Academy of Management Learning & Education Developing

Developing Leaders forDecision Making Under Stress:

Wildland Firefighters in theSouth Canyon Fire and Its

AftermathMICHAEL USEEM

Wharton School, University of Pennsylvania

JAMES COOKU.S. Forest Service and National Interagency Fire Center

LARRY SUTTONU.S. Bureau of Land Management and National Interagency Fire Center

To identify the sources of ineffective leadership decisions, we focus on ten decisionsmade by a leader of a wildland firefighter crew during the fatal South Canyon fire ofJuly 5–6, 1996. The decisions of team leaders in fire zones are unusually clear-cut andconsequential for the goals of the enterprise, but they are not unlike decisions faced bymanagers of most organizations. We suggest that three factors—underpreparation, acutestress, and ambiguous authority—can result in suboptimal decisions by team leaders ona fireline. Through detailed evaluation of the team leader’s ten most consequentialdecisions in the South Canyon fire, we conclude that five were relatively optimal for thetriple objectives of safety, speed, and suppression, but five others proved suboptimal.Much of their suboptimality is traced to the fact that the team leader was undertrainedfor leadership decision making, faced intense stress, and operated without clearauthority. In the wake of this firefighting disaster—14 men and women lost their lives—the fire service created a development program using both classroom and experientialmethods for preparing its leaders to make good and timely decisions. The South Canyonfire and its aftermath point to the value of explicit preparation in leadership decisions byboth fire services and business schools as part of their efforts to enhance strategicthinking and other essential leadership attributes for achieving organizational goals inhigh stress environments.

........................................................................................................................................................................

Wagner Dodge led his crew of 16 young men tocombat the forest fire in Mann Gulch, Montana onAugust 5, 1949. All but one had parachuted into the

canyon at 4 p.m. for what looked to be a routinemission. But a series of leadership choices resultedin their entrapment, and by 5:56 p.m. all but threehad been fatally burned.

Donald Mackey helped oversee 49 firefightersspread out to combat a fire on Storm King Moun-We are grateful to editor James Bailey and two anonymous

reviewers for their very helpful suggestions and guidance dur-ing the review process. We are thankful for the useful commen-tary and insights of those who joined us during a day-long walkof the South Canyon fire zone on Storm King Mountain on May29, 2002, including affiliates of several wildland firefightingagencies: Kim Bang, Grant Beebe, Tim Blake, Sarah Doehring,Pam Ensley, Deb Epps, Anthony Escobar, Jim Glenn, JimKitchen, Mark Linane, Bob Leighty, Nancy Lull, Greg Power,George Steele, and Steve Thomas; officers of the U.S. Marine

Corps: Bob Baird, Eric Carlson, and Cheston Souza; author JohnMaclean; and affiliates or graduates of the Wharton School:Mark Davidson, Neil Doherty, Bruce Newsome, and BarbaraShannon. We are deeply indebted to Donald Mackey, whosedecisions during the South Canyon fire have profoundly in-formed our thinking and conclusions.

� Academy of Management Learning & Education, 2005, Vol. 4, No. 4, 461–485.

........................................................................................................................................................................

461

tain, Colorado on July 6, 1994. Some of the 49 hadparachuted onto the mountain that day; others hadcome by helicopter, still others by foot. A chain ofleadership decisions also resulted in their entrap-ment, and by 4 p.m. it appeared that the MannGulch disaster was about to repeat itself.

Bad luck and a fatal confluence of environmen-tal factors contributed to the flaming ambush ofthe firefighters in both Mann Gulch and on StormKing Mountain, but leadership decisions were alsocritical in both. Those most directly responsibleon-site had faced a sequence of decision pointsduring their fateful hours in the fire zone, and theirdecisions at those points helped bring the teams tothe brink of disaster.

Those most directly responsible on-sitehad faced a sequence of decision pointsduring their fateful hours in the fire zone,and their decisions at those pointshelped bring the teams to the brink ofdisaster.

The human consequences of suboptimal deci-sions by fire leaders are compellingly clear, and,conversely, optimal leadership decisions are noless vital for successfully suppressing a fire. Thesame is evident in company management: Faultyleadership decisions helped bring down suchfirms as Enron and WorldCom, and quality deci-sions have been vital for the prosperity for compa-nies from eBay to Southwest Airlines.

To help identify what makes for good leadershipdecisions, we have chosen to focus on wildlandfirefighters, those who combat forest, brush, andgrass fires in the backcountry. The decisions ofwildland fire crew leaders are unusually clear-cutand consequential for the goals of the enterprise—but at root they are not unlike decisions faced bymanagers of most organizations. Wildland fire-fighting offers a particularly effective vehicle forunderstanding the predicates of good leadershipdecision making whatever the venue.

The specific focus of our analysis is the set ofleadership decisions among the firefighters onStorm King Mountain on July 5 and 6, 1994 in whathas come to be known as the South Canyon fire.Our method is to examine the events of the fire insufficient detail to pinpoint the critical decisionsand then extrapolate their implications for leader-ship decisions in other settings and institutions,including company management. Our purpose isnot only to identify the factors that facilitate orundermine optimal leadership decisions, but also

to help specify the developmental steps that orga-nizations can take to help their leaders reach bet-ter decisions on behalf of the enterprise. An under-lying purpose is to draw greater attention todecision making by those in leadership positions.

LEADERSHIP DECISIONS

Making decisions is what most managers do muchof their day. Hundreds are minor, but some aresignificant: commencing a project or hiring a staff.A few are momentous: blowing a whistle or redi-recting a career—or even launching an enterpriseor saving a firm. Made well, good decisions be-come the foundation of personal advancement;made poorly, they can end an otherwise promisingcareer.

Made well, good decisions become thefoundation of personal advancement;made poorly, they can end an otherwisepromising career.

Apart from personal consequences, decision mak-ing is at the heart of virtually all managementwork and a key driver of organizational outcomes(Barnard, 1968; Donaldson, 1983; Finkelstein &Hambrick, 1996).

Decisions take on special significance whenmade by those in leadership positions becausethey impact the fate of many others and possiblyeven the enterprise itself. We define leadershipdecisions to be those moments when an individualwith organizational responsibility faces a discrete,tangible, and realistic opportunity to commit en-terprise resources to one course or another on be-half of the enterprise’s objectives. Making nochoice in the face of such an opportunity—whetherconsciously recognized or not—is also deemed tobe a decision. Few leadership decisions are takenwithout the input and influence of others, and infocusing on the decisions of an individual, we donot mean to imply that they are solely responsiblefor the decisions, but rather that they take a prin-cipal role in making the decisions.

Although the varied options in a decision mayall be plausible, we focus on the fact that some aremore optimal than others for achieving the enter-prise’s goals. This analytic focus is akin to thepragmatic focus of many firefighters. Good deci-sion making, offers one experienced urban fire-fighter, “is about taking action now in order to giveyourself and your organization a shot at the bestpossible future” (Salka, 2004: 105). Our focus is also

462 DecemberAcademy of Management Learning & Education

similar to what some analysts have stressed indecision making. In The Functions of the Executive,for instance, Chester Barnard refers to organiza-tional decisions as a “choice of means to accom-plish ends which are not personal” (1968: 186). Sim-ilarly, “top management teams make strategicdecisions,” offered another observer, and their“quality” had direct bearing on “organizationalperformance” (Amason, 1996: 123). Our concern ison why some leadership decisions are suboptimalfor achieving the “best possible future” and “endswhich are not personal,” and how firms can betterprepare leaders to make decisions for “organiza-tional performance.”

Making quality leadership decisions on behalfof the organization has been well recognized byacademic investigators as a defining aspect ofleadership. Yukl (1989), for instance, identifiedgood decision making as one of the key compo-nents of leadership. Yet detailed study of leadersin the act of making decisions has often provenchallenging. Organizational leaders communicatein public but decide in private, making that aspectof their leadership more difficult to study. Mostwork of senior management is well shrouded, andas a consequence, in the words of one inside ob-server who knew it well, “most executive decisionsproduce no direct evidence of themselves” (Bar-nard, 1968: 193). This is one reason that research oncorporate governance has focused more often onthe composition and policies of boards of directorsthan the decisions made by directors inside theboardroom. Since “most of the work done by aboard takes place in the privacy of the boardroom,”noted two governance observers, “rarely, if ever”does information “escape to the outside world,”and most such decisions are thus largely invisibleto the outside world (Carter & Lorsch, 2004: 163).The actual decisions of those in leadership posi-tions should nonetheless be subject to direct exam-ination when feasible if we are to understand whatoften makes the difference for an outcome in a firezone, executive office, or boardroom.

EXAMINING LEADERSHIP DECISIONS

Underpinning our analysis is the premise that dis-tinct leadership decisions can be isolated andevaluated against the firm’s goals. The task willnot be easy not only because such decision makingis typically shrouded in secrecy, but also becausemanagers sometimes report that there was no sin-gle decision moment behind their company’s com-mitment to launch a product or enter a market.Researchers similarly report that some decisionsevolved out of numerous discussions, multiple

players, and unanticipated events, with few sharp-edged decision moments (Langley, Mintzberg,Pitcher, Posada, & Saint-Macary, 1995). And forsome organizations, their goals are so ambiguousthat the quality of decisions on their behalf is sim-ply beyond evaluation.

While such descriptions are no doubt correct forsome decisions in some organizations, they are notapplicable to all decision makers, especially inleadership positions. It is there that the decisioncontext, enterprise goals, and conflicting demandsare most likely recognized, analyzed, and incorpo-rated into a relatively specific decision to take thefirm down one path or another. It is there that thecalling for reasoned and judicious considerationsof the alternatives is most paramount, and it isthere that authority most clearly resides for select-ing one path over another. Although enormous po-litical and historical forces shaped the U.S. plan toattack the Afghan government in 2002 and Iraqgovernment in 2003, for instance, detailed insiderdescriptions of how those policies were chosenpoint unequivocally to specific decisions by thepresident (Woodward, 2003, 2004).

For these reasons we have chosen to focus onleadership decisions in a setting akin to the WhiteHouse, where they are relatively explicit. Whileappreciating that the broader context is importantfor fully understanding what transpired on in theSouth Canyon fire, our aim here is not to whollyexplain the event (and others have sought to do so:see Maclean, 1999; Butler et al., 1998). It is rather tounderstand one critical contributing element—thedecisions of the team leader—and why some deci-sions fell short of achieving the team’s objectives.

After appraising a diverse range of thinking andresearch on organizational decision making, Lan-gley and her colleagues urged that to “appreciatethe rich relationship between commitment and ac-tion, or to detect the roles of insight, inspiration,and emotion [in decisions] will require researchersto zoom in closer to the people and processes understudy” (1995: 276). Our agenda is to appreciate therichness by zooming in on the key figure during theSouth Canyon fire.

LEADERSHIP DECISIONS AND LEADERSHIPATTRIBUTES

The underfocus on leadership decisions may haveresulted in a partial misspecification of the behav-ioral foundation of leadership. Many observershave concluded that organizational leadershipcalls for at least four major attributes: strategicthinking about the organization’s environment,mobilization of its resources to achieve its strat-

2005 463Useem, Cook, and Sutton

egy, execution of the strategy, and selflessness(see, e.g., Bennis, 2003; Bossidy & Charan, 2002;Collins, 2001; Gardner, 1993; George, 2003; Tichy,1997; Useem, 1998).

Yet a leader’s personal reputation for these qual-ities may largely depend on the quality of theunderlying decisions he or she has previouslytaken. Thinking strategically depends on makinggood decisions to acquire and analyze informationon the environment; mobilizing resources requirescorrect decisions in assigning people and capitalto the task; executing effectively is contingentupon timely and appropriate decisions to imple-ment a plan; and selflessness depends upon sub-ordinating self-interest to collective purpose whenthose objectives diverge in a decision (see Table 1).

If the leader’s decisions have been made well,they come to constitute what we often attribute tothe leader as strategic thinking, resource mobili-zation, effective execution, and personal selfless-ness. If taken poorly, by contrast, the leader’s de-cisions diminish these same four definingattributes of leadership.

Greater attention to leadership decisions shouldprovide greater insight into how four leading at-tributes of leadership are strengthened or weak-ened. Analysis of the critical leadership decisionsin the South Canyon fire is thus also intended tocontribute to our understanding of how managersenhance or diminish their leadership attributes.

IMPROVING LEADERSHIP DECISIONS

All organizations have an interest in ensuring thattheir managers make optimal decisions, espe-cially when they occupy leadership roles, but ef-fective decision making is not a natural capacity.Managers tend to fall recurrently into what Russoand Schoemaker (1990) have termed “decisiontraps,” and a host of reasons for such suboptimaldecisions has been identified by a range of inves-tigators (Chiles, 2001; Hammond, 2000; Kahneman,2003; Klein, 1998, 2003; Russo & Schoemaker, 2002).

Despite the recurrent suboptimality, most re-searchers are optimistic about the capacity of or-ganizations to enhance managers’ decision mak-ing. Drawing upon a host of studies, for instance,Bazerman (2002: 152) has concluded that “we allhave plenty of room to improve our judgment,” andself-conscious learning how to make better deci-sions is a proven avenue for enhancement.

Other investigators as well have concluded thatleadership decisions can be improved. In the pri-vate sector, for example, many have pointed to theimportance of properly designed pay and promo-tion incentives for aligning executive decisionswith investor objectives. Well-designed incentivecompensation and promotion systems are viewedas optimizing the likelihood that company leaderswill consistently take decisions that benefit share-holders, not just themselves (e.g., McKenzie & Lee,1998).

Since decision making of those in leadershippositions has been relatively understudied, how-ever, the organizational roadmap for improvingleadership decisions has also remained relativelyunderdeveloped. Research studies indirectlytouching on this terrain suggest that such a road-map should make a difference. Druskat andWheeler (2003), for example, found that successfulwork teams were overseen more often than othersby managers who were particularly good at ac-quiring critical information and building good re-lations between the team and outside constituen-cies. Put differently, when overseeing leadersdecided to inform and empower their work teams,the teams performed more effectively, and suchcapacities should be subject to improvementthrough training and development.

Although the improvement of decision makingby those in leadership positions is a desirableorganizational objective in itself, it should alsoenhance the major attributes that are taken to bedefining of effective organizational leadership.One way to ensure quality leadership at the top ofan organization—summed up in the four recog-nized attributes of strategic awareness, resourcemobilization, effective execution, and personalselflessness—is to assure that its leaders make theeffective choices that give rise to such attributes.

THE SOUTH CANYON FIRE AS INSTRUCTION

Our method for helping to draw the roadmap foreffective leadership decision making is to examinethe events of the South Canyon fire in sufficientdetail to pinpoint the critical decisions and thenextrapolate their implications for improving lead-ership decision making in other institutions, in-

TABLE 1Leadership Attributes and Underlying Decision

Quality

LeadershipAttribute

Underlying DecisionQuality

Strategic thinker Acquire and analyze data on environmentAble to mobilize Assign people and capitalAble to execute Timely and appropriate implementationPersonal

selflessnessSubordinate private to collective purpose


cluding company management. Weick (1993, 1995)used a similar terrain and method to establish theimportance of what he termed organizational “sen-semaking”—the effort by people to make theirworld understandable and accountable. The col-lapse of sensemaking during the 1949 blow-up inMontana proved to be a critical factor, Weick dis-covered, in the fatal consequences that followed.“The Mann Gulch disaster,” he found (1993: 649),“can be understood as a dramatic failure of lead-ership,” and as we turn to the South Canyon firedisaster, it too can be understood as a failure ofleadership. In this instance, we trace the leader-ship failure not to the collapse of organizationalsensemaking but to suboptimal leadership deci-sion making. Our study may be viewed as comple-mentary to Weick’s focus on sensemaking: Bothsensemaking and decision making can be power-ful determinants of how leaders react to adversityand capitalize on opportunity.

We have chosen the South Canyon fire becauseit has been the subject of extensive official studyand secondary analysis. Like Civil War battlesand Space Shuttle disasters, we have the benefit ofan exceptionally well-documented record on thedecisions taken by those responsible for the event(for Civil War decisions, see Sears, 2003; for SpaceShuttle decisions, see Columbia Accident Investi-gation Board, 2003).

In addition to an official investigation on the fire(South Canyon Fire Accident Investigation Team,1994; Butler et al., 1998), we drew upon two othersources of information in preparing this analysis.The first is direct personal inspection of the SouthCanyon fire zone in June, 2002. We conducted thevisit as a walking seminar in the company of 17wildland firefighters (including one of the survi-vors, Sarah Doehring). The walking seminar pro-vided extended opportunities to discuss the lead-ership decisions with experienced firefighters whothemselves had taken thousands of decisionswhile leading fire teams of their own. The secondsource of information is the professional experi-ence of the second and third authors, who haveserved as wildland firefighters for more than 2decades and have taken leadership roles in dozensof fire zones. For additional perspective on leader-ship decisions in wildland fires, we also drewupon secondary analysis of the Mann Gulch andSouth Canyon fires conducted by Maclean (1999),Maclean (1990), Putnam (1995), Roberto and Ferlins(2003a, 2003b), Weick (1993, 1996) and others, includ-ing this article’s first author (Useem, 1998).

Our objective is neither to criticize nor commendthe decisions taken on the mountain, but rather tostudy them for what can be learned from them. Our

language may appear critical or complementary attimes, but it is intended to reveal enduring aspectsof effective leadership decision making ratherthan affixing blame or praise. To wildland fire-fighters, those who fell in the line of duty are he-roes: They placed themselves in harm’s way toprotect others and paid the ultimate price. Yet fire-fighters also feel it is their duty to unflinchinglyexamine past tragedies to determine what wentwrong in order to prevent similar calamities in thefuture, and since leadership decisions were a crit-ical factor, attention must be directed at them.

We begin with an appraisal of the distinctiveobjectives of wildland firefighting that are thefoundation for evaluating the decision making bythose who lead firefighters. We identify three fac-tors that are likely to result in suboptimal leader-ship decisions in firefighting. With these objec-tives and factors in mind, we then focus on oneleader’s decisions during the South Canyon fire,examining his choices in some detail as they bearupon the concepts in question. We conclude withan identification of the steps taken by U.S. agen-cies in the aftermath of the South Canyon fire toimprove decision making nationally among thosewho lead others into fire zones.

SAFETY, SPEED, AND SUPPRESSION: THE TRIPLEGOALS OF LEADERSHIP DECISIONS INFIREFIGHTING

In attacking fires in the wilderness, firefightersform into crews ranging from 3 to 20 members.Firefighting crews are rapidly deployed, com-bine with other crews to combat larger fires, andthen just as quickly break up and redeploy toother incidents. Crew leaders operate collabora-tively but also independently. During multiple-crew blazes, as was the case in the South Can-yon fire, one individual takes overallresponsibility as the incident commander. Theleadership responsibility is total: “On any inci-dent, large or small,” states one of the fire ser-vice manuals, “the Incident Commander has ul-timate responsibility for the effective and safeexecution” of all aspects of the attack (IncidentCommand System, 1994: 3– 4; National WildfireCoordinating Group, 2004; Maclean, 2003; Pyne,1997; Thoele, 1995). That responsibility places apremium on ensuring that leadership decisionsoptimally contribute to the three primary goals offirefighting: safety, speed, and suppression.

The premier criterion for decision making by firecrew leaders and incident commanders is thesafety of their team. Like friendly fire on the bat-tlefield, the possibility of loss is recognized but


must always be prevented. Fatal injuries are nomore tolerable in firefighting than is fraudulentaccounting in business or bogus stories in journal-ism. Since risk is always present, however, wild-land fire leaders must be able to appraise it andtake appropriate steps to mitigate it. Despite thepremise that no loss is acceptable, crew leadersand incident commanders frequently direct theirteams into venues where peril looms large.

A second criterion for decision making by crewleaders and incident commanders is speed. In fire-fighting, the clock is one of the main enemies.Hesitation or equivocation may do more than delaya solution: They can radically compound the prob-lem. In product markets, short-term can be months;in stock markets, days; in fire zones, hours. A nas-cent 10-acre fire, small matter in the wildland play-book, can explode in minutes—if not quickly sup-pressed—into a 1,000-acre conflagration. Notsurprisingly, the official firefighters’ manual ex-horts firefighters to “[m]ake sound and timely de-cisions” (Incident Operations Standards WorkingTeam, 2002: iv).

A final criterion for decision making for crewleaders and incident commanders is the technicaldecisions required to actually suppress the fire:How many firefighters are required, where shoulda fireline be constructed, what aerial reconnais-sance is needed? Good and opportune resolutionof such questions is essential for success on anyfireline. All three decision criteria are encoded inthe primary rule for successful engagement ofwildland firefighters that speaks to both safetyand speed: “Fight Fire Aggressively but Provide forSafety First” (Incident Operations Standards Work-ing Team, 2002).

UNDERPREPARATION IN LEADERSHIPDECISIONS

Wildland firefighters often assume leadershiproles with little warning, and the venues for theexercise of their leadership are always new. Suchassignments are analogous to the appointment offreshly commissioned officers to command sol-diers in combat, or of seasoned officers to taketroops onto an unknown battlefield. Unlike gradu-ates of the military academies or war colleges,where leadership decisions have long held a cen-tral place in the curriculum, newly appointed wild-land crew leaders have traditionally taken chargewith virtually no formal leadership preparation(Freedman, 2000; Ruggero, 2001).

Unlike graduates of the militaryacademies or war colleges, whereleadership decisions have long held acentral place in the curriculum, newlyappointed wildland crew leaders havetraditionally taken charge with virtuallyno formal leadership preparation.

Prior research reveals that the absence of prepa-ratory experience weakens a capacity for makingeffective decisions. Consider one adversary ofgood decisions: overconfidence, a moment when aleader believes that a decision outcome is morelikely than the factual situation would predict.Studies confirm that excess audacity is most prev-alent when leaders face decisions on products andmarkets with which they are less familiar. Simonand Houghton (2003), for instance, examined confi-dence among product managers of small computersoftware and hardware firms when they intro-duced radically new products to the market. Themore pioneering the new products—and thus theless familiar the market—the more the productmanagers became overconfident about their likelysuccess. Other research confirms that overconfi-dence can be reduced through straightforwardtechniques such as explicitly identifying the prosand cons of a decision before making it, implyingthat the absence of preparation in such techniqueswould heighten the likelihood of an overconfi-dence bias (Metcalf, 1998).

A similar suboptimal mind-set for decision mak-ing may be anticipated among firefighters sincethey are constantly forced to make decisions inunfamiliar terrains with little or no leadershippreparation. Prior to the South Canyon fire, thefederal agencies responsible for wildland fire-fighting provided virtually no leadership training.A “can do” attitude on the part of firefighters wasessential for their daunting challenges, but unlessdisciplined through explicit leadership develop-ment, it may result in overconfidence that a givenstrategy will be effective and safe.

ACUTE DISTRESS AND LEADERSHIP DECISIONS

Wildland fires can reach 2,500° Fahrenheit, raceforward at 25 miles per hour, and leap overheadwithout warning. On occasion such fires “blowup,” an inflection point when they acquire a manicmomentum of their own. Like avalanches and tor-nadoes, a blowup is one of nature’s most terrifyingspectacles, and fire leaders know that their crews


could bear personal witness. Tension is thus ever-present in a fire zone, and since team leaders carrypersonal responsibility for the lives of others, theirstress can become acute. Research has confirmedthat when individuals are under time pressure orperforming multiple tasks at the same time, theyare more likely to take suboptimal decisions for ahost of reasons (including a reluctance to searchfor relevant information), and much of the stressexperienced by firefighters is a direct product ofthe urgent and diverse demands imposed on crewleaders and incident commanders when con-fronted by a fast-evolving fire (Janis & Mann, 1977;Finucane, Alhakami, Slovic, & Johnson, 2000; Gil-bert, 2002).

Prior research also demonstrates that the ad-verse effects of underpreparation on decision mak-ing become most pronounced under acute stress. Astudy by Fiedler (1992) of captains and lieutenantsamong urban firefighters, for example, found thatthe performance of seasoned officers improved un-der the stress of a fire, but the performance ofless-prepared officers declined. Consistent withthis finding that experience pays, a battalion chiefwith the New York Fire Department urged from hisown 25-year experience that managers pay partic-ular attention to their own inner voice when takingdecisions under stress. For “making the right callwhen the heat is on,” he writes, “intuition is reallyyour subconscious trying to offer up the benefits ofa lifetime’s worth of experience” (Salka, 2004: 122).Drawing on extensive research on those facingdifficult decision points, Klein (2003) similarly con-cluded that intuition—if well honed and informedby experience—improves decision making understress. A fire crew leader or incident commanderwho is relatively underprepared for leadershipmay thus be expected to make decisions less wellunder the pressure that is often experienced infront-line firefighting.

AMBIGUOUS AUTHORITY AND LEADERSHIPDECISIONS

The decision-making burden on fire leaders ismade even greater by three organizational factorsthat are especially prevalent in wildland firefight-ing. First, crew leaders guide a workforce that islargely seasonal, many employed for the summermonths but otherwise unengaged. Second, theycollaborate with other agencies over which theyhave no control. And third, as fire crews combineinto temporary organizations on larger blazes,crew leaders and incident commanders find them-selves working with, reporting to, or instructing

other crews and leaders whom they had never met.All three sources of ambiguous authority work to

undermine optimal decision making: the first be-cause crews are often underdeveloped as teams;the second because related parties bring self-interested agendas to bear; and the third becausecrew leaders and incident commanders must or-chestrate others that they sometimes scarcelyknow. The weak relations among the various par-ties can result in a crew leader or incident com-mander receiving less information than is avail-able within the teams for making informeddecisions. In their study of aircraft carrier flightdecks, for instance, Weick and Roberts (1993) foundthat responsible officers were more likely to com-mit errors during stressful landing episodes whentheir collective mindfulness and mutual heedful-ness was insufficiently developed or became im-paired by the rush of events. The consequences ofambiguous authority are thus likely to include areduced flow of information to the fire leader, aweakened commitment by the leader to exerciseauthority, and diminished team compliance withthe leader’s instructions (implied, e.g., by Edmond-son & Watkins, 2003).

LEADERSHIP DECISIONS, ATTRIBUTES, ANDOBJECTIVES

Good leadership decisions, we have argued, are asource of both leadership attributes and organiza-tional results. Those with leadership responsibili-ties are thus likely to look less like leaders andalso to fall short of their leadership objectiveswhen their decisions are suboptimal. For leadersof firefighters, more specifically, both their reputa-tions and the triple goals of safety, speed, andsuppression are likely to be impaired by the threemajor sources of suboptimality that they face—underpreparation, acute stress, and ambiguousauthority—as depicted in Figure 1.

Underdevelopment, stress, and ambiguity wereprecisely the circumstances that firefighter DonMackey faced on the afternoon of July 6, 1994 as heled an assemblage of firefighters in its attack onthe South Canyon fire. As we shall see, those fac-tors weakened his capacity to make optimal deci-sions that would lead his firefighters to suppressthe South Canyon fire and ensure their safety indoing so. By time the fire ran its course, the threesuboptimizing factors had contributed to one of thegravest disasters of American wildland firefight-ers ever (Maclean, 2003).


LEADERSHIP DECISIONS DURING THE SOUTHCANYON FIRE

“Okay, everybody out of the canyon!” radioed DonMackey to his beleaguered firefighters at 4:06 p.m.as they raced along a narrow path they had beenclearing for the past several hours. Wrappingaround the West Flank of a ridge from the ZeroPoint to Lunchspot Ridge, Mackey’s 2,100-foot fire-line was to serve as a narrow fuel-free barrier tothwart the fire’s downhill spread.

But now the fireline had become an escape path,and the crew’s only way out—although even thenan uncertain way—to safety. Eighteen firefighterswere running for their lives, and as they sprintedtoward what would become known as the ZeroPoint—the place where their trail reached the topof the mountain ridge—the blowup was threaten-ing to overtake them before they could reach thecomparative safety of the ridge’s backside just be-yond the Zero Point. Six firefighters would outpacethe fire, but for Mackey and 11 others, their raceagainst time would be lost (Figure 2).

To understand how Mackey’s world had becomeso perilous, we return to the prior afternoon whenhe had parachuted onto the ridge where the SouthCanyon fire had been smoldering for 3 days.Mackey had been the first in a group of eightsmokejumpers with the U.S. Forest Service to jumpinto the zone, landing at 5:45 p.m. The smokejump-ers followed their customary protocol of namingthe first on the ground to be the “jumper-in-charge.” It was the crew’s first parachute drop of

the season, and it was to be Mackey’s first serviceas a leader of a crew that large. He readily shoul-dered the responsibility despite an absence of for-mal training in the leadership it would require (theaccount that follows draws upon Butler et al., 1998;Maclean, 1999; South Canyon Fire Accident Inves-tigation Team, 1994; and direct inspection of thefire zone).

Another firefighter—Butch Blanco—already heldoverall responsibility for the fire as its incidentcommander. He had hiked up the mountain earlierthat day with a crew of 7 firefighters from the U.S.Bureau of Land Management (BLM) and had con-structed a fireline to contain the blaze. Just 15 min-utes before Mackey landed on the ridge, however,Blanco had begun walking his own crew off themountain to the nearby town of Glenwood Springsfor the night. The blaze had seemed small andslow-moving enough to permit a night’s retreat,and Blanco’s crew in any case did not have theproper gear to spend the night on the mountain.

Later that evening, Mackey radioed Blanco thatflames had crossed the fireline that Blanco’s crewhad built earlier in the day, and by 10:30 p.m. thefire had alarmingly expanded to 50 acres, up from30 acres just several hours earlier. Mackey, nowdeciding as the incident commander in Blanco’sabsence, radioed a request at 11:30 p.m. for theaddition the next morning of two high-performanceteams known as Type 1 or “hotshot” crews, a totalof 40 new firefighters. Hotshot crews are known fortheir dedication and discipline.

FIGURE 1Sources of Suboptimal Leadership Decision Making


Butch Blanco was formally the incident com-mander. Yet in Blanco’s absence from the moun-tain that evening, Don Mackey was already mak-ing decisions as if he had become the de factoleader. Mackey’s evening request for the Type 1crews was his first substantial decision (Decision1). During the 16 hours ahead, Don Mackey made anumber of decisions on the mountain, some ofthem in collaboration with Butch Blanco, and wefocus on the ten most consequential. Our focus onthe ten decisions provides a compact method forunderstanding why some leadership decisions be-came suboptimal.

The most consequential decisions are those thathad greatest bearing on the crew’s triple objectivesof speed, safety, and suppression. In identifyingthe ten, we draw upon our collective experience in

reviewing the reports on the fire, walking and ex-amining the fire zone (a total of 12 days on StormKing mountain among us), and study of leadershipdecisions in related fire zones such as MannGulch. We believe that other qualified observerswould pick much the same decision set, and weare reasonably confident that these are the tenleadership decisions that had greatest impact onwhether the fire would be swiftly and safely sup-pressed.

TEN CONSEQUENTIAL LEADERSHIP DECISIONS

For Don Mackey and his smokejumpers, it was afitful evening on the mountain ridge on July 5 assummer temperatures plunged to 58°. Awake at2 a.m., Mackey noted that the fire was exception-

FIGURE 2Ridge on Storm King Mountain Where the South Canyon Fire Overran Don Mackey and 13 Other

Firefighters on July 6, 1994. Note. Photo and named features from Butler et. al., 1998, p. 5; original photoby J. Kautz, U.S. Forest Service.


ally active for that time of the night, and withdawn’s first light on July 6, he requested fixed-wing aerial surveillance of the fire (Decision 2).The BLM office responsible for such aircraftpromptly denied the request for an “eye in the sky”and instead assigned a light helicopter the dualtask of ferrying gear and aerial observation. At thetime, the South Canyon fire was just one of severaldozen fires being actively fought in the area, and itwas still low on the priority list for receiving scarcefirefighting resources. Later, that helicopter wouldbe conscripted into dropping water onto the fire aswell. The rejection of Mackey’s request for dedi-cated, continuous airborne surveillance of thespreading fire was an early warning that organi-zational complexity and the resulting ambiguity inauthority would thwart this leader’s capacity toacquire the vital information he would later needfor making informed decisions.

While Mackey was attempting to get a handle onthe fire, Butch Blanco had been climbing back upto the Storm King ridge with a crew of 11 firefight-ers drawn from both the Forest Service and BLM.Blanco’s crew crested the ridge by 8 a.m., andMackey huddled with Blanco at 8:45 a.m. to pooltheir overnight intelligence and decide upon theday’s strategy.

At 10:27 a.m., a second group of smokejumpers—one of the two Type 1 crews requested by Mackeythe night before—floated onto the ridge. They ar-rived as a crew but hardly a team: Two of thefirefighters were based in Montana, two in Wash-ington, four in Idaho. As the diversity of personnelescalated, so too would the ambiguity of Mackey’sauthority over them.

With the fire advancing and firefighters assem-bling, Blanco and Mackey decided that their ownaerial reconnaissance had now become essential,and they asked the helicopter pilot to fly them overthe fire zone (Decision 3). What they witnessed at9:30 a.m. was worrisome: The fire had now ex-panded to 125 acres, more than double the daybefore. Aggressive action was called for, andMackey proposed cutting a fireline very close tothe flames on the West Flank of the ridge. His linewould slope sharply down—a 55% gradient inplaces—for 300 feet below the Zero Point, and then,having descended beneath the lower rim of the firethat was crawling down the west face, the line wasto turn left and extend horizontally across theslope. Blanco approved Mackey’s plan. It was notMackey’s decision alone since he was not formallythe incident commander, but he had proposed thestrategy and now became its prime mover.

Mackey initiated the West Flank fireline at 11:30a.m. (Decision 4) although its course conflicted

with several established decision rules. “Downhillfireline construction is hazardous in steep terrain,fast-burning fuels, or rapidly changing weather,”warned the wildland firefighters’ manual. Mackeywould soon be facing not just one of those condi-tions but all three. “Downhill fireline construction,”the manual cautioned, “should not be attemptedunless there is no tactical alternative.” And, al-though the manual says to “avoid locations thatrequire a steep uphill escape route,” the firelinewould later become an uphill escape route (Inci-dent Operations Standards Working Team, 2002:7–8). As the conditions became more challenging,Mackey’s lack of leadership training likely contrib-uted to his decision’s violation of several standardoperating procedures that an experienced andtrained leader would have felt more compelled tofollow.

Unknown to Mackey, beyond a vertical ridgelinerunning down the fireline slope—Lunchspot Ridgeidentified in Figure 2—the fire had already burnedbelow where the horizontal section of his firelinewas to be cut. Moreover, as Mackey commencedwork on the fireline, Lunchspot Ridge blocked hisview of the downhill fire movement. Standard fire-fighting policy requires a crew leader to know thata fireline will be free of fire below it through theposting of a “lookout who can see the fire” (Inci-dent Operations Standards Working Team, 2002: 8).But no observer was so posted, blinding Mackey tothe explosive conditions that were then developingbelow his line.

Other factors of course may have contributed toMackey’s absence of attention to such principles.Sometimes standard procedures are so routinelyviolated that the enterprise’s culture comes tohonor the principles as much in the breach as theenforcement. Seasoned urban firefighters, for in-stance, tend to break with certain required rou-tines between fires more often than beginners (Fie-dler, 1992). Experienced directors and executives atEnron and Tyco, presumably well aware of securi-ties laws, nonetheless frequently violated stan-dard accounting and tax policies (Useem, 2004b).Still, leadership development programs are de-signed to ensure that those with responsibility donot routinely violate principles essential for thesuccess or survival of the enterprise. In the case ofthe midday development on July 6 in the SouthCanyon fire, this would have meant avoidance of adownhill fireline and assignment of a lookout toguard against a fire below the fireline.

Mackey’s own crew was initially skeptical of hisdecision to create this particular line. One of hissmokejumpers radioed back after receiving Mack-ey’s instruction: “You sure you want us to do that?


Go down that side?” Mackey reaffirmed, but waschallenged again: “Are there any safe spots downthere?” Mackey offered reassurance—“It doesn’tlook too bad”—but his smokejumpers momentarilyrefused to move: “We’re going to wait for you tocome down here and explain some stuff to us,”said one. Once face to face, Mackey elaborated,“Let me have a big crew and we’ll do this, we’ll dofine,” and now his words proved more persuasive.His directly expressed confidence in his firelinedecision nudged the crew downward, though infact it was a misplaced reassurance stemming inpart from his overconfidence in a terrain and tacticof which he was little familiar (Maclean, 1999: 77).

At 11:30 a.m., the smokejumpers moved downhillto start the West Flank fireline. When Butch Blancosoon asked Michelle Ryerson, one of the firefight-ers who had walked up with him that morning, totake a group down the line to support Mackey’seffort, she flatly refused. Although Blanco askedRyerson to support Mackey rather than Mackeyhimself, it was on Mackey’s behalf since it wasMackey who had proposed the strategy and wasnow executing it. Mackey’s premise for doing“fine”—a big crew on the line—was proving unten-able from the outset in the absence of his unam-biguous authority to place a big crew on the line.With the benefit of hindsight, it was evident that alethal combination of underpreparation, intensify-ing stress, and ambiguous authority was alreadytaking form, but it was not so evident to Mackey inthe fire zone at the time.

Mackey’s uncertainty of authority occasioned bythe ad hoc melding of crews from diverse localesand several agencies became further exacerbatedat 1 p.m. when a team of 20 hotshots from Prinev-ille, Oregon began arriving on the ridge by heli-copter. On the mountain now with one missionwere three leaders and three crews with firefight-ers from five states (see Table 2).

When the first hotshot subgroup landed, it wasmomentarily in the fire zone without its ownleader, Tom Shepard. Normally the hotshot super-visor would have arrived with the initial insertion,but Shepard had been bumped from the first heli-

copter flight because of a weight restriction, aseemingly minor snag whose consequences wouldsoon loom large.

In Shepard’s absence, Blanco instructed the firstwave of Oregon hotshots on the ground to go downthe West Flank fireline to join Mackey’s group thatwas working to extend the line. When Shepardlanded a few minutes later, Blanco told Shepardthat the first hotshots had already descended ontothe line. The Prineville boss acquiesced in the de-cision that had been taken in his absence, reas-sured in part by the fact that his hotshots would bejoining smokejumpers already working on the lineunder Mackey’s direction. It was fair to as-sume—as Shepard did—that an experienced andtrained jumper-in-charge would not place his ownfirefighters at undue risk.

Hotshot crew leader Tom Shepard did not walkdown to inspect the line, choosing instead to re-main on the ridge top. Since Shepard had justarrived on the scene and had performed no recon-naissance of his own, he was of necessity relyingentirely upon Mackey’s vigilance for the protectionof his hotshots. Without explicit acknowledgement,Mackey was further acquiring the essence of inci-dent commander. Still, because it came withoutformal recognition and it extended across teamsover which he held no formal power, his emergentauthority remained equivocal.

Butch Blanco had started as the incident com-mander before Don Mackey’s smokejumpers hadarrived the prior afternoon, but Blanco’s nighttimeabsence from the fire zone had ceded much controlto Mackey. It was Mackey, for instance, who askedfor the insertion of Type 1 crews and for continuousaerial surveillance on July 6, not Blanco. In a fur-ther shift of control, Blanco had given oversight ofthe central strategy of the day—construction of theWest Flank fireline—entirely to Mackey. Blanconever descended to have a direct look at the fire-line nor further instructed Mackey in its making.Mackey was now directly supervising 24 firefight-ers, nearly half of those on the mountain, all at theforefront of the day’s main attack on the fire.

The ambiguity was enough to cause Mackey to

TABLE 2Fire Crews Fighting South Canyon Fire on the Afternoon of July 6, 1994

Crew Type Smokejumpers Local Firefighters Hotshots

Crew leader Don Mackey Butch Blanco Tom ShepardFederal agency U.S. Forest Service U.S. Forest Service and Bureau

of Land ManagementU.S. Forest Service

Members andlocation

16 from Idaho, Montana, andWashington

11 from Colorado 20 from Oregon


wonder if he might have even become the incidentcommander. Around 2 p.m., Mackey mused abouthis emergent authority with fellow smokejumper,Kevin Erickson. He asked Erickson whether hethought Butch Blanco was the incident com-mander, or Mackey himself. Erickson responded “Idon’t know.” Mackey added, “neither do I,” but hetook no subsequent steps to clarify his place in theday’s hierarchy (Decision 6).

Michelle Ryerson radioed Mackey at 3:30 p.m.from the crest of the ridge and asked him to decidewhether a water drop by the helicopter should bedirected on top of the ridge as requested by severalfirefighters close to her, or instead should be di-rected near the most forward firefighters withMackey on the West Flank fireline. Mackey de-cided the top was more important at the moment.The query to him on how to allocate a criticalresource—and his resolution of the issue withoutconsulting Blanco—again served to reinforce thefact that he was practically functioning as the in-cident commander. But Ryerson’s earlier refusal tojoin Mackey on the line also served as a reminderthat his authority was not all that clear cut.

Yet, without formal assignment of incident com-mand, Mackey would not feel impelled to takecomplete responsibility for the safety of the 49 fire-fighters now spread across the ridge. Like leadersof any organization, incident commanders are ex-pected to be completely aware of a threateningenvironment as it evolves, and acceptance of thatresponsibility here would have dictated the as-signment of fire lookouts, radio contact with allparties, and personal positioning for optimal ap-praisal. “Look up, down and around” exhorts thefire manual. But without the creation of such ob-servers, communication, or positioning—not sur-prising since Mackey was not unambiguously incharge and new to leadership and untrained init—he could not appreciate that the ridge was onthe cusp of a blowup.

Further limiting Don Mackey’s situationalawareness was a critical weather forecast for theafternoon that never reached him. The local mete-orologist correctly predicted that a cold front wouldsurge through the region around 3 p.m., generatingno rain but plenty of wind. Standard operatingprocedures required that such information be con-veyed to fire crews, but bureaucratic bungling pre-vented its delivery that day. And it would be thatsudden wind—gusting from 30- to 45-miles-perhour soon after 3:20 p.m.—that drove so much oxy-gen into the fire zone that a blowup became a nearcertainty within the hour.

The supply side of the equation failed to deliverthe weather warning, but the demand side also did

not seek it. The principles of incident commanddictate constant awareness of evolving weatherconditions, but Mackey did not request up-to-dateinformation on the day’s atmospherics (Decision 7).With his hands full supervising the front line,Mackey could take the absence of a radioedweather warning to him to signify no dire threats.But had Mackey preemptively sought the forecast,even as late at 4 p.m.—well after the winds hadbegun to kick up—he would still have had time toevacuate everybody off the West Flank fireline be-fore the blowup. Here was a decision that couldhave made up for the earlier suboptimal decisions,even just minutes before the conflagration over-whelmed the firefighters. Again, the fact thatMackey was not definitively in charge of the firemay have cast doubt on whether it would evenhave been appropriate for him to demand an up-dated forecast, and that reluctance may have beenreinforced by his own underpreparation in the es-sential operating principles of fireline leadership.

In Mackey’s two preceding decisions—not toclarify who was in charge or to seek a weatherforecast—he had faced opportunities to move tohis firefighters in one direction or another. We de-fine leadership decisions to include those mo-ments when no choice was made, and now a dis-crete, tangible, and realistic opportunity to pursuean alternative course had presented itself. In fac-ing the opportunities underlying Decisions 6 and 7,had Mackey taken those decisions other than theway he did (doing nothing), the subsequent expe-rience of the firefighters on the West Flank firelinewould have been very different. If Mackey hadclarified that he was in fact in charge—an optionfacing him at Decision 6—that realized responsi-bility might well have compelled him to seekweather data—an option at Decision 7. He thenwould have most likely appreciated that an immi-nent arrival of violent winds on Storm King Moun-tain was on the verge of endangering all those onthe downhill fireline.

We can also identify other moments of foregoneopportunity to pursue alternative courses duringthe South Canyon fire. Mackey, for instance, couldhave simply decided to walk his team off themountain earlier in the day. But within the contextof his unfolding experience on the mountain, thatwould not have been a realistic option, since inci-dent commanders, de facto or otherwise, virtuallynever stage a full-scale retreat. Decisions 6 and 7in which Mackey did not make an active choiceamong alternatives are thus included among hismost critical decisions because they rank amongthe most discrete, tangible, and realistic decisionsthat he faced.


At 3:30 p.m., Don Mackey took note of the newlygusting winds, and he assigned firefighters KevinErickson and Sarah Doehring to patrol the firelinefor burning debris and hot spots that might beblown across the line. He also radioed a group ofhotshots working at the end of the fireline near theLunchspot Ridge (so named because a group hadstopped for a food break there at 2 p.m.) and askedthem to improve the line in their area. At 3:30, mostof the firefighters on the West Flank fireline wereworking between 1,450 to 1,880 feet from the ZeroPoint.

By 4 p.m., the signs of a blowup were fast mate-rializing below Lunchspot Ridge: Smoke swirled,flames churned, air thundered. Standing close toLunchspot at 4:06 p.m., Mackey instructed eightnearby smokejumpers to run directly up the ridgeto an area where he knew that the fire had previ-ously burned off sufficient ground cover to serve asa small enclave, a safe zone if the fire came up theridge (Decision 8). “Go up!” he shouted, “there’sgood black further up.” They would indeed findthat “good black” and survive when the fire soonshot up the ridge. This was a swift leadershipdecision that proved life saving.

But rather than run up Lunchspot Ridge with theeight other firefighters, Mackey ran back along theWest Flank fireline to help push the remainingfirefighters on that line to the safe zone beyond theZero Point (Decision 9). He radioed ahead: “Okay,everybody out of the canyon!”

In going back across the West Flank firelinerather than up Lunchspot Ridge Mackey made theleader’s ultimate decision (Decision 10). Although farriskier than going up, moving across would helpsave those still on the fireline. At a moment whenteam interest and self-interest were radically diverg-ing, Mackey focused entirely on the first.

At a moment when team interest andself-interest were radically diverging,Mackey focused entirely on the first.

Already sensing the emerging blowup on theirown, the West Flank firefighters had begun to rushalong the fireline toward the Zero Point, andMackey dashed along that line to catch up withthem. By 4:09 p.m., Mackey had reached the tailend of the 18 fleeing smokejumpers and hotshotsnear a tree that was just 450 feet from the ZeroPoint (marked in Figure 2).

Mackey’s last directive almost worked. He and 11others had come within 120 to 280 feet of the ZeroPoint and the safety of the back slope before they

were overtaken by the flames between 4:13 and4:14 p.m. The most forward of the nonsurvivors,Scott Blecha, was less than a minute from reachingthe Zero Point. Two others elsewhere on the moun-tain became entrapped as well, making for a finaldeath toll of 14 firefighters. Five days later, a farlarger fire crew finally brought the South Canyonfire under control.

LEADERSHIP DECISIONS APPRAISED

Don Mackey’s decisions had enormous impact onthe firefighters’ experience during the South Can-yon fire. Yet he was certainly not the only decisionmaker on the mountain. Butch Blanco and TomShepard added their own important decisions tothe matrix, as did off-site officials. The latter, forinstance, never delivered the high-wind alert tothose on the front line. Others dispatched DonMackey onto the mountain with insufficient re-sources in the first place.

Still, Don Mackey’s 10 leadership decisions be-came critical on the ground, and their impact issummarized in the “Result” column of Table 3. The10 decisions are also displayed in Figure 3 as anoverlay on the diagram in Figure 2 of the fire ridgeon South King Mountain.

By integrating the authors’ professional experi-ence with detailed study of the incident for whichthe decision making record is exceptionally welldocumented, we have concluded that 5 of Mackey’s10 decisions proved relatively optimal while theother 5 were less so. We cannot be completely sure,of course, since we do not have the controlled ele-gance of a laboratory experiment or the statisticalvalue of a sample. Still, 5 decisions proved advan-tageous in that they arguably improved the likeli-hood that Mackey and his team would swiftly andsafely suppress the South Canyon fire:

• Decision 1 at 10:30 p.m. on July 5 to request twoadditional elite crews secured and assignedthe people that would be required the next dayto combat the rapidly expanding fire.

• Decision 3 at 9:30 a.m. on July 6 to conductaerial surveillance significantly improved DonMackey’s information on the environment.

• Decision 8 at 4:06 p.m. to dispatch 8 smoke-jumpers up Lunchspot Ridge successfullymoved them into a life-saving area.

• Decision 9 at 4:06 p.m. to evacuate the WestFlank fireline helped successfully move 6 fire-fighters toward another life-saving area.

• Decision 10 at 4:06 p.m. to go back across theWest Flank fireline rather than up LunchspotRidge placed team survival ahead of personalself-interest.

The five other decisions proved less optimal inthat they lessened the likelihood that Mackey and


his team would swiftly and safely suppress thefire. Keep in mind that this assessment has thebenefit of coming after the incident, and its inten-tion is to improve leadership decisions in the fu-ture rather than find fault. The suboptimality of thefive decisions can be explained in part by one ormore of the three factors identified at the outset aspotentially undermining optimal leadership deci-

sion making—under preparation, working underacute stress, and ambiguous authority:

• Decision 2 at 5:30 a.m. on July 6 to requestcontinuous aerial surveillance that went unful-filled limited Don Mackey’s information on,and thus understanding, of the fire. The intentbehind his decision was sound: He sought the“eye in the sky” so that he would have real-

TABLE 3Appraisal of 10 Leadership Decisions by Don Mackey During the South Canyon Fire on July 5–6, 1994

No.Decision by

Donald Mackey Time

Source ofSuboptimal

Decision Making Appraisal ResultDecision Quality 3Leadership Attribute

1 Requested twoadditional crewsfor the next day

10:30 p.m.,July 5

Effectiveassessment andprompt decision

Acquired theadditional personnelrequired

Assign people 3able to mobilize

2 Requestedcontinuous fixed-wing aerialsurveillance

5:30 am,July 6

Ambiguousauthority

Right decision butthwarted by poorinteragencycoordination

Prevented appraisal offire’s rapid downhillspread

Acquire data 3strategic thinker

3 Conducted ownaerialsurveillance

9:30 a.m. Acquired goodoverview of fire

Enhanced intelligenceon the fire’s growth


4 Started buildingdownhill WestFlank firelinewithout lookout

11:30 a.m. Underpreparationand ambiguousauthority

Aggressive andunsafe

Placed firefighters inperil

Implement 3 ableto execute

5 Added hotshots toWest Flankfireline

1 p.m. Underpreparationand workingunder acutestress

Taken by Blanco inabsence ofShepard; implicitin Mackey’sstrategy

Drew additionalfirefighters ontounsafe terrain

Assign people 3able to mobilize

6 Did not clarify whowas in charge ofincident

2 p.m. Underpreparation,acute stress,and ambiguousauthority

Left authorityambiguous whenit should havebeenunequivocal

Resulted ininsufficient effort togain full situationalawareness as wellas lack of clarityamong followers asto who was incharge


7 Did not secureweather update

3 p.m. Underpreparationand ambiguousauthority

Left Mackeyunaware of theimminent arrivalof a cold frontwith high winds

Firefighters continuedfireline work despitegusty conditionsahead


8 Sent 8 smoke-jumpers upLunchspot Ridgeto safe area

4:06 p.m. Clear-mindedinstructionduringincreasinglystressfulmoments

The 8 smokejumperssurvived

Assign people andimplement 3 ableto mobilize andexecute

9 Evacuated WestFlank fireline

4:06 p.m. Optimal given theconditions at themoment

Helped save 6firefighters, but toolate for 11 others

Assign people andimplement 3 ableto mobilize andexecute

10 Personally soughtto clear the WestFlank fireline

4:06 p.m. Placed crew safetyahead ofpersonal risk

Sacrificed himself forthe benefit of others

Focus on goal 3selflessness


time data on areas of the fire that he could notsee directly. The fact that he was unable toobtain the requested resource may haveseemed unremarkable at the time, since fire-fighters are often denied such requests be-cause of competing priorities, safety concerns,or resource constraints. Still, arguably contrib-uting to the rejection of his request was the factthat he was not formally the incident com-mander or otherwise of sufficient influence inthe hierarchy to succeed in pressing the casefor aerial surveillance, which he explicitly re-quested and in hindsight clearly needed. Theoutcome was to leave him partially blindedfrom what within hours would evolve into afatal conflagration. With more clearly delin-eated authority, Mackey might well have se-cured the overview he sought of a fire that wasalready dangerously expanding.

• Decision 4 at 11:30 a.m. to start constructing thedownhill West Flank fireline would place anumber of firefighters in peril. We cannot becertain that a better trained supervisor wouldhave made a safer decision under the stress ofthe moment, but it is arguable that a well-

prepared leader would have been less san-guine about transgressing standard operatingprocedures and more likely to take extra mea-sures to guard against downhill dangers. Amore experienced and trained leader mightalso have been more uncomfortable predicat-ing the moment’s prime strategy on an essen-tial resource not yet firmly in hand: the “bigcrew” that Mackey required if his line was to“do fine” never did materialize. A leader withunequivocal authority might also have beenmore successful in deploying the big crew thatwas required.

• Decision 5 at 1 p.m. to bring the newly arrivedhotshots onto the downhill fireline, good fromthe standpoint of building the ample force thatMackey judged was required, had the effect ofplacing more firefighters in harm’s way. Again,the overoptimism of underpreparation for lead-ership likely played a role here, exacerbatedby the worsening fire conditions and growinganxieties on the mountainside. With greatertraining in realistic appraisals of market con-ditions, a leader should be more savvy about

FIGURE 3Leadership Decisions by Don Mackey During the South Canyon Fire on July 5–6, 1994. Photo: Jim Kautz,

U.S. Forest Service.


the proper allocation of resources given theincreasingly dangerous conditions.

• Decision 6 at 2 p.m. not to clarify who was incommand of the incident resulted in Mackey’sunderinvestment in the requisite situationalawareness. The ambiguity of authority stem-ming from the diverse units on the mountain,the stress of firefighting, and sheer fatigueprobably contributed to this suboptimal deci-sion. So too did underpreparation: With greaterexperience and training comes greater appre-ciation for the requirement of unequivocal clar-ity in who carries ultimate responsibility on theline.

• Decision 7 at 3 p.m. not to secure a weatherforecast left those on the West Flank firelinevulnerable to an unanticipated blowup. Again,the uncertainty of whether Mackey was inci-dent commander was likely a factor, especiallyin combination with his underpreparation andhis other preoccupations. The fire was becom-ing more threatening, Mackey had far morefirefighters under his wing with the arrival ofthe hotshots, and without a warning light inthe back of his mind from prior experience andtraining, the danger of a missing weather fore-cast was not necessarily top of mind.

LEADERSHIP DECISIONS AND LEADERSHIPATTRIBUTES IN FIRES AND COMPANIES

The five good decisions were consonant with fourattributes expected of a team leader: strategicthinking, resource mobilization, effective execu-tion, and personal selflessness. As summarized inthe last column of Figure 2, Don Mackey mobilizedthe resources required to achieve the mission (De-cision 1); worked to better inform his strategicthinking about the fire (Decision 3); mobilized histeam and executed swiftly when essential (Deci-sions 8 and 9); and risked his own life for thesurvival of others (Decision 10).

The five suboptimal decisions by contrast under-mined the same four attributes: Don Mackey fellshort of the strategic attentiveness to the fire envi-ronment essential for leading his team (Decision2); executed an excessively risky strategy (Decision4); mobilized firefighters onto a dangerous terrain(Decision 5); did not sharpen strategic awarenessof who was in charge (Decision 6); and did not thinkstrategically about the threatening weather (Deci-sion 7).

Behind these four leadership attributes, then,were a set of specific decisions that either pro-duced or diminished the attributes. Although lead-ership decisions are usually harder to see or wit-ness than the leadership attributes, we have seenthat the decisions in this case underpinned theattributes. The road to better leadership attributes,then, is to understand why some decisions are

suboptimal while others are not. And for that, wehave here identified three primary factors that canlessen decision quality: limited preparation, expe-riencing acute stress, and ambiguous authority. Ifall three are concurrently present, as they wereduring the South Canyon fire, the combination canbecome especially lethal for leadership decisionsand thus the attributes of leadership.

If we had focused on Mackey’s leadership at-tributes rather than the underlying leadership de-cisions, we would have run the risk of being misledby the attributes if they were only modestly asso-ciated with the underlying decisions. What maylook like good leadership from the outside cansometimes mask poor leadership decisions on theinside.

Consider the leadership of Enron shortly beforethe company’s collapse in the fall of 2001. Theboard of directors displayed the right attributes foreffective leadership at the top: with 13 members,its midsize was deemed right for effective over-sight; with 11 nonexecutives, it was seen as right-fully dominated by outsiders; and with separationof chair and chief executive, the chair could pre-sumably assure that the CEO would do right by theshareholders (Sonnenfeld, 2002).

Viewed from without, the Enron board’s leader-ship attributes would seem optimal for reachingjudicious decisions, but in actuality, many of thedecisions inside the boardroom proved very inju-dicious. When CFO Andrew Fastow asked theboard on June 28, 1999 to okay one of his specialpurpose entities, for example, approval requiredsuspension of the company ethics code, since itstipulated that “even the appearance of an im-proper transaction must be avoided” and no em-ployee could “gain separately” from company ser-vice. Yet the proposed partnership entailed both(Powers, Troubh, & Winokur, 2002; U.S. Senate,2002).

The proposal for the ethics code suspension ar-rived in the Enron directors’ fax machines just 3days before their board meeting. The agenda wasjam-packed with other matters, no board commit-tee vetted the proposed suspension, the directorsconducted the meeting by telephone, and the tele-conference ended within an hour. The ethics sus-pension sailed through, and in the months thatfollowed, the CFO not only did “gain separately”from the partnerships—pocketing $30 million—buthis “improper transactions” with the partnershipsinflicted fatal damage on the firm. The problemwas not the outward attributes of the Enron board-room—it was a well sized blue-chip set of indepen-dent directors led by a nonexecutive chairman—but the abbreviated and hands-off manner in


which those in the room allowed themselves toreach their suboptimal decisions (Useem, 2003; Lor-sch, 2002; Carter & Lorsch, 2004).

Leadership attributes can thus be misleading inpredicting actual leadership decisions—and thusdangerously deceptive since they would some-times appear to be good predictors. During theCivil War, for instance, Union General George B.McClellan was beloved by his soldiers and re-spected by his president for his leadership of theArmy of the Potomac, but his failure to make timelydecisions for the army during the battle of Antie-tam forever defined him as a failure for the Union.By contrast, Confederate General Robert E. Leewas revered by his soldiers and admired by hispresident for his leadership of the Army of North-ern Virginia, and his timely decisions for the armyon such battlefields as Chancellorsville wouldcome to define him as an icon of the Confederacy.For McClellan, the attribute of charisma served asa poor predictor of leadership decisions on thebattlefield, but for Lee, a potent forecast (Freeman,1998; Sears, 1999; Useem, 2001).

DEVELOPING LEADERSHIP DECISION MAKINGIN THE WILDLAND FIRE SERVICE

Some of the nation’s most elite wildland firefight-ers were on Storm King Mountain on July 6. Hotshotcrews and smokejumpers are among the best tech-nically trained of the country’s vast corps of wild-land firefighters. Yet more than a dozen firefight-ers lost their lives in what most qualified observershave concluded was a preventable disaster. Ifeven those technically well prepared in fire sup-pression could be caught by a blowup, it was notfor lack of formal fire-fighting skills. The disaster,rather, derived in part from an underdevelopmentof leadership skills, especially for making qualitydecisions under demanding and ambiguous con-ditions.

Donald Mackey had parachuted into the firezone as a crew member, became jumper-in-chargeon the ground, and soon assumed virtual com-mand of a multicrew firefighting incident. His re-sponsibilities expanded overnight from follower-ship to leadership, and he willingly rose to theoccasion. Organizations often depend upon peoplelike Mackey to get a job done, and if they are to getthe leadership they require, training and develop-ment in the essence of leadership decision makingis a prerequisite.

The presence of strong technical capabilitiescombined with weak leadership capacities amongthose fighting the South Canyon fire was a directproduct of U.S. policies prior to 1994. The federal

agencies responsible for wildland firefighting foryears had focused on fire behavior rather thanhuman behavior. In retrospect, that was akin to abusiness concentrating on engineering or an armyon ballistics, important, but providing no founda-tion for the well-conceived and well-timed leader-ship decisions required in moments of pressureand ambiguity.

Appreciating the disastrous human conse-quences of suboptimal leadership decisions in thewake of the South Canyon fire, federal firefightingagencies commissioned an independent analysisof safety practices on the fireline. The outsidestudy called for the U.S. government to establish aleadership development process for firefighterswith a special focus on training for good decisionmaking under conditions of stress (TriData Corpo-ration, 1998).

In response, the National Wildfire CoordinatingGroup, a consortium of federal and state wildlandfirefighting agencies, established the WildlandFire Leadership Development Program in 2001 withthe goal of explicitly enhancing “decision-makingskills” so that responsible firefighters could “makesound and timely decisions.” The second and thirdauthors of this article have been responsible forbuilding that program (the program is describedby Cook, 2002; Sutton, 2002; Wildland Fire Leader-ship Development Program, 2004).

Prior to the South Canyon fire, the wildland fire-fighting services had offered brief supervisorytraining but none in leadership. A decade after thefire, an array of courses had been established toprovide in-depth training in leadership decisionmaking under tension in fast-changing, unfamil-iar, and complex environments. They are designedin accord with one of the main principles that nowguide all wildland fire leadership development:“Make sound and timely decisions” (Wildland FireLeadership Development Values and Principles,2004).

In a first-level course for all firefighters (“HumanFactors on the Fireline”) the focus is on situationalawareness and decision making with an emphasison ensuring that personal stress and private atti-tudes do not undermine clear-mindedness. In oneexercise, for example, participants are providedinformation on trends in temperature, humidity,and clouds over a particular terrain as a cold frontapproaches, and they are asked to forecast theway the wind will shift and what slope will be-come most dangerous—and thus how the fire crewshould be instructed as a consequence. In a sec-ond-level course entitled “Followership to Leader-ship,” the fundamentals of leadership are empha-sized, for example, ethical decision making, and


participants devote a day in the field to makingdecisions in small teams. A third-level course,“Fireline Leadership,” is designed for those re-sponsible for a fire crew and emphasizes commu-nicating intent and managing stress, and here par-ticipants study models of decision making, engagein tactical decision simulations adapted from theMarine Corps, and master after-action reviews ofkey decisions.

In a fourth-level course, “Incident Leadership,”for “leaders of leaders”—those who would serve asincident commanders with several fire crews dur-ing an extended attack—the focus is on identifyinghow ambiguous authority can undermine effectivedecisions, how decision errors can be detectedearly, and how decision tempo can be maintained.Here participants engage in role and computersimulations of fire combat. This is the trainingcourse that Don Mackey should have had but couldnot have experienced, since it was then unavail-able. In fact, a person performing the job of DonMackey on a fire today would have received morethan 80 hours of formal leadership training if theyhad completed the courses now available.

In a capstone course for “leaders of organiza-tions”—those who run incident managementteams that oversee dozens of units on a majorfire—the focus is on integrating information fromnumerous sources to reach informed and timelydecisions and on communicating strategic intentto the field in a way that frontline leaders can takeappropriate decisions. Had Don Mackey beenproperly trained before his dispatch to Storm KingMountain and had he then swiftly and safely sup-pressed that fire, this would have been the coursefor a subsequent promotion that he would havelikely received (Wildland Fire Leadership Develop-ment Training Courses, 2004).

Running through all these courses in the Wild-land Fire Leadership Development Program is anemphasis on building the right decision-makingskills for safety, speed, and suppression in a firezone. Learning how to acquire and analyze theright data, allocate people and resources to theessential tasks, implement in a timely fashion, andfocus on the goals of the enterprise regardless ofpersonal concerns are central to the curriculum.The coursework also emphasizes learning to copewith ambiguous authority and personal stress. Al-though those sources of suboptimal decision mak-ing cannot be eliminated, the federal initiative isintended to reduce the third cause of suboptimaldecisions, underpreparation for making leader-ship decisions, especially when authority is uncer-tain and stress is intense.

DEVELOPING LEADERSHIP DECISION MAKINGOUTSIDE THE CLASSROOM

To reinforce the leadership decision-making les-sons of the classroom, the Wildland Fire Leader-ship Development Program created a set of out-of-classroom learning experiences that draw on theconcept of the battlefield “staff ride” long used bythe armed forces. The military staff ride is de-signed to provide officers a deeper understandingof military strategy by walking such battlefields asGettysburg and Normandy to reconstruct the keydecisions (e.g., U.S. Military Academy, 2004; U.S.Marine Corps University, 2004). Although distinctfrom what is often termed outdoor experientiallearning, staff rides offer some of the same powerto educate in ways that classrooms cannot. By wit-nessing other settings, mentally reconstructing thedecisions taken in those settings, and then draw-ing implications for one’s own settings, such out-of-classroom experiences can instill the principlesof leadership more enduringly than classroom ex-perience alone (Meyer, 2003).

The staff rides of the fire leadership programoffer analysis of firefighting strategies and deci-sions in several fire zones. The purpose, in theprogram’s words, is to “put participants in theshoes of the decision makers on a historical inci-dent in order to learn for the future.” These staffrides are intended to build the “decision-makingskills of leaders” by analyzing “fireground deci-sions made by leaders of the past in concert withan on-scene study of the actual terrain.” Whilewalking the terrain, participants evaluate whatdecisions they would have taken had they been inthe shoes of the incident commander (WildlandFire Leadership Development Staff Ride Library,2004).

Staff rides in fire zones have become as federalpolicy directive states, “a decision-making andleadership development tool” for “fire manage-ment personnel at all levels.” Implementing thedirective, the director of the U.S. Forest Service’sFire and Aviation Management informed nationalforest supervisors that the “staff ride concept hasproven many times over that leadership and deci-sion making skills are enhanced through an indepth examination of the fire dynamics and hu-man behavior” (Wildland Fire Leadership Devel-opment Staff Ride Library, 2004; Quintanar, 2003).

The South Canyon fire itself has come to serve asone of the frequently utilized staff-ride venues. Nu-merous groups have walked the West Flank fire-line on Storm King Mountain to review the leader-ship decisions of Don Mackey and others. Ahotshot fire crew from Redding, California even


reoriented its entire training program around lead-ership decision making, and the culmination of its6-week program is an extended walk in the SouthCanyon fire zone on Storm King Mountain (Red-ding Interagency Hotshot Crewmember Report,2003). Indicative of the learning legacy of such staffrides, Redding participants reported that their dayon Storm King Mountain:

• “drove home” that “it is important to have aclear head and act aggressively”;

• enhanced “awareness” of “situations that re-quire quick decisions”;

• “enabled me to identify the error chains andhopefully given me the strength to break thembefore something like this happens again”;

• made me “realize” that “good leadership traitsare very, very important”;

• “was without a doubt . . . the most beneficialtraining and learning experience of my firecareer”; and

• “was one of the most moving and educationalexperiences of my firefighting career.”

As their walk on the mountain came to end, onegroup of Redding firefighters stood near the end ofthe fireline where Don Mackey had shouted,“Okay, everybody out of the canyon!” The facilita-tors asked the firefighters to race against Mackey’sclock to reach the safety of the top. Many ran alongthe same fireline that Mackey had built and used,others bolted directly for the top, but whatever thepath, most failed to reach the safety of the ZeroPoint or ridge top within the few minutes that hadremained for most of Mackey’s crew. As “I ran upthe West Flank fireline,” said one, “I was able toimagine the fire, the smoke, and the heat belowme, and even experienced a few chaotic thoughtswhile I second-guessed my decision to take theWest Flank escape route.”

RESEARCH ON THE DEVELOPMENT OFLEADERSHIP DECISION MAKING

Academic research substantiates the learning pre-mises underlying the Wildland Fire LeadershipDevelopment Program. Dean and Sharfman (1996),for example, found that when manufacturing man-agers make strategic decisions based upon carefulanalysis of good information, their decisions aremore likely to be successfully executed than whenthe managers are less analytic and less informed.By implication, training those in leadership posi-tions to acquire essential information—such as ter-rain conditions and weather forecasts—and thento take reasoned decisions based on the datashould improve the quality of their decisions.

Moreover, research by Eisenhardt (1989) andJudge and Miller (1991) on managerial decision

making in fast-moving environments revealed thateffective decisions depend upon the availabilityand exploitation of good information on the mar-ket—and that well-informed, rapid decisions un-der such conditions led to stronger company per-formance. By inference, training those inleadership positions in what Eisenhardt hastermed “high-velocity environments”—whether inproduct manufacturing or wildland firefight-ing—to make decisions quickly using quality datashould enhance the quality of their leadership de-cisions when on the line.

Studies of what Kahneman (2003) terms the “sys-tematic biases” that separate the choices that peo-ple actually make from what would be posited byrational-agent models reveal that biases can bereduced when managers draw upon intuition thatis built upon prolonged practice and effective useof feedback to learn from practice (Klein, 1998;Gawande, 2002). For younger firefighters new toleadership assignments, leadership courses andstaff rides should thus serve to accelerate acquisi-tion of the experience required for sound intuitionand thereby reduce the gap between their optimaland actual choices.

Decision research also implies that staff ridesmay be a particularly powerful training methodfor overcoming the tendency of firefighters toinadequately appraise low-likelihood but high-consequence risks. Studies report that people oftenperceive low-probability events to have zero like-lihood (Kunreuther, 2001). Since the probability ofhuman loss in most wildland fires is very low,some firefighters may tend to assume it could nothappen on their watch. This may have been afactor in Don Mackey’s overly risky decisions, suchas building the downhill fireline. But when low-probability risks are presented in concrete formbefore real-time decisions have to be taken—aswhen firefighters walk the South Canyon fire zoneand consider the decisions that contributed to the14 victims’ entrapment—the research would implythat when the firefighters subsequently lead ateam into an active fire zone, they should betterunderstand that their decisions could result—withsmall but nonzero probability—in human fatality.

Such experiential learning will be all the stron-ger if firefighters undergo staff rides to severalzones, not just one, since other research confirmsthat learning from experience can be best gener-alized to other contexts if is derived from severalcontexts (Bazerman, 2002). Thus, walks on StormKing should be followed by walks in Mann Gulchand kindred zones, and to provide for a range ofsuch opportunities, the Wildland Fire LeadershipDevelopment Program developed a host of staff


walks (Wildland Fire Leadership DevelopmentStaff Ride Library, 2004).

Confirming the indelible impression that con-crete embodiment can provide, one of the Reddinghotshots reported an emotional reaction uponnearing the area where many of the fire victimsperished, just several hundred feet from the ZeroPoint. “A pair of skis formed an X over one crossmaking the spot where one of the firefightersdied,” he recalled, and “I immediately became par-alyzed and stunned.” The Redding firefighter hadencountered the skis when testing to see whetherhe could reach the Zero Point in the time that Mack-ey’s crew had. Having failed to do so, he remindedhimself “this meant that if I was on the mountain in1994, I too would have perished.” Upon that real-ization, he urged all hotshot crew members to“climb that mountain to see and experience theemotional connections.” His experience-basedsuggestion is corroborated by Bazerman’s re-search-based assessment that an “event thatevokes emotions and is vivid, easily imagined, andspecific” will have greater hold on an individual’smemory than unemotional and bland events—andthus be more informative of one’s future decisions(Redding Interagency Hotshot Crewmember Re-port, 2003; Bazerman, 2002: 7).

In the aftermath of the South Canyon fire, thewildland fire service has used classroom and ex-periential learning to attack two of the three rootcauses of the suboptimal decisions on the moun-tain: inadequate preparation for leadership deci-sion making under stress. Separately, the fire ser-vice has attacked the third root cause—ambiguityof authority—by sharpening and better instillingthe principles of unequivocal leadership responsi-bility when on a fireline.

DEVELOPING LEADERSHIP DECISION MAKINGIN ALL ORGANIZATIONS

In drawing upon the decisions of Don Mackey andothers on that fateful day in 1994, the Wildland FireService has been respectfully learning from thosewho made the ultimate sacrifice to help ensurethat it is not repeated. We have sought to do thesame here, and we would suggest extending whathas been learned from the South Canyon fire di-saster to the leadership of any organization,whether a business firm, medical center, or publicagency (several such lessons have been developedfor business by Charlier, 1994; Roberto, 2001).

Most company managers will never find them-selves on a fireline, but their leadership decisionswill find parallels with those of the fireline. Labor-ing under intense investor, competitor, or customer

pressures, and working with uncertain lines of au-thority, managers must often make critical deci-sions on staffing personnel (akin to Mackey’s De-cisions 1 and 5), market analysis (Decisions 2, 3,and 7), competitive strategy (Decision 4), division ofresponsibility (Decision 6), workplace safety (Deci-sions 8 and 9), and placement of company goalsahead of career objectives (Decision 10). Makingoptimal leadership decisions at such points can beimportant for both company results and personallivelihoods, just as they were for Mackey’s crewsand his own livelihood. Studying the sources ofsuboptimal decisions on Storm King Mountain cantherefore yield fresh insights into leading underanalogous circumstances at home.

Whether a fireline or company venue, makingquality leadership decisions under conditions ofstress and ambiguous authority is not a naturalcapacity, and we conclude from study of the SouthCanyon fire experience and its aftermath that allorganizations would do well to prepare their lead-ers for demanding decisions under uncertain cir-cumstances. Moreover, if organizations want thosein responsible positions to bring four major lead-ership attributes to their office—strategic thinking,resource mobilization, effective execution, and per-sonal selflessness—they will also want to focus onimproving their leaders’ decision-making capabil-ities. This will be especially important when theirmanagers have been recently promoted or facenew markets, confront chaotic conditions, or workwith less than authoritative control.

Since many managers face one of those condi-tions and some confront all, initiatives to developgood leadership in most organizations would bewise to start with the essentials of building qualityleadership decisions. This would apply to govern-ing bodies as well. Carter and Lorsch (2004: 35)have noted that although demands for governancereform have mainly focused on “what is visible” ofthe board, the “real action is in the boardroomitself.” If directors and trustees are to lead alongwith their appointed executives, improving theirdecisions behind closed doors should be a com-pany priority.

Effective preparation of leaders, however, is sub-ject to its own set of hazards. These include over-control, inadequate customization, and misleadingmetrics, and the building of an effective leadershipdevelopment program will require sustained andthoughtful management attention to avoid the pit-falls (Conger & Fulmer, 2003; Ready & Conger, 2003;Gandossy & Effron, 2004).

Since business schools carry responsibility forpreparing their students to shoulder organiza-tional leadership, the present analysis also im-


plies that they should strengthen the capacities oftheir students to make leadership decisions withambiguous authority in work environments thatare often demanding and frequently changing.Given the impact of increasingly global productand equity markets on company operations, thepressures for quality decision making are likely tointensify in the future, making preparation in lead-ership decisions even more important for manage-ment education in the years ahead (Useem, 2004a).Studies of MBA programs reveal that well-designed curriculums can improve a host of man-agement capabilities (e.g., Boyatzis, Stubbs, & Tay-lor, 2002). The experience of those during the SouthCanyon fire suggests the seminal importance oflearning to make quality leadership decisions inthose programs.

Many business schools have establishedcourses on leadership during the past decade, andthe South Canyon fire also points to going beyondfostering leadership attributes to cultivating thecapacities for making leadership decisions. How tothink strategically, mobilize resources, executestrategy, and act selflessly are often at the core ofsuch courses (including one offered by the firstauthor to executive MBA students). What is alsorequired is a focus on learning how to make goodand timely decisions when in a position of respon-sibility and facing discrete, tangible, and realisticopportunities to take the enterprise in one direc-tion or another. Drawing upon one of the guidingobjectives of the Wildland Fire Leadership Devel-opment Program, the business school courseworkshould be designed to prepare future managers to“make sound and timely decisions.” Cases andexercises would help students enhance their situ-ational awareness and clear-mindedness understress, know when and how to clarify their ownauthority and that of others, and master the art ofconveying strategic intent so that subordinatescan make good and timely decisions of their own.

In keeping with the Wildland Fire LeadershipDevelopment program, business schools wouldalso be advised to reinforce their classroom les-sons on leadership decision making with out-of-classroom learning experiences. Even the best de-signed classroom-based leadership curriculumsdo not succeed as well as they should in fullyengendering the main capacities that constituteleadership, and they may be particularly inade-quate for developing an ability to make good andtimely leadership decisions (Day, 2000). Akin to theconcept of the staff ride for firefighters, the out-of-classroom learning experiences would place fu-ture managers “in the shoes of the decision makers

on a historical incident in order to learn for thefuture.”

Out-of-classroom learning experiences for busi-ness students need not be limited to company in-cidents. By examining decision moments in venuesranging from Civil War battlefields to Himalayanmountain ascents and even the South Canyon firezone, future managers should gain improved in-sight into the perils and pluses of decision makingin their own settings. Akin to the staff ride on StormKing Mountain that firefighters now use and ac-tion-learning projects that many companies havecome to use for much the same purpose (Dolitch &Noel, 1998), leadership ventures should be de-signed to be tangible and engaging—and thus in-structively memorable (Hirst, Mann, Bainc, Pirola-Merlod, & Richvera, 2004)—and to combineanalytic ideas from the classroom with tangibleapplications in the field, and thus, be analyticallyinformed as well (several such ventures and theirlearning value for leadership decision making aredescribed in Useem, Davidson, & Wittenberg, 2005).

LEADERSHIP DECISIONS MAKE A DIFFERENCE

The thrust of our analysis might seem to go againstthe conclusions of investigations of other accidentsand disasters in which the source was seen as lessthe results of pinpointed suboptimal individual de-cisions than the product of error chains, impairedsensemaking, or bureaucratic complexity (e.g., Al-lison, 1971; Perrow, 1999; Vaughan, 1996; ColumbiaAccident Investigation Board, 2003). In his analysisof the accidental downing of two U.S. helicoptersby two U.S. F15 fighters over northern Iraq in 1994,for instance, Snook (2000: 207) concluded that thedowning should be viewed as “the behavior ofactors struggling to make sense, rather than asrational attempts to decide.”

From a distant perspective, the tragic turn ofevents during the South Canyon fire should beviewed in much the same way: Setting the stagefor Don Mackey’s decisions were larger contribut-ing factors not of his own making, including thedelayed and inadequate assignment of firefight-ing resources to the South Canyon blaze, the fail-ure of the engaged federal agencies to unequivo-cally designate an incident commander and todeliver a critical weather warning, and the ab-sence of sufficient training in the essence of inci-dent leadership not only for himself but also forothers.

Within that set stage, however, we have none-theless found it useful to examine the leadershipdecisions in their own right since they made theirown independent contribution to the outcome. Don


Mackey was dealt a poor hand by the governmentagency that dispatched him to the mountain, butsome of his decisions on the slopes helped makeup for the insufficient training and support that hehad been rendered by the government. His eighthand ninth decisions at 4:06 p.m. to evacuate theWest Flank fireline and Lunchspot Ridge, and histenth decision to run across the fireline rather thanup the ridge, no doubt contributed to the survival ofas many as 14 firefighters. Conversely, his otherleadership decisions that could have gone a differ-ent way—to avoid a downhill fireline, to seek aweather forecast, to resolve the ambiguity of whowas in charge—contributed to the entrapment of atleast 12 who did not survive.

Whether on the slope of Storm King Mountain, inthe boardroom of Enron, or on the battlefield ofAntietam, readers will find it analytically useful tofocus on specific leadership decisions and theirimpact on the goals of the enterprise. We shouldseek to learn more about how leadership decisionswithin such settings can be critical to the suc-cesses or setbacks of an organization—and howthey can be improved.

Organizations would themselves benefit from animproved appreciation for what best contributes toquality leadership decision making within theirwalls. With better understanding, programs can bebetter designed to help incident commanders andenterprise managers better avert the unforced er-rors to which they are prone when working understress and ambiguity. While not overlooking theimportance of error chains, reduced sensemaking,and organizational bottlenecks, the U.S. agenciesresponsible for firefighting are now seeking to pre-pare their field leaders for quality decisions in thechallenging moments yet to come—and other or-ganizations would do well to consider the same.

REFERENCES

Allison, G. T. 1971. Essence of decision: Explaining the Cubanmissile crisis. New York: HarperCollins.

Amason, A. C. 1996. Distinguishing the effects of functional anddysfunctional conflict on strategic decision making: Resolv-ing a paradox for top management teams. Academy ofManagement Journal, 39: 123–148.

Barnard, C. I. 1968. The functions of the executive. Cambridge,MA: Harvard University Press.

Bazerman, M. 2002. Judgment in managerial decision making,(5th ed.) New York: Wiley.

Bennis, W. 2003. On becoming a leader (expanded ed.). NewYork: Perseus.

Bossidy, L., & Charan, R. 2002. Execution: The discipline of get-ting things done. New York: Crown Business/RandomHouse.

Boyatzis, R. E., Stubbs, E. C., & Taylor, S. N. 2002. Learningcognitive and emotional intelligence competencies throughgraduate management education. Academy of Manage-ment Learning and Education, 1: 150–162.

Butler, B. W., Bartlette, R. A., Bradshaw, L. S., Cohen, J. D.,Andrews, P. L., Putnam, T., & Mangan, R. J. 1998. Fire be-havior associated with the 1994 South Canyon Fire on StormKing Mountain. Colorado: U.S. Forest Service (http://www.fs.fed.us/rm/pubs/SouthCanyon).

Carter, C. B., & Lorsch, J. W. 2004. Back to the drawing board:Designing corporate boards for a complex world. Boston,MA: Harvard Business School Press.

Charlier, M. 1994. Nature’s ambush: How two firefighters sur-vived the tragedy of South Canyon Blaze. Wall Street Jour-nal, August 22: A1, 6.

Chiles, J. R. 2001. Inviting disaster: Lessons from the edge oftechnology. New York: Harper Business.

Collins, J. 2001. Good to great: Why some companies make theleap . . . and others don’t. New York: HarperBusiness.

Columbia Accident Investigation Board, 2003. Report. Washing-ton, DC: National Aeronautics and Space Administrationand Government Printing Office.

Conger, J. A., & Fulmer, R. M. 2003. Developing your leadershippipeline. Harvard Business Review, December: 76–84.

Cook, J. 2002. Leadership toolbox: From wildland firefighters.Wharton Leadership Digest, October (leadership.whar-ton.upenn.edu/digest/10-02.shtml).

Day, D. V. 2000. Leadership development: A review in context.Leadership Quarterly, 11: 581–613.

Dean, J. W., Jr., & Sharfman, M. P. 1996. Does decision processmatter? A study of strategic decision-making effectiveness.Academy of Management Journal, 39: 368–396.

Dotlich, D. L, & Noel, J. L. 1998. Action learning: How the world’stop companies are recreating their leaders and themselves.San Francisco: Jossey-Bass.

Donaldson, G. 1983. Decision making at the top: The shaping ofstrategic direction. New York: Basic Books.

Druskat, V. U., & Wheeler, J. V. 2003. Managing from the bound-ary: The effective leadership of self-managing work teams.Academy of Management Journal, 46: 435–457.

Edmondson, A. C., Roberto, M. A., & Watkins, M. D. 2003. Adynamic model of top management team effectiveness:Managing unstructured task streams. Leadership Quarterly14: 297–325.

Eisenhardt, K. M. 1989. Making fast strategic decisions in high-velocity environments. Academy of Management Journal,32: 543–576.

Fiedler, F. E. 1992. Time-based measures of leadership experi-ence and organizational performance: A review of researchand a preliminary model. Leadership Quarterly, 3 (Spring):4–23.

Finkelstein, S., & Hambrick, D. C. 1996. Strategic leadership: Topexecutives and their effects on organizations. St. Paul, MN:West Publishing.

Finucane, M. L., Alhakami, A., Slovic, P., & Johnson, S. M. 2000.The affect heuristic in judgments of risks and benefits.Journal of Behavioral Decision Making, 13: 1–17.

Freedman, D. H. 2000. Corps business: The 30 management prin-ciples of the U.S. Marines. Harper Business.


Freeman, D. S. 1998. Lee’s lieutenants: A study in command. NewYork: Scribner.

Gandossy, R., & Effron, M. 2004. Leading the way: Three truthsfrom the top companies for leaders. New York: Wiley.

Gardner, J. 1993. On leadership. New York: Free Press.

Gawande, A. 2002. Complications: A surgeon’s notes on an im-perfect science. New York: Henry Holt.

George, B. 2003. Authentic leadership: Rediscovering the secretsto creating lasting value. San Francisco: Jossey-Bass.

Gilbert, D. T. 2002. Inferential correction. In Thomas Gilovich,Dale Griffin, & Daniel Kahneman, Eds., Heuristics and bi-ases: The psychology of intuitive thought. New York: Cam-bridge University Press.

Hammond, K. R. 2000. Judgment under duress. New York: OxfordUniversity Press.

Hirst, G., Mann, L., Bainc, P., Pirola-Merlod, A., & Richvera, A.2004. Learning to lead: The development and testing of amodel of leadership learning. The Leadership Quarterly, 15:311–327.

Incident Command System. 1994. National training curriculum:Organization overview. Boise, ID: National Interagency FireCenter.

Incident Operations Standards Working Team, National Wild-fire Coordinating Group. 2002. Operational LeadershipGuide. In Incident response pocket guide. Boise, ID: Na-tional Interagency Fire Center.

Janis, I., & Mann, I. 1977. Decision making: A psychologicalanalysis of conflict, choice, and commitment. New York:Free Press.

Judge, W. Q., & Miller, A. 1991. Antecedents and outcomes ofdecision speed in different environmental contexts. Acad-emy of Management Journal, 34: 449–463.

Kahneman, D. 2003. Maps of bounded rationality: Psychology forbehavioral economics. American Economic Review, 93:1449–1475.

Klein, G. 1998. Sources of power: How people make decisions.Cambridge, MA: MIT Press.

Klein, G. 2003. Intuition at work: Why developing your gut in-stincts will make you better at what you do. New York:Currency Doubleday.

Kunreuther, H. C. 2001. Protective decisions: Fear or prudence.In Stephen J. Hoch and Howard C. Kunreuther, (Eds.), Whar-ton on making decisions, New York: Wiley.

Langley, A., Mintzberg, H., Pitcher, P., Posada, E., & Saint-Macary, J. 1995. Opening up decision making: The viewfrom the black stool. Organization Science, 6: 260–279.

Lorsch, J. W. 2002. Board failed in its duty, but others must learn.Boston Globe, April 27: E3.

Maclean, J. N. 1999. Fire on the mountain: The true story of theSouth Canyon Fire. New York: William Morrow & Co.

Maclean, J. N. 2003. Fire and ashes: On the front lines of Amer-ican wildfire. New York: Henry Holt & Co.

Maclean, N. 1990. Young men and fire. Chicago, IL: University ofChicago Press.

McKenzie, R. B., & Lee, D. R. 1998. Managing through incentives:How to develop a more collaborative, productive, & profit-able organization. New York: Oxford University Press.

Metcalfe, J. 1998. Cognitive optimism: Self-deception or memo-

ry-based processing heuristics. Personality and Social Psy-chology Review, 2: 100–110.

Meyer, J. P. 2003. Four territories of experience: A developmentalaction inquiry approach to outdoor-adventure experientiallearning. Academy of Management Learning & Education,2: 352–263.

National Wildfire Coordinating Group. 2004. Fireline handbook.Boise, ID: National Interagency Fire Center.

Perrow, C. 1999. Normal accidents: Living with high-risk tech-nologies. Princeton, NJ: Princeton University Press.

Powers, W. C., Jr., Troubh, R. S., & Winokur, H.S., Jr. 2002. Reportof investigation. Special Investigative Committee of theBoard of Directors of Enron Corporation. Houston, TX: EnronCorporation.

Putnam T. 1995. The collapse of decision making and organiza-tional structure on Storm King Mountain. Missoula Technol-ogy and Development Center, U.S. Forest Service.

Pyne, S. J. 1997. Fire in America: A cultural history of wildlandand rural fire. Seattle, WA: University of Washington Press.

Quintanar, R. 2003. “Staff Rides,” memo of July 29, 2003, to“Forest Supervisors and Directors,” U.S. Forest Service.

Ready, D. A., & Conger, J. A. 2003. Why leadership-developmentefforts fail. Sloan Management Review, Spring: 83–88.

Redding Interagency Hotshot Crewmember Report. 2003. SouthCanyon Fire Staff Ride, Glenwood Springs, Colorado, May25–29, 2003. Redding, CA: Author.

Roberto, M. A. 2001. Making difficult decisions in turbulenttimes. Ivey Business Journal, May/June: 15–20.

Roberto, M. A., & Ferlins, E. M. 2003a. Storm King Mountain(case). Boston, MA: Harvard Business School.

Roberto, M. A., & Ferlins, E. M. 2003b. Fire at Mann Gulch (case).Boston, MA: Harvard Business School.

Ruggero, E. 2001. Duty first: West Point and the making of Amer-ican leaders. New York: HarperCollins.

Russo, J. E., & Schoemaker, P. J. H. 1990. Decision traps: Tenbarriers to brilliant decision-making and how to overcomethem. New York: Simon & Schuster.

Russo, J. E., & Schoemaker, P. J. H. 2002. Winning decisions:Getting it right the first time. New York: Currency Double-day.

Salka, J., with J. Neville. 2004. First in, last out: Leadershiplessons from the New York Fire Department. New York:Portfolio/Penguin.

Sears, S W. 1999. George B. McClellan: The young Napoleon.New York: Da Capo Press.

Sears, S W. 2003. Gettysburg. Boston, MA: Houghton Mifflin.

Simon, M., & Houghton, S. M. 2003. The relationship betweenoverconfidence and the introduction of risky products: Evi-dence from a field study. Academy of Management Journal,46: 139–149.

Snook, S. A. 2000. Friendly fire: The Accidental shootdown of U.S.Black Hawks over Northern Iraq. Princeton, NJ: PrincetonUniversity Press.

Sonnenfeld, J. A. 2002. What makes great boards great. HarvardBusiness Review, September: 106–113.

South Canyon Fire Accident Investigation Team. 1994. SouthCanyon Fire investigation of the 14 fatalities that occurredon July 6, 1994 near Glenwood Springs, Colorado. Washing-


ton, DC: U.S. Forest Service and Bureau of Land Manage-ment.

Sutton, L. 2002. Leadership on the line: Wildland firefighters.Wharton Leadership Digest, January (leadership.wharton.upenn.edu/digest/01-02.shtml).

Thoele, M. 1995. Fire line: The summer battles of the west.Golden, CO: Fulcrum Publishing.

Tichy, N. M. 1997. The leadership engine: How winning compa-nies build leaders at every level. Harper Business.

TriData Corporation. 1998. The wildland firefighter safetyawareness study. Arlington, VA: TriData Corporation.

U.S. Marine Corps University. 2004. Conduct of the staff ride.www.mcu.usmc.mil/mcu/Staff-Ride/Conduct.htm.

U.S. Military Academy. 2004. Staff rides. www.dean.usma.edu/history/web03/staff%20rides%20site/sret%20pages/staff%20ride%20home03.htm.

U.S. Senate, Permanent Subcommittee on Investigations of theCommittee on Governmental Affairs. 2002. The role of theboard of directors in Enron’s collapse. Washington, DC: U.S.Senate (http://news.findlaw.com/hdocs/docs/enron/senpsi70802rpt.pdf).

Useem, M. 1998. The leadership moment: Nine true stories oftriumph and disaster and their lessons for us all. New York:Random House.

Useem, M. 2001. Leading up: How to lead your boss so you bothwin. New York: Random House.

Useem, M. 2003. Corporate governance is directors making de-cisions: Reforming the outward foundations for inside de-cision making. Journal of Management and Governance, 7:241–253.

Useem, M. 2004a. Corporate governance and leadership in aglobalizing equity market. In Hubert Gatignon & John Kim-berly, (Eds.), The alliance on globalizing: Strategies forbuilding successful businesses. New York: Cambridge Uni-versity Press.

Useem, M. 2004b. The essence of leading and governing isdeciding. In Robert Gandossy & Jeffrey Sonnenfeld (Eds.),Leadership and governance from the inside out. New York:Wiley.

Useem, M., Davidson, M., & Wittenberg, E. 2005. Leadershipdevelopment beyond the classroom: The power of leader-ship ventures to drive home the essence of decision mak-ing. International Journal of Leadership Education, 1: 159–178.

Vaughan, D. 1996. The Challenger launch decision: Risky tech-nology, culture, and deviance at NASA. Chicago, IL: Uni-versity of Chicago Press.

Weick, K. E. 1993. The collapse of sensemaking in organizations:The Mann Gulch disaster. Administrative Science Quar-terly, 38: 628–652.

Weick, K. E. 1995. Sensemaking in organizations. ThousandOaks, CA: Sage Publications.

Weick, K. E. 1996. Drop your tools: An allegory for organizationalstudies. Administrative Science Quarterly, 41: 301–313.

Weick, K. E., & Roberts, K. H., 1993. Collective mind in organi-zations: Heedful interrelating on flight decks. Administra-tive Science Quarterly, 38: 357–381.

Wildland Fire Leadership Development Training Courses. 2004.www.fireleadership.gov/courses/courses.html

Wildland Fire Leadership Development Program. 2004. www.fireleadership.gov.

Wildland Fire Leadership Development Staff Ride Library. 2004.http://www.fireleadership.gov/toolbox/staffride/index.html.

Wildland Fire Leadership Development Values and Principles.2004. http://www.fireleadership.gov/values_principles.html.

Woodward, B. 2003. Bush at war. New York: Simon & Schuster.

Woodward B. 2004. Plan of attack. New York: Simon & Schuster.

Yukl, G. A. 1989. Managerial leadership: A review of theory andresearch. Journal of Management Development, 15: 251–289.

James R. Cook is the train-ing projects coordinator forthe U.S. Forest Service at theNational Interagency FireCenter in Boise, Idaho. Mr.Cook has worked in thewildland fire service since1975, including 18 years as aHotshot Crew superinten-dent. He has worked onmore than 450 fire assign-ments in 20 states. He hasbeen in his current job since1999 where his primary fo-cus has been working on anational level initiative to

establish a formal leadership development program for wild-land firefighters. He can be reached at [email protected].

Larry Sutton is the trainingunit leader for the U.S. Bu-reau of Land Management(BLM) at the National Inter-agency Fire Center. Mr. Sut-ton has been involved infirefighting since 1979, re-sponding to hundreds ofwildland fires. He is a mem-ber of a Type 1 Incident Man-agement Team, one of 17 suchteams organized by the wild-land fire agencies to respondto emergency incidents nation-ally. For the past several years,

he has worked to build a leadership development program for wild-land firefighters. He can be contacted at [email protected].


Michael Useem is professorof management and directorof the Center for Leadershipand Change at the WhartonSchool of the University ofPennsylvania. He is coau-thor and coeditor of UpwardBound (2003), and author ofLeading Up (2001) and TheLeadership Moment (1998).In collaboration with theU.S. Forest Service, Bureauof Land Management, Na-tional Interagency Fire Cen-ter, and U.S. Marine Corps,he has co-organized with

James Cook and Larry Sutton learning visits to the site of the1949 fire in Mann Gulch, Montana and the 1994 South Canyonfire on Storm King Mountain, Colorado. He can be reached [email protected].


Documents

Academy of Management Learning & Education Developing