, but his ubiquity model of groupthink suggested that a threat to shared social identity was the driving force.

JOURNAL OF LEADERSHIP STUDIES • Volume 4 • Number 4 • DOI:10.1002/jls 33

leaders. The shared commitment to reaching the sum- mit, the closed nature of the leadership, the extreme cir- cumstances, and ultimately the subsequent consensus seeking influenced the leaders’ decision to push on well beyond a safe turn-around time.

MISTAKE OR AVOIDABLE BLUNDER?

Classifying the Everest disaster as a case of groupthink requires answering a second key question: “Was the de- cision to continue objectively faulty?” An Everest climb is a dangerous undertaking, even under the best of con- ditions with the most skilled and well-led climbers. Of the expeditions climbing Everest between 1980 and 2002, a total of 91 climbers and 38 porters (sherpas) died. During the spring season of 1996, the fatality rate reached 7.5% for the expeditions who climbed the South Col route (the common route and the one climbed by Hall’s and Fischer’s teams). Approximately one in 10 expeditions suffered the death of more than one team member. Everest is considered one of the most dangerous climbs in the world, second only to the mountain called K2 in difficulty (Huey et al., 2007). However, of the climbers attempting the summit Hall’s team suffered a death rate of closer to 30%, and Fischer’s team would have had a similar fatality rate if not for the skill of the guides in rescuing a number of stranded clients.

Defective decision making turned this hazardous summit attempt into a fatal one. The decision to climb far past the designated turn-around deadline provides a bright-line indicator of error. A feature story published six weeks prior to the expedition stressed Fischer’s phi- losophy about turn-around times: “Every climber has a set of personal guidelines that he or she follows, Little Stay Alive Rules. One of Fischer’s is the Two O’Clock Rule. If you are not on top by two, it is time to turn around. Darkness is not your friend.” As Krakauer writes, “Certainly time had as much to do with the tragedy as the weather, and ignoring the clock can’t be passed off as an act of God. Predetermined turn-around times were egregiously ignored” (1997, p. 273).

Climbing experts agree that the groups should not have pushed on beyond the turn-around time. Hillary, after hearing the details of the failed expeditions, con- cluded that tragedy was not an “accident,” for it was caused by the slow ascent to the summit (quoted in

Dowling, 1996). Ed Viesturs, an expert on high- altitude climbing, was present on the mountain in 1996. He watched in disbelief as the teams pushed for the summit, thinking, “Guys, you left at midnight. It’s two o’clock! It’s going to be three or four before you get to the summit.” As he watched the climbers continue upward he wondered, “Dudes, what are you doing? Wake up! Guys, turn around, turn around” (Viesturs, 1996, p. 1). However, Fischer’s and Hall’s expeditions failed to adhere to the strict two o’clock rule—a mis- take that turned deadly.

GROUPTHINK ANTECEDENTS AND SYMPTOMS

Although the decision to continue climbing was a faulty one, unless the groups exhibited the key antecedents and symptoms the teams could not have experienced groupthink, at least as originally defined by Janis (1972). The third component of arguing for a group- think analysis of a particular case is to illustrate that the key antecedent conditions and symptoms of groupthink were present. In the current study, we consider the mul- tidimensional nature of group cohesion (Dion, 2000) before elaborating on the critical role of the leaders, and provocative context in contributing to the consensus seeking and ultimately the poor decision to ignore the turn-around time.

Cohesion. The first and most critical catalyst of group- think is cohesion. Early work on cohesion suggested it was a culmination of factors, such as attraction to the members, prestige of the group, and collective pursuit of goals (Festinger, 1950). However, subsequent empiri- cal analyses, including Janis’s case studies, typically op- erationalized cohesion in terms of interpersonal attraction (Lott & Lott, 1965). During the 1950s through the early 1970s, the dominant perspective was that cohesion was a one-dimensional construct that had similar impacts on relevant outcomes despite potential subcomponents. After Cartwright (1968) and Hackman (1976) raised concerns about this unitary definition, stud- ies began to offer evidence for multidimensionality (e.g., Carless & De Paola, 2000; Cota, Evans, Dion, Kilik, & Longman, 1995). This history is important for our paper. Although Janis may not have directly artic- ulated that interpersonal attraction was the only source

34 JOURNAL OF LEADERSHIP STUDIES • Volume 4 • Number 4 • DOI:10.1002/jls

less experienced than the leader. On Everest commercial expeditions, the leadership style of the head guides is often more directive. As Krakauer (1997) noted, “For safety’s sake, a responsible guide . . . simply can’t afford to let each client make important decisions independ- ently” (p. 168). Rather, the leader guides the group. This philosophy is functional in theory and perhaps necessi- tated by the situation, but such a directive leadership style makes groups significantly more prone to group- think (Kramer, 1998).

The nature of directive leadership raises the issue of what constitutes a group decision. For example, is it that the leader obtains a vote, getting input from each and every member? In past groupthink analyses, this has not typically been the case. For example, the United States clearly did not vote to fail to carefully monitor Japanese warship movements in the Pacific prior to Pearl Harbor, yet this is a classic groupthink example. Is a group meeting required for a group decision? Again, for many of the famous groupthink groups, such as the Wa- tergate burglars, the members did not “meet” or directly discuss their series of choices to cover up the incident. In the current case-study analyses, we are not suggesting that at some point during the climb a leader directly stated to all members, “I’ve decided the expeditions should continue on beyond the turn-around time; does everyone agree?” The style employed on summit day makes such a direct collective decision impossible, as does the nature of directive leadership. We argue that such an explicit announcement of a group choice is not required for something to be a group decision. Rather, we propose that the climbers proceeded to the summit beyond the turn-around time under the direction of the leader, and this was precipitated by groupthink consen- sus seeking.

On Everest, neither the second guides nor the clients voiced their concerns about the groups’ decision to con- tinue the push toward the summit well beyond a safe turn-around time. Krakauer noted that Beidleman (a guide on Fischer’s team) acknowledged, after the ex- pedition, that “I tried not to be too pushy. As a conse- quence, I did not always speak up when maybe I should have” (Krakauer, 1997, p. 260). Experienced and capa- ble guides such as Beidleman censured themselves in large part due to the directive leadership and in part due, to the desire to reach the team goal of the summit.

of cohesion necessary for groupthink, Janis’s case stud- ies focused primarily on the interpersonal nature of co- hesion, as did subsequent empirical investigations of groupthink.

Our model, in contrast, draws on studies of perfor- mance groups that suggest cohesion is often based on commitment to a task, rather than to the group and its members. Studies of a variety of task-oriented groups, such as teams, military squads, and expeditions, find that when asked to describe their team’s cohesiveness members stress the quality of their group’s ability to per- form as a unit (Guzzo, 1995). Similarly, the expeditions on Everest exhibited high task cohesion as team mem- bers shared a common goal. Each member trained for months or years; endured extreme ailments such as painful coughing spurts, frigid temperatures, months away from family and friends; and paid as much as $65,000 to be part of the expedition. To succeed, each person needed to focus, put personal issues aside, and function as part of a unified group committed to reaching the summit. Emerson (1966) noted in his eval- uation of Everest and goal striving that, in considera- tion of the heavy toll and long commitment required for high-altitude expeditions, group goal-oriented mo- tivation had to be very powerful to overcome such a harsh setting. These groups come together with the sole purpose of reaching the summit. Unlike club-style ex- peditions among friends, commercial expeditions are groups that come together with the sole purpose of reaching the summit. These groups therefore typically lack interpersonal cohesion and loyalty to one another; rather, the groups on Everest share commitment to the collective goal of reaching the summit (Kayes, 2006). This extreme task cohesion can result in what moun- taineers often call “summit fever.” We suggest not only that the deterioration in group performance is due to an excessive desire to achieve the desired goals but also that this commitment, combined with directive leader- ship and other antecedents of groupthink, contributes to consensus seeking.

Directive Leadership. Directive leadership entails telling followers what needs to be done and giving appropriate guidance along the way (House & Mitchell, 1974). Directive leadership is most often used when the task is unstructured and complex and when the follower is

JOURNAL OF LEADERSHIP STUDIES • Volume 4 • Number 4 • DOI:10.1002/jls 35

Janis argued that when leadership is overly directive, group processes form the basis for and sustain consen- sus behaviors that can lead to groupthink. It was not that Beidleman or the clients lacked the experience to offer valuable feedback. Rather, groupthink triggered self-censorship, with the result that qualified guides failed to express concerns. Unintended as it may have been, a norm was created in which potential concerns about safety came second to protecting the goal of reaching the summit of Everest, which is stressed by the paying clients. We suggest that this norm developed out of the directive leadership style coupled with high task cohesion.

Provocative Situational Context. Cohesive groups cope well with routine problems, but the advantages of cohe- sion may be lost not only if a leader is overly directive but also if groups face a stressful environment (Driskell, Salas, & Johnston, 1999). The imminence of death, the intense amount of time and money invested in reaching the top, and the uncontrollable objective dangers (e.g., avalanches, icefalls, crevasses, poor weather) make stress an inevitable part of mountaineering expeditions. These stressors are constantly present in mountaineering, but in May 1996 the addition of journalists, an IMAX film crew, and two expeditions competing for the same clien- tele created additional anxieties. When Beck Weathers (a client on Hall’s team) was interviewed by ABC News about having a reporter on his expedition team, he replied, “It added a lot of stress. I was always a little con- cerned about the idea. . . . you know, this guy’s going to come back and write a story that’s going to be read by a couple of million people” (Krakauer, 1997, p. 174).

Another component of a provocative situational con- text is the role of recent failures. This antecedent condi- tion could be seen on Hall’s team, the one that endured the majority of casualties. The year before, while guid- ing on Everest, Hall had turned all his clients around just below the summit because the team had failed to reach the top before the turn-around time. Hansen, a client on Hall’s team, had failed on the past expedition to make the summit, and this year Hall was determined to help him reach the top. A year later, on the 1996 ex- pedition, Hansen made it to the summit with the sup- port of Hall, but not until 4:00 p.m. However, this time turned out to be too late for a safe descent.

In summary, our analysis of the events of the May 1996 Everest tragedy illustrates the presence of the key groupthink antecedents, including high task cohesion, directive leadership, and a provocative context. Thus far, we sought to elaborate on Janis’s original model by highlighting the multifaceted nature of group cohesion and the critical role of the leader. However, it is neces- sary not only to illustrate the antecedents but also to elaborate on the symptoms of groupthink. Janis (1982) noted that cohesiveness, leadership, and decisional stress may cause a group to experience groupthink, whereas the factors of overestimation of the group, closed- mindedness, and pressure toward uniformity are symp- toms suggesting a group has fallen prey to groupthink (Henningsen, Henningsen, Eden, & Cruz, 2006). We now assess the events of May 1996 and examine which symptoms of groupthink were present.

Groupthink Symptoms: Overestimation of the Group. Janis argued that overestimation of the group increases because of feeling “close” to the group members. In the current analysis, overestimation of the group’s ability is also a groupthink symptom, though one based on an exaggerated commitment to the group goal. The clients’ commitment grew as they watched their talented head guides lead them safely to high camp, through difficult logistical challenges and complex route finding. The clients and guides developed an illusion of invulnera- bility and thus failed to focus on another extremely dan- gerous aspect of the climb: late afternoon storms high on the mountain. The characteristics of optimism, the leader’s assurance of success, and the belief that some- thing never done before can be accomplished are strongly reminiscent of groupthink.

Pressure Toward Uniformity. In the traditional group- think model, uniformity characterizes groups experi- encing groupthink because members do not wish to offend one another and risk undermining the group’s bond. In the present framework, uniformity is high be- cause all the members shared the same goal. Each ex- pedition wanted to ensure a chance at the summit, and dissenting information would have hindered that goal. For example, Lene Gammelgaard on Fischer’s team re- marked, “Can’t help but admire Scott’s decision. This kind of gambling must be what’s gotten him to the summit so

36 JOURNAL OF LEADERSHIP STUDIES • Volume 4 • Number 4 • DOI:10.1002/jls

expeditions viewed one another with some degree of dis- trust, given that they were competitors seeking the same goal. The teams also tended to stereotype some of the other groups present on the mountain that season, among them the Taiwanese and South African expedi- tions. This stereotyping of outgroups is characteristic of typical groupthink-based decision making in which in- group members are viewed as “morally” right whereas outgroups are viewed as enemies or morally wrong. Krakauer (1996) described how Hall reacted after Ian Woodall, the South African expedition leader, declared that the South Africans would go for the summit when- ever it suited them. Hall exclaimed, “I don’t want to be anywhere near the upper mountain when those pun- ters are up there” (Krakauer, 1997, p. 142).

Discussion In summary, using reviews of personal accounts, media descriptions, online interviews, and past empirical pa- pers describing the May 1996 Everest disaster, we of- fered evidence for an adapted framework through which the groupthink phenomenon can be examined. These events on Everest, viewed in the context of groupthink, form a coherent pattern illustrating the applicability of task cohesion and directive leadership in explaining how groups can respond to a stressful and extreme environ- ment. As opposed to the original groupthink model pro- posed by Janis, which focused on interpersonal cohesion, we proposed that task cohesion can contribute to an un- derstanding of how such a skilled group of climbers with prominent leaders decided to ignore the turn-around time. We sought to use this provocative case analysis as a means of introducing our scholarly analysis of groupthink driven by high task cohesion and directive leadership. We recognize that groupthink is only one explanation in a multifaceted process. The weather pat- terns, communication and lack thereof with other ex- peditions, high altitude, and other less controllable factors are also present on mountaineering expeditions.

Our goal in this Everest case study analysis was not to lay blame on the expedition leaders. Rather, we sought to use this analysis to illustrate our theoretical proposi- tion, namely, that task cohesion may serve as a catalyst of groupthink decision making when coupled with di- rective leadership and a provocative context. Replace- ment of interpersonal cohesion with task cohesion as a

many times. I would have chosen a wider margin of safety and waited below for more stable conditions. But I want to summit and have no scruples. Apparently no- body else does either” (Gammelgaard, 1996, p. 160). Boukreev, one of the guides, admitted later that he chose to go along with the group, even though he had misgivings: “I tried not to be argumentative, choosing instead to downplay my intuitions” (Boukreev & DeWalt, 1997, p. 140). The illusion of unanimity stems in part from the silence of self-censorship and occurs when members share the illusory belief that they are unanimous in their judgments. Clients and guides self- censored information that could have been critical to the groups’ safety because of a vehement commitment to reach the summit and a desire not to impede prog- ress. These two expeditions failed to establish an envi- ronment where diversity of opinion and involvement of all team members was encouraged. Group members viewed dissenting opinions as potentially detrimental to accomplishing the ultimate goal. However, express- ing disagreement might have prevented a risky decision, one that turned fatal (see Hobman, Bordia, & Gallois, 2003, for a discussion of beneficial task conflict and value dissimilarity in groups).

Closed-Mindedness of the Group. Collective rational- ization and ingroup-outgroup biases were also evident. Both teams rationalized their choices by referring to the basic goal: to reach the summit. They became so com- mitted to their goal that they rationalized pushing on well beyond a safe turn-around time. In fact, Hall was so committed to reaching the summit that, prior to the May 1996 expedition, he practically guaranteed Hansen a chance to reach the top of Everest; Hall even offered him a discount on the guiding fee to return since Hansen came very close to reaching the summit the previous year (Krakauer, 1997). Although this may be a common prac- tice in keeping client satisfaction high in order to pro- mote referrals, Hall felt an added stress to help Hansen reach the goal of the summit.

Stereotypes of outgroups, another groupthink symp- tom reflective of closed-mindedness, are most promi- nent in political decisions regarding attacking or invading another nation. Among mountaineering groups on Everest, decisions mostly concerned intragroup pro- cesses. However, outgroups were still present. The two

JOURNAL OF LEADERSHIP STUDIES • Volume 4 • Number 4 • DOI:10.1002/jls 37

driving force in the current analysis parallels research in which limited support has been offered for the role of interpersonal cohesion in groupthink laboratory studies (for reviews, see Aldag & Fuller, 1993; Park, 1990, 2000; see also Esser, 1998; Neck & Moorhead, 1995). For example, Flowers (1977), Fodor and Smith (1982), and Callaway and Esser (1984) all found no detrimen- tal effects of interpersonal cohesion on group decision making. Indeed, some empirical studies have demon- strated that high interpersonal cohesion leads members to exhibit less self-censorship (Leana, 1985).

If experimental studies lack support for the importance of interpersonal cohesion, why do many case analyses (e.g., Cuban Missile Crisis, Challenger Space Shuttle Dis- aster, Columbia Space Shuttle Disaster, Gulf War II) fur- nish evidence that interpersonal cohesion plays a role in faulty decision making (Moorhead, Ference, & Neck, 1991)? In laboratory tests, the role of cohesion could be difficult to discern, owing to the lack of other neces- sary conditions of groupthink. It is challenging in the lab, with ad hoc groups, to create the status differenti- ation, time pressure, and stressful circumstances that surround real-world decision-making situations.

Additionally, we suggest that researchers must distin- guish between excessive task cohesion and interpersonal cohesion in examining groupthink decision making. As Bernthal and Insko (1993) noted, “A logical step in the groupthink research is to launch an effort to compare social-emotional and task-oriented cohesive groups in the groupthink model” (p. 68). Furthermore, in our analysis of the Everest case, we argue that a key com- ponent of the groupthink model that must be incorpo- rated into investigations is the role of directive leadership. Leaders who encourage open discussion can harness task cohesion for increased performance rather than stifling dialogue through directive styles.

LIMITATIONS

The selection of a single case study design naturally brings forth limitations. One of the main concerns of case study research is the potential lack of generalizabil- ity. Thus the setting, the groups, and the exact details of the spring 1996 climbing season on Mount Everest can only be seen as a pilot context of group decision- making processes. On the other hand, this also repre- sents a key goal of conducting a case study analysis. We

hoped that by understanding this particular case we might eventually also learn something that could be ap- plied to other groups and situations.

Another potential limitation of case study analysis is the lack of experimental control and the post hoc the- orizing. For example, it is less clear in our analysis which of the antecedents is the leading cause of defective decision making. However, case studies retain more of the “noise” of real life, and in studying groupthink this allows all the antecedents and symptoms to be more readily studied. For example, a provocative situation and directive leadership, although readily available for study in case analyses, can be difficult to introduce to ad hoc groups in lab settings. Future research could build on the current analysis and test the assumptions using more traditional, laboratory-based, a priori hypothesis testing to offer support for the proposed model.

Despite limitations, case studies are fruitful grounds for conceptual and theoretical development. To our knowledge, we are the first to propose a framework that replaces interpersonal cohesion with task cohesion to examine groups who come together with the sole pur- pose of reaching a shared goal other than securing a con- sensual decision. The case study analysis builds the foundation from which future research can examine the role of task cohesion, leadership, and extreme envi- ronments within the groupthink framework. For exam- ple, avenues exist for future laboratory and case study research to examine how the interplay between task and interpersonal cohesion can help to reduce poor decision making on mountaineering expeditions and in extreme contexts in general. Early work by Bernthal and Insko (1993) started to explore this question, but more work is needed that incorporates leadership theory. For exam- ple, on a larger scale researchers could experimentally manipulate task versus interpersonal cohesion, directive versus open leadership, norms versus no procedures for decision making, or constraints versus unlimited time to establish what combination of conditions most com- monly leads to risky decisions. This research could in turn help leaders make more informed decisions.

APPLICATIONS AND IMPLICATIONS

The model proposed in the current paper can increase the applicability of groupthink to real-world decisions made by groups in pursuit of a common goal. Considering the

38 JOURNAL OF LEADERSHIP STUDIES • Volume 4 • Number 4 • DOI:10.1002/jls

form of groupthink. Additionally, these groups will be prone to use directive leadership styles to try to achieve efficiency and clear delineation of roles. As illustrated in the current analysis, this combination can result in risky decision making.

What, then, can leaders do to reduce faulty decision making? Recognizing when groupthink antecedents are present, leaders could consider adopting leadership styles that foster more input and membership participation. For example, recent research highlights shared leader- ship as an efficient and effective method of facilitating group processes. Shared leadership, an emergent group dynamic in which leadership is intentionally distributed across all team members and not solely held by one des- ignated person, fosters a climate in which members share information openly and benefit from the reciprocal in- fluence of peers (Carson, Tesluk, & Marrone, 2007). As research related to the benefits of shared leadership con- tinues to emerge, a shift in leadership toward a more in- clusive style may help alleviate the pitfalls of high task cohesion in the context of groupthink.

Summary In the current paper, we argued that the shared pursuit of a common goal coupled with directive leadership and a provocative situation can lead groups composed of well-functioning team members and skilled leaders to take excessive risks. Although additional empirical re- search is still needed to investigate a task-cohesion- focused groupthink framework, the present research of- fers initial evidence of its applicability and can enhance our understanding of group decision making. Leaders who are aware of the potentially deleterious effects of overly high task cohesion may foster vigilant decision making and a more optimal decision-making process (e.g., Janis & Mann, 1977; McCall, Trombetta, & Nattrass, 2002). However, prior to putting workplace implementations into practice, more systematic empir- ical research is needed. We hope our case study analysis of the Everest disaster fosters such future explorations.

References Abbasi, S. M., & Hollman, K. W. (1991). Dissent: An important but neglected factor in decision making. Management Decision, 29, 7–11.

growing use of teams in organizational functioning, the study of decision making from a task cohesion-focused perspective may have additional applications. For example, we suggest that the proposed framework can be useful in examining other climbing disasters, such as on Mount McKinley in 1967 when eight of twelve expedition mem- bers died attempting to summit. The tragedy has been proposed by mountaineering pundits as one of the most mysterious and controversial disasters in the history of mountaineering (Tabor, 2007). Tabor’s work pieces to- gether personal interviews, unpublished correspondence, diaries, and government documents. Future research could use this existing documentation of the event to conduct an analysis of the applicability of a task-cohesion-focused groupthink framework.

In addition to extreme environments such as moun- taineering, organizational groups whose main purpose is to reach a common goal could also be examined through the lens of a groupthink model focused on task cohesion. For example, subsequent accounts of the NASA Space Shuttle Challenger disaster illustrate the tragic consequences resulting from a myopic focus on accomplishing a task (e.g., Moorhead et al., 1991). Another potential case is the build-up to the year 2000 information technology changeover. In this context, many groups formed with limited scope in response to what the media termed the Y2K issue (Schiano & Weiss, 2006). In these groups, the solitary goal was to limit the downtime and damage done to information technology systems as the new millennium approached. However, the narrow focus on the collective goal, in- fluenced by factors such as time pressures and lack of norms for decision making procedures, led to poor de- cision making (Schiano & Weiss).

In today’s rapidly changing business environment, it is often more the rule than the exception that ad hoc groups are convened to address a solitary issue and then are redistributed to the next task (e.g., DiMaggio, 2001). These groups may prove especially fruitful for groupthink model applications. Ad hoc groups are un- likely to exhibit high interpersonal cohesion; rather, they are gathered for the sole purpose of completing a task, and thus their collective commitment to the goal is high. These groups may be more likely to suffer from the detrimental effects of overly high task cohesion, in- troduced in the current paper, than from Janis’s original

JOURNAL OF LEADERSHIP STUDIES • Volume 4 • Number 4 • DOI:10.1002/jls 39

Aldag, R. J., & Fuller, S. R. (1993). Beyond fiasco: A reappraisal of the groupthink phenomenon and a new model of group decision processes. Psychological Bulletin, 113, 533–552.

Baron, R. S. (2005). So right it’s wrong: Groupthink and the ubiq- uitous nature of polarized group decision making. Advances in Ex- perimental Social Psychology, 37, 219–253.

Bernthal, P. R., & Insko, C. A. (1993). Cohesiveness without group- think: The interactive effects of social and task cohesion. Group & Organization Management, 181, 66–87.

Boukreev, A., & DeWalt, G. W. (1997). The climb: Tragic ambi- tions on Everest. New York, NY: St. Martin’s Press.

Callaway, M. R., & Esser, J. K. (1984). Groupthink: Effects of co- hesiveness and problem-solving procedures on group decision mak- ing. Journal of Social Behavior and Personality, 12, 157–164.

Carless, S. A., & De Paola, C. (2000). The measurement of cohesion in work teams. Small Group Research, 31, 71–88.

Carson, J. B., Tesluk, P. E., & Marrone, J. A. (2007). Shared lead- ership in teams: An investigation of antecedent conditions and per- formance. Academy of Management Journal, 50, 1217–1234.

Cartwright, D. E. (1968). The nature of group cohesiveness. In D. Cartwright and A. Zander (Eds.), Group dynamics (pp. 91–109). New York, NY: Free Press.

Cota, A. A., Evans, C. R., Dion, K. L., Kilik, L., & Longman, R. S. (1995). The structure of group cohesion. Personality and Social Psy- chology Bulletin, 21, 572–580.

DiMaggio, P. (2001). Introduction: Making sense of the contem- porary firm. In P. DiMaggio (Ed.), The twenty-first century firm (pp. 3–13). Princeton, NJ: Princeton University Press.

Dion, K. L. (2000). Group cohesion: From “field of forces” to mul- tidimensional construct. Group Dynamics: Theory, Research, and Practice, 4, 7–26.

Dowling, C. G. (1996). Death on the mountain. Life, August, 32–46.

Driskell, J. E., Salas, E., & Johnston, J. H. (1999). Does stress lead to a loss of team perspective? Group Dynamics: Theory, Research, and Practice, 3, 291–302.

Elmes, M., & Berry, D. (1999). Deliverance, denial, and the death zone: A study of narcissism and regression in the May 1996 Everest climbing disaster. Journal of Applied Behavioral Science, 35, 163–187.

Elmes, M., & Frame, B. (2008). Into hot air: A critical perspective on Everest. Human Relations, 61, 213–241.

Emerson, R. M. (1966). Mount Everest: A case study of communi- cation feedback and sustained group goal-striving. Sociometry, 29, 213–227.

Esser, J. K. (1998). Alive and well after 25 years: A review of group- think research. Organizational Behavior and Human Decision Pro- cesses, 73, 116–141.

Festinger, L. (1950). Informal social communication. Psychological Review, 57, 271–282.

Flippen, A. R. (1999). Understanding Groupthink from a self- regulatory perspective. Small Group Research, 30, 139–165.

Flowers, M. L. (1977). A laboratory test of some implications of Janis’s groupthink hypothesis. Journal of Personality and Social Psy- chology, 35, 888–896.

Fodor, E. M., & Smith, T. (1982). The power motive as an influence on group decision making. Journal of Personality and Social Psychol- ogy, 42, 178–185.

Gammelgaard, L. (1996). Climbing High. New York, NY: Perennial Press.

Graydon, D., & Hanson, K. (Eds.). (1997). Mountaineering: Free- dom of the hills. Seattle, WA: Mountaineers Press.

Guzzo, R. A. (1995). Introduction: At the intersection of team ef- fectiveness and decision making. In R. A. Guzzo & E. Salas (Eds.), Team effectiveness and decision making in organizations (pp. 1–8). San Francisco, CA: Jossey-Bass.

Hackman, J. R. (1976). Group influences on individuals. In M. Dunnette (Ed.), Handbook of Industrial and Organizational Psychology (pp. 1455–1525). Chicago, IL: Rand McNally.

Henningsen, D. D., Henningsen, M.L.M., Eden, J., & Cruz, M. G. (2006). Examining the symptoms of Groupthink and retroactive sensemaking. Small Group Research, 37, 36–64.

Hobman, E. V., Bordia, P., & Gallois, C. (2003). Consequences of feeling dissimilar from others in a work team. Journal of Business and Psychology, 17, 301–325.

House, R. J., & Mitchell, T. R. (1974). Path-goal theory of leader- ship. Contemporary Business, 381–398.

Huey, R. B., Salisbury, R., Wang, J. L., & Mao, M. (2007). Effects of age and gender on success and death of mountaineers on Mount Everest. Biological Letters of the Royal Society, 498–500.

Janis, I. (1972). Victims of groupthink. Boston, MA: Houghton-Mifflin.

Janis, I. (1982). Groupthink (2nd Ed.). Boston, MA: Houghton- Mifflin.

Janis, I., & Mann, L. (1977). Decision making: A psychological anal- ysis of conflict, choice, and commitment. New York, NY: Free Press.

Kayes, D. C. (2004). The 1996 Mt. Everest climbing disaster: The breakdown of learning in teams. Human Relations, 57, 1236–1284.

40 JOURNAL OF LEADERSHIP STUDIES • Volume 4 • Number 4 • DOI:10.1002/jls

Park, W. W. (2000). A comprehensive empirical investigation of the relationships among variables of the groupthink model. Journal of Organizational Behavior, 21, 873–887.

Roberto, M. A. (2002). Lessons from Everest: The interaction of cognitive bias, psychological safety, and system complexity. Califor- nia Management Review.

Raven, B. H. (1998). Groupthink, Bay of Pigs, and Watergate recon- sidered. Organizational Behavior and Human Decision Processes, 73, 352–362.

Schiano, W., & Weiss, J. W. (2006). Y2K all over again: How group- think permeates IS and compromises security. Business Horizons, 49, 115–125.

Street, M. D., & Anthony, W. P. (1997). A conceptual framework establishing the relationship between groupthink and escalating commitment behavior. Small Group Research, 28, 267–293.

Tabor, J. M. (2007). Forever on the mountain: The truth behind one of the most tragic, mysterious, and controversial disasters in moun- taineering history. New York, NY: Norton.

Viesturs, E. (1996). Ed Viesturs on 1996: Turn around guys. Na- tional Geographic Adventure. Retrieved from http://www.national geographic.com/adventure/everest/ed-viesturs-6.html.

Weathers, B., & Michaud, S. (2001). Left for dead: My journey home from Everest. New York, NY: Dell.

Whyte, G. (1998). Recasting Janis’ groupthink model: The key role of collective efficacy in decision fiascoes. Organizational Behavior and Human Decision Processes, 73, 185–209.

Jeni L. Burnette is an assistant professor in the Department of Psychology at the University of Richmond. She received her Ph.D., in social psychology, from Virginia Commonwealth University in 2006. She can be reached at jburnet2@ richmond.edu.

Jeffrey M. Pollack is an assistant professor of management in the Robins School of Business at the University of Rich- mond. He holds a Ph.D. from Virginia Commonwealth University. He can be reached at [email protected].

Donelson R. Forsyth is a professor in the Jepson School of Leadership Studies at the University of Richmond where he holds the Thorsness Chair in Ethical Leadership. He received his Ph.D., in social psychology, from the University of Florida in 1978. He can be reached at [email protected].

Kayes, D. C. (2006). Destructive goal pursuit: The Mount Everest dis- aster. Hampshire: Palgrave-Macmillan Press.

Kowert, P. A. (2002). Groupthink or deadlock: When do leaders learn from their advisors? New York: State University of New York Press.

Krakauer, J. (1997). Into Thin Air. New York, NY: Villard.

Kramer, R. M. (1998). Revisiting the Bay of Pigs and Vietnam de- cisions 25 years later: How well has the groupthink hypothesis stood the test of time? Organizational Behavior and Human Decision Pro- cesses, 73, 236–272.

Kruglanski, A. W., Pierro, A., Mannetti, L., & De Grada, E. (2006). Groups as epistemic providers: Need for closure and the unfolding group-centrism. Psychological Review, 113, 84–100.

Leana, C. R. (1985). A partial test of Janis’ groupthink model: Ef- fects of group cohesiveness and leader behavior on defective decision making. Journal of Management, 11, 5–18.

LeBlanc, R., & Gillies, J. (2005). Inside the Boardroom. Mississauga, Ontario: John Wiley & Sons, Canada.

Lott, A. J., & Lott, B. E. (1965). Group cohesiveness as interper- sonal attraction: A review of relationships with antecedent and con- sequent variables. Psychological Bulletin, 64, 259–309.

Mangione, L., & Nelson, D. (2003). The 1996 Mount Everest tragedy: Contemplation on group process and group dynamics. In- ternational Journal of Group Psychotherapy, 53, 553–573.

McCall, M., Trombetta, J., & Nattrass, K. (2002). Limiting under- age alcohol purchases: An application of the consumer decision model. Journal of Business and Psychology, 17, 19–29.

Moorhead, G., Ference, R. J., & Neck, C. P. (1991). Group decision fiascoes continue: Space shuttle Challenger and a revised groupthink framework. Human Relations, 44, 539.

Moorhead, G., Neck, C. P., & West, M. S. (1998). The tendency to- ward defective decision making within self-managing teams: The relevance of groupthink for the 21st century. Organizational Behav- ior and Human Decision Processes, 73, 327–351.

Neck, C. P., & Moorhead, G. (1995). Groupthink remodeled: The importance of leadership, time pressure, and methodical decision- making procedures. Human Relations, 48, 537–557.

Nutt, P. C. (2002). Why decisions fail: Avoiding the blunders and traps that lead to debacles. San Francisco, CA: Berrett-Koehler.

Park, W. W. (1990). A review of research on Groupthink. Journal of Behavioral Decision Making, 3, 229–245.

Copyright of Journal of Leadership Studies is the property of John Wiley & Sons, Inc. and its content may not

be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written

permission. However, users may print, download, or email articles for individual use.