United States Department of Agriculture CCrewrew CCohesion,ohesion, Forest Service Technology & WWildlandildland FFireire TTransition,ransition, Development Program 5100 Fire aandnd FFatalitiesatalities February 2002 0251-2809-MTDC

Jon Driessen, Ph.D. Sociologist

USDA Forest Service Technology and Development Program Missoula, MT

TE02P16—Fire and Aviation Management Technical Services

February 2002

The Forest Service, United States Department of Agriculture (USDA), has developed this information for the guidance of its employees, its contractors, and its cooperating Federal and State agencies, and is not responsible for the interpretation or use of this information by anyone except its own employees. The use of trade, firm, or corporation names in this document is for the information and convenience of the reader, and does not constitute an endorsement by the Department of any product or service to the exclusion of others that may be suitable.

The U.S. Department of Agriculture (USDA) prohibits discrimination in all its programs and activities on the basis of race, color, national origin, sex, religion, age, disability, political beliefs, sexual orientation, or marital or family status. (Not all prohibited bases apply to all programs.) Persons with disabilities who require alternative means for communication of program information (Braille, large print, audiotape, etc.) should contact USDA’s TARGET Center at (202) 720-2600 (voice and TDD). To file a complaint of discrimination, write USDA, Director, Office of Civil Rights, Room 326-W, Whitten Building, 1400 Independence Avenue, SW, Wash- ington, D.C. 20250Ð9410, or call (202) 720-5964 (voice and TDD). USDA is an equal opportunity provider and employer. 1 Contents

Introduction ______3 Part I—Crew Cohesion at the Mann Gulch, South Canyon, and Thirtymile Fires ______4 Cohesion and Fire Crews—A Long-Standing Concern ______4 Intracrew and Intercrew Cohesion Problems in Three Firefighting Tragedies ______4 The Mann Gulch Fire—Failure in Intracrew Cohesion ______4 The South Canyon Fire—Failure in Intercrew Cohesion ______5 The Thirtymile Fire—Failure of Intracrew and Intercrew Cohesion ______5 Come Down Out of the Rocks—That’s Not the Place to Be ______6 More Intercrew Cohesion Problems at the Thirtymile Fire ______7

Part II—The Discovery of Cohesion and the Welfare of People ___ 7 Cohesion—A Definition ______7 Suicide—The Problem of Cohesion in Modern Human Groups ______7 Accidents and Cohesion in Forest Service Crews______7

Part III—Firefighter Fatalities and Transition Fires______8 Types of Wildland Fires ______8 Stages in Wildland Firefighting ______8 Initial Attack Stage ______8 Extended Attack Stage ______8 The Ambiguous and Dangerous Transition Stage—Fire and People ______8 Transition and Fatalities—The Munson and Mangan Studies ______9

Part IV—Future Implications ______10 Focus on the Money—Fires in Transition ______10 Creating Context—Developing a Classification of Transition Fires ______10 Writing a Common Playbook—Connecting Context and Action ______10 Finding the Unwritten Typology and Unwritten Playbook ______11 Simulation Training ______11 Serendipitous Outcomes ______11

Summary ______12 References ______13

Appendix A—Some Impressions of Different Wildland Fire Crews and Their Cohesion ______14

Appendix B—Fire Crew Cohesion: Previous Recommendations______15

Acknowledgments

I called on Chuck Whitlock to help me write this paper. I bounced many of my ideas off him because of his valuable knowledge of the firefighting community. A colleague, Raymond Gold, reviewed drafts of the paper and made suggestions that were to the point about sharing rules and game playing. I also used data from fire investigation reports, social science publications, popular books, fire handbooks, my own reports on Forest Service work crews, and my observations while on the investigation team at the Thirty- mile Fire. Several insights in the paper arose from my ongoing involvement in a project at the Missoula Technology and Development Center, Avoiding Entrapment in Wildfire.

From our publications shop in the center, two additional people deserve special thanks. Bert Lindler patiently worked with me carefully critiquing and editing multiple drafts of the paper. Sara Lustgraaf rallied to meet the short deadline for the paper. She did a wonderful job with the typing and layout. 2 Introduction

ike wildfire, my research into the The concept of cohesion, or how fire paper discusses crew cohesion during topic of cohesion in fire crews crews are or are not bonded, is woven different stages of firefighting. I draw LLsought its own direction. The topic through all four parts of this paper. In special attention to the role crew took me into considerations of the safety part I, I look at the role of crew cohesion cohesion plays during the dangerous of firefighters and eventually into the in three tragic fires. In part II, I examine fire transition period. In the final part, I connection between crew cohesion and how the concept of group cohesion discuss some implications the paper fire fatalities. I relied on numerous sources, became central in sociological studies has for future studies and for wildland which I have acknowledged, to help of human groups. I also discuss my firefighter training, concluding with a develop this paper. studies of cohesion and safety in Forest brief summary. Service field crews. The third part of the

3 Part I—Crew Cohesion at the Mann Gulch, South Canyon, and Thirtymile Fires

Cohesion and Fire The Mann Gulch Fire— Failure in Intracrew Cohesion Crews—A Long- As the entity of a crew dis- Standing Concern In South Canyon Revisited (1995), Weick solved, it is not surprising that compared the Mann Gulch and South the final command from the “crew” Experts who have studied why people Canyon Fire disasters. He points to the leader to jump into an escape fire died fighting wildfires have long noted a lack of crew cohesion in both fires. The was heard not as a legitimate connection between fire crew cohesion Mann Gulch crew collapse is a clear order but as the ravings of some- and fatalities. Let’s look at this connection example of the consequences of a lack one who had “gone nuts.” in the context of three firefighting of intracrew cohesion. Weick wrote: tragedies.

Since the crew did not know The South Canyon Fire— each other well, since Dodge knew Failure in Intercrew Cohesion only three of them, since several were on their first jump, and since The South Canyon fire illustrates the Intracrew and Inter- Dodge himself was rusty on lead- absence of intercrew cohesion. It would crew Cohesion Problems ing a crew, it was imperative to be hard to argue that the Prineville build some common understand- Hotshots lacked intracrew cohesion. in Three Firefighting ings and common action into Smokejumpers and helitack crews at the Tragedies this assortment of strangers. South Canyon fire also probably had That didn’t happen. cohesion within their own ranks. Strong Intracrew cohesion is the cohesion within intercrew cohesion did not exist among a single crew fighting a fire. Intercrew the three crews fighting the fire. Quoting cohesion refers to cohesion between Weick, quoting from Norman Maclean’s from the report of the incident investiga- different crews fighting the same fire as book, Young Men and Fire (1992), points tion team (USDA Forest Service 1995): well as their cohesion with fire managers. out that the smokejumpers at Mann Gulch were probably not a crew at all. They were In this paper I will focus on intracrew and merely “loosely coupled” firefighters with- intercrew cohesion of the ground crews. out any intracrew cohesion. Weick wrote: Cohesion with fire managers is critical during fire transition, but it will not be considered here. I will use the Mann Gulch, South Canyon, and Thirtymile Fires to illustrate the dangers when Intracrew Breakdown of intracrew and intercrew cohesion are not cohesion intracrew cohesion established before crews chase fires that blow up. Until recently, few of my studies have focused on wildland firefighting crews. I will blend my ideas on cohesion with some brilliant insights provided by Karl Weick, who wrote about the collapse of sensemaking in firefighting crews (Weick 1993, 1995).

4 Part I—Crew Cohesion at the Mann Gulch, South Canyon, and Thirtymile Fires

As is typical in extended Intercrew cohesion attack situations, firefighting groups arrived on the fire at intervals from dispersed loca- tions and blended into the existing organization.

Weick focused on the word “blended,” when he wrote:

The key word there is “blended.” Blending sounds like something that occurs automati- Breakdown of intercrew cohesion cally, not something that people work at. Many would say it is hard to blend into an “existing organization” if that organization itself is invisible, as was the case for some people at South Canyon. Some people trying to blend did not know who the Incident Commander was, or which radio traffic had the force of authority, or what the suppression strategy was, since it seemed counterintuitive.

The Thirtymile Fire—Failure of Intracrew and Intercrew Cohesion

As one of the investigators of the Thirty- mile Fire, I was able to make firsthand observations of crew cohesion. I found problems with both intracrew and inter- crew cohesion was apparent when I people with whom they shared intracrew crew cohesion. The Northwest Regulars found that firefighters from each of the cohesion. This clustering into cohesive No. 6, the type II crew that was entrapped, two districts drew close to each other crews explains why the four people who was made up of a mixture of two crews while they were in the “safety zone” died on the rocks belonged to the Naches from two different districts. When these waiting for the fire to pass. crew. The crewmembers from the Lake/ crews were configured into a single type Leavenworth Ranger District were on the II crew, they lacked intracrew cohesion. For the most part, the crew from the Lake/ road with their familiar crew leader. Leavenworth Ranger District gathered on The two crews had not trained or worked the road. The Naches crew, except for one In this tragic incident, intracrew cohesion together. For the most part, they were person, gathered together on the rocks probably saved the lives of the Lake/ two crews of strangers caught up fighting above the road. As the fire approached, Leavenworth crew. They “hung together” a fire that suddenly escaped. The type people sought security; they wanted to be and deployed their shelters on the road. II crew boss/incident commander was close to their trusted crew leaders and However, the same intracrew cohesion unfamiliar to the Naches crew. Home fellow crewmembers. These were the spelled doom for the Naches crew.

5 Part I—Crew Cohesion at the Mann Gulch, South Canyon, and Thirtymile Fires

Come Down Out of the of control and order by falling back into Naches crew leader had more legitimacy Rocks—That’s Not the Place their intracohesive groups and taking for the Naches crew than the formally to Be direction from their home crew leaders. assigned power of the overall type II Weick wrote about it this way: crew boss. The Naches crew members and their leader followed the basic social Let me try to explain why I feel the squad rule that governs people during anomic boss from the Naches crew did not comply “Collegial authority patterns moments (when rules governing sense- with “orders” from the crew boss who overlay bureaucratic ones as the making and action collapse). During such reportedly told him three times: “Come tempo of operations increase. times, people naturally seek to be in their down out of rocks. That’s not the place Formal rank and status decline intracohesive groups with their trusted to be.” As the type II crew waited in the as a reason for obedience.” leaders. safety zone, they had collective uncer- tainty about what whether to deploy on the road or on the rocks. Social psychologist, The Lake/Leavenworth crew boss, in Karl Weick (1993), refers to such periods addition to being the incident commander, of extreme uncertainty as “cosmology More Intercrew Cohesion was also the overall crew boss formally episodes…when people suddenly, and Problems at the Thirtymile in charge of both crews making up the deeply feel that the universe is no longer Fire type II crew. While waiting for the fire to a rational, orderly system. What makes pass, the Lake/Leavenworth crew stayed such an episode so shattering is that both Engine Crew No. 704 and the Entiat inter- cohesive and remained near their crew the sense of what is occurring and the agency hotshot crew were also fighting boss on the road. However, in this situa- means to rebuild that sense collapse.” the fire. There can be little doubt the tion, the crew boss’ formal authority over engine and hotshot crews had high levels the Naches crew did not carry much Weick noted that when groups are con- of intracrew cohesion. However, the weight. The members of the Naches crew fronted by anomie (a collapse of social investigation report shows all sorts of waited together on the rocks above the stability) during “cosmological episodes” confusion regarding failures to blend road and died with their natural intracrew they will always seek to restore a sense their efforts into a single intercohesive leader. The collegial authority of the firefighting unit with the type II crew.

6 Part II—The Discovery of Cohesion and the Welfare of People

Cohesion—A Definition Suicide—The Problem work practices that fostered their cohesion. of Cohesion in Modern Generally they talked about the importance When the word cohesion is used in rela- of staying focused on good production tion to groups, people are describing how Human Groups and how this focus depended on physical closely tied together they are as a group. fitness, work skills, safety awareness, But the meaning of cohesion extends far At the turn of the 20th century, French interpersonal harmony, and good beyond this. People in cohesive groups sociologist Emile Durkheim conducted supervision. speak openly of themselves as “a little a study of suicide (Durkheim 1897). He family.” They often talk about their cohe- concluded that suicide rates vary inversely Perhaps the most important discovery I sion in spiritual and reverential ways. with the degree of social integration made in these studies is this: Members of military, sports, and work (cohesion) of people’s groups. In other teams may use the word “cohesion” to words, people belonging to groups with describe a kind of intensified spiritual low cohesion had higher suicide rates Crew cohesion is “made” by state of interpersonal connectedness or than those who belonged to highly cohe- individual workers themselves a special group “chemistry.” When people sive groups. He used the word “anomic,” when they establish agreements experience cohesion in the groups they meaning “without rules” to describe groups about the rules that govern a belong to, they speak as if they have been with weak cohesion. His landmark study host of their day-to-day work blessed with a special kind of strength laid the groundwork for future sociological practices. that enables them to overcome great studies of cohesion in different kinds of obstacles. The special strength provided groups. by cohesion has not gone unnoticed by I found that members of cohesive crews sociologists. talk frankly with one another about their ongoing expectations. These expectations Sociologists have long used the idea of govern such things as work pace, rest cohesion in studies of a variety of small periods, decisionmaking, humor, warnings groups. These studies have focused on Accidents and of danger, requests for help, assistance cohesion in families, military units, sports Cohesion in Forest for fellow crewmembers, complaints, teams, neighborhoods, church congre- sharing food, and other practical matters gations, labor unions, street gangs, and Service Crews that bear directly on maintaining their especially small work groups. In this paper, cohesion. Cohesion, and the protection I want to connect the idea of cohesion to In my studies of Forest Service field crews it affords individual workers, comes about wildland firefighting crews. Before I do (Driessen 1986, 1996), I discovered that only after crews have tested and negoti- so, it is important to discuss how cohesion a modified version of Durkheim’s propo- ated acceptable norms governing their became such a central concept in socio- sition applied to accident rates. Accidents work practices. It takes time for this logical studies. in field crews were inversely correlated cohesion to develop. In my studies, I with the cohesion in the crews. In other found it takes from 6 to 8 weeks for words, the greater the crew cohesion, the individual seasonal workers to “click” into fewer the accidents. People working in crews. When this happened, individual cohesive field crews were fully aware that workers bonded into intracohesive crews. their cohesion helped protect them from They were filled with pride about their dangers inherent in the work. I listened production and trusted one another like to workers and their supervisors describe a “little family” (Driessen 1986, 1996).

7 Part III—Firefighter Fatalities and Transition Fires

ased on my recent work on fire- Initial Attack Stage At this critical transition period when the related projects, I have some fire may be expanding, often rapidly, type BBimpressions about the levels of The Fireline Handbook describes initial I crews and single resources can be cohesion in different types of fire crews. attack as the action taken by resources requested to fight the fire. More type II My impressions are in appendix A. These (people and equipment) that are the first crews can also be ordered. impressions can be used as hypotheses to arrive at the incident. All wildfires that to guide future studies of cohesion in the are controlled by suppression forces different kinds of fire crews. For now, I undergo initial attack. The number and want to focus on the importance of fire type of resources responding to the initial The Ambiguous and crew cohesion at the stage in firefighting attack varies depending on the fire danger, Dangerous Transition when most fatalities happen. To do this, fuel type, values at risk, and other factors. Stage—Fire and People I need to briefly discuss different types of fires and different stages in fighting During initial attack, the fires are relatively The “transition stage” is confusing because fires. small, probably type IV or V. District crews the fire community uses the word “tran- will probably be assigned to suppress a sition” in two senses. First, transition small fire. It is not unusual for the initial refers to a time when a fire is changing attack fire crew to have the support of a from a small, type IV or V fire to a much single resource such as a dozer, engine larger type III fire. In ordinary language, crew, or some sort of aircraft, (such as a this means the fire has grown big. Some Types of Wildland Fires helicopter with a water bucket). Smoke- fires grow big suddenly. Firefighters use jumpers are usually assigned to carry different words to describe this moment: According to the Fireline Handbook out initial attack on small fires in remote blowing up, taking off, losing control, or (National Wildfire Coordinating Group locations. As fires grow in complexity and making a run. During transition, the fire 1998), “typing fires provides managers size, type I crews, type II crews, contract has quickly expanded beyond the capa- with additional information in selecting crews, and single resources can be city of the resources that were initially the best resource for the task.” Wildfire assigned (see appendix A for a description assigned to control it. typing provides managers a context for of the types of crews). determining the types of crews to assign The Fireline Handbook describes a fire to fires. Fires are identified as type I, II, transition stage this way: III, IV, and V. The lower the number, the larger and more complex the fire. In other Extended Attack Stage words, type I and II fires are large and Transition from an Initial Attack complex. Type IV and V fires are smaller If type IV or V fires are not contained Incident to an Extended Attack and less complex. within a 24-hour work shift, or if the fire Incident. Early recognition by the grows rapidly, the suppression moves Initial Attack IC (Incident Com- from “initial attack” to “extended attack.” mander) that the initial attack The Fireline Handbook describes extended forces will not control a fire is attack as: important. As soon as the Initial Attack IC recognizes that additional Stages in Wildland resources are needed or knows Firefighting …a wildfire that has not been additional forces are enroute, the contained/controlled by the initial IC may need to withdraw from attack forces and additional fire- direct fireline suppression and Each fire is also identified based on the fighting resources are arriving, must prepare for the transition to stage of attack. The stages are identified enroute, or being ordered by the the Extended Attack. as initial attack, extended attack, and initial attack commander. An transition. extended attack fits into the Type III incident as regards complexity.

8 Part III—Firefighter Fatalities and Transition Fires

The second meaning of transition is a study, Wildland Firefighter Entrapments Mangan points to other reasons why transfer of the command established (Munson 2000), found that 43 percent of transition stages are so dangerous: during initial attack. As new crews and the firefighting fatalities occurred during resources are assigned for extended type IV or V fires. Twenty-nine percent attack, new crew bosses and a new inci- of the fatalities occurred during type III Most of the burnover events dent commander are placed in control. fires. These three types of fires accounted occurred during the initial attack They establish new strategy and tactics. for 72 percent of all wildland fire fatalities or extended initial-attack phase. The Fireline Handbook warns that at this from 1976 to 1999. Many of these deaths This is when the firefighters are transition stage, “fire crews may need to occurred when fires were rapidly tran- often involved in independent disengage from fighting the fire.” During sitioning or had transitioned into type III action, either as members of a initial attack, crews are relatively orga- fires. small crew, an engine, or even nized. When the fire makes a sudden as individuals. The higher levels transition, or blows up, there is an urgent Another recent study, Wildand Fire Fatal- of incident management teams need for organizational controls to make a ities in the United States (Mangan 1999), are not on the scene, communi- transition. Two recent studies have found also pointed to the danger of the transition cation may be confused, fire that when both the fire and the firefighting stage. Mangan wrote: weather and behavior conditions organization are in transition, fire crews may not be widely known or are at maximum risk. recognized, and the chain of The other dangerous phase of command may not be well a wildfire is the “transition phase,” established. when the fire has escaped initial Transition and Fatalities— attack efforts and higher level The Munson and Mangan incident management teams are My own observations as well as the ob- Studies being brought in. During this phase servations from other recent studies all some confusion may exist over focus attention on the likelihood of crew areas of responsibility; locations cohesion problems during the transition During the fire transition stage, the fire of different resources such as stage in wildfires and in prescribed burns has grown or is growing rapidly. However, crews, engines, or line overhead; that have gotten out of control. the resources, tactics, strategies, and or appropriate radio frequencies organizational structures are not yet in for tactical operations. This is often place to engage the fire in an extended the time the fire is exceeding the attack. Perhaps most importantly, inter- capability of the initial attack. crew cohesion is not in place either. How dangerous is this period? A recent

9 Part IV—Future Implications

Focus on the Money— quickly establish intercrew cohesion complete, it needs to be taught to all before starting an extended attack. The firefighters, especially crew leaders and Fires in Transition first step in creating this language is to fire managers. A shared language for develop a precise classification system the different types of fire transitions will When Jesse James was asked why he of the types of transition fires. allow firefighters to place their collective robbed banks, he supposedly replied: actions in a common context. “That’s where the money is!” If up to 72 As the word is now used in firefighting, percent of all fire fatalities occur during transition refers to very complex changing fires in transition, we should focus future fire events as well as to changing com- studies and training there. mand structures. The first task is to describe the properties of different types Previous recommendations to pay more of transition fires. From these descriptions, Writing a Common attention to problems in fire crew cohesion a classification scheme can be con- have been very general (Putnam 1995; structed to provide crew bosses with a Playbook—Connecting Tri-Data 1998; USDA Forest Service shared language. This shared language Context and Action 2001). See appendix B for a more detailed will permit them to distinguish between discussion of these recommendations. different types of transitions that are now An analogy to baseball may help. All Future work should concentrate on ways lumped into one word. With new classi- baseball players share a complex typology to rapidly establish crew cohesion within fications of transition fires, crew bosses of baseball situations—full count, bases the contexts of different types of fire will be able to talk precisely. Using a loaded with two outs, and so forth. These transitions. common language will give crew bosses types of baseball scenarios are like types a common understanding of how to act of fire transition situations. Not only do the and react together, depending on the players recognize the different situations, type of transition fire they identify. they also have a shared “playbook” that calls for specific strategy and tactics Because the present terminology used for depending on the different types of situ- Creating Context— fire transitions is not precise, individuals ations. The playbook dictates when to Developing a Classifi- are free to come up with their own inter- walk a player, when to steal, or where to pretation of the fire situation. These throw the ball. A common playbook for cation of Transition different interpretations make it extremely different types of transition fires would Fires difficult for different crews and their leaders specify what actions to take depending to know how to act and react collectively on the type of fire situation. When there Initially, studies need to focus on creating (with intercrew cohesion) before starting is a shared typology and a common a more precise language for fire crews, an extended attack. Studies can construct playbook, even strangers can quickly play crew bosses, and fire managers. This a specific typology of transition stages a game together. language is needed by crew bosses to in wildland fires. Once this typology is

10 Part IV—Future Implications

Finding the Unwritten scientists. Birds, rocks, and trees don’t fire crew bosses and managers. During have their own language. However, in training, it will be possible to determine Typology and cultural sciences, we can construct typ- who does or does not have the ability to Unwritten Playbook ologies of fire transition based on the identify types of transition fires and to language actually used by experienced practice using the playbook to establish My previous studies of field crews and wildland fire crews and their leaders. For intracrew and intercrew cohesion before their supervisors demonstrate that it is this project, we can record the natural starting extended attack (Klein 2001). reasonable to assume that the fire language typology actually used by crews community already has an unwritten and their leaders during different types classification of the types of transition of fire transitions (Schutz 1962). This fires as well as an unwritten playbook. typology can then be taught to all wildland Good crew bosses use these unwritten firefighters. The playbook can be devel- typologies of fire transitions and the oped by recording the actions (strategy Serendipitous unwritten playbook to guide their actions and tactics) used by good leaders during as they establish and maintain crew co- different types of fire transitions (Garfinkel Outcomes hesion when fires are in transition. Every 1967; Gumperz and Hymes 1972; Rose year there are thousands of different types 1992; and Driessen 1997). Studies of the sort described in this paper of transition wildfires. The vast majority almost always produce unexpected dis- of these situations are recognized and coveries. These discoveries will lead to the potential crew cohesion problems exciting new solutions to current problems associated with them are successfully (Reichenbach 1953; Glaser and Strauss managed. The successes come about 1967). As studies proceed, it will be because crews and their leaders are using Simulation Training important to take advantage of these dis- this unwritten typology to recognize fire coveries. I would expect a variety of new study proposals and policy recommenda- transitions and the unwritten playbook Simulation training for wildland firefighters to select strategies and tactics. tions to emerge during the course of the can be based on recognizing types of fire work. Many of the proposals will focus on transitions and using strategy and tactics In the physical sciences, classification ways to accelerate intracrew cohesion in identified in the playbook. This training the different types of fire crews discussed schemes or typologies have to be con- can also be designed to test future wildland structed using the technical language of in appendix A.

11 Summary

oor intercrew and intracrew cohe- of command, it is difficult for crew bosses field crews show that during the transition sion during the fire transition stage to create the minimally required inter- stage in fires, it is unrealistic to expect PPis a major factor in wildland fire crew cohesion before starting extended different fire crews and fire managers to fatalities. Cohesion problems that were attack. blend together into a single intercohesive quite different existed in crews on the unit. The time it takes to create this Mann Gulch, South Canyon, and Thirty- Sociologists know from their studies that cohesion could be greatly reduced by mile Fires. Because of the rapid growth cohesive groups are safer than groups focusing on the kinds of studies and of such fires and the associated transition with little or no cohesion. My studies of training I have suggested in this paper.

12 References

Anon. 2001. National interagency hotshot Klein, Gary. 2001. Sources of power. Reichenbach, Hans. 1953. The rise of crew operations guide. London: The MIT Press. scientific philosophy. Berkeley, CA: University of California Press. Driessen, Jon. 1986. Supervisor and the Maclean, Norman. 1992. Young men and work crew. Missoula, MT: United States fire. Chicago: University of Chicago Rose, Edward. 1992. The worulde [sic]. Department of Agriculture, Forest Press. Boulder, Co: The Waiting Room Press. Service, Missoula Technology and Development Center. Maclean, John N. 1999. Fire on the Rose, Edward. 1962. Two papers on the mountain. New York: William Morrow use of words as cultural units. Decisions, Driessen, Jon. 1996. Making a crew. and Company, Inc. Values and Groups. Vol. 2. Oxford, Missoula, MT: United States Depart- England: Pergamon Press. ment of Agriculture, Forest Service, Mangan, Richard. 1999. Wildland fire Missoula Technology and Development fatalities in the United States: 1990 to Schutz, Alfred. 1962. Collected papers. Center. 1998. Tech. Rep. 9951-2808-MTDC. Vol. 1. The Hague: Nartinus Nijhoff. Missoula, MT: United States Depart- Driessen, Jon. 1997. Raymond L. Gold’s ment of Agriculture, Forest Service, TriData Corp. 1998. TriData report, phase ethnographic method in sociology. Missoula Technology and Development III: Implementing cultural changes for Qualitative Inquiry. 3(4): 387Ð402. Center. safety. Arlington, VA: TriData Corp.

Driessen, Jon. 1997. Worldly interpreta- Munson, Steve. 2000. Wildland firefighter USDA Forest Service. 1995. Report of tions of a suspicious story. Ethnographic entrapments: 1976 to 1999. Tech. Rep. the South Canyon fire accident investi- Studies. 2: 3Ð15. 0051-2853-MTDC. Missoula, MT: United gation team. States Department of Agriculture, Driessen, Jon. 1997. A changing Forest Forest Service, Missoula Technology USDA Forest Service. 2001. Thirtymile Service work culture: training crew and Development Center. fire investigation report. leaders. Tech. Rep. 9767-2816-MTDC. Missoula, MT: United States Department National Interagency Fire Center. 1999. Weick, K. E. 1993. The collapse of sense- of Agriculture, Forest Service, Missoula National interagency mobilization guide. making in organizations: the Mann Technology and Development Center. NFES 2092. Boise, ID: National Inter- Gulch disaster. Administrative Science agency Fire Center. Quarterly. 16: 571Ð593. Durkheim, Emile. 1897. Le Suicide. Paris: F. Alcan. National Wildfire Coordinating Group. Weick, K. E. 1995. South Canyon revisi- 1998. NWCG fireline handbook. ted: lessons from high reliability Garfinkel, Harold. 1967. Studies in ethno- organizations. In: Putnam, Ted. 1995. methodology. Englewood Cliffs, NJ: Putnam, Ted. 1995. Findings from the Findings from the wildland firefighters Prentice-Hall. wildland firefighters human factors human factors workshop. Tech. Rep. workshop. Tech. Rep. 9551-2855- 9951-2855-MTDC. Missoula, MT: U.S. Glaser, Barry; Strauss, A. 1967. The MTDC. Missoula, MT: United States Department of Agriculture, Forest discovery of grounded theory. Chicago: Department of Agriculture, Forest Service, Missoula Technology and Aldine. Service, Missoula Technology and Development Center: 42Ð53. Development Center. Gumperz, John J.; Hymes, D. 1972. Directions in sociolinguistics. New York: Holt, Rinehart and Winston, Inc.

13 Appendix A—Some Impressions of Different Wildland Fire Crews and Their Cohesion

I will focus on type I and type II crews, assembled, many of the members may Engine Crews—The appendix in the engine crews, smokejumpers, contract not know one another. Based on this Fireline Handbook (National Wildfire crews, and single resources, as described social arrangement, one would not expect Coordinating Group 1998) describes the in the Fireline Handbook (National Wild- strong intracrew cohesion, including bonds makeup of engine crews. The number of fire Coordinating Group 1998). to the assigned crew boss. There may be people in an engine crew varies according strong intracrew cohesion within the to the size of the fire engine pump and separate district crews. However, until tank capacity. Most Forest Service engine Type I Crews (Hotshots)—The the different crews have worked together crews have about three members, with National Interagency Mobilization Guide for a while as a type II crew, their overall one person assigned as the captain, or (National Interagency Fire Center 1999) intracrew cohesion is probably low. Further engine boss. These small crews that describes type I crews as the primary study is needed to determine what it takes work together throughout a fire season firefighting force. Each crew must meet for type II crews formed from different may have the potential to reach the intra- the minimum standards specified in the crews to develop intracrew cohesion. crew cohesion levels found in hotshots. Interagency Hotshot Crew Operations However, cohesion among engine crew- Guide (Anon. 2001) These standards members also needs to be studied. specify working and training requirements, Smokejumpers—In the National Inter- experience levels, and the assignment agency Mobilization Guide, smokejumpers of permanent supervisors. Because of are described as a national resource. Single Resources—Another category these requirements, crewmembers are Their primary mission is initial attack. of people available to assist in fire sup- able to establish deep understandings Everyone I’ve talked to said it would be pression is identified as single resources. of each other as people, work out their wrong to think that smokejumpers work Some examples of single resources are own internal division of labor, and learn in cohesive crews. Different jumpers are dozer operators, aircraft pilots, and engine how to fight fire together as a tightly knit assigned for each initial attack. Because crews. Future studies of these different crew. In most instances, one would they work together in this fashion, smoke- types of single resources should focus expect high cohesion in hotshot crews. jumpers don’t have continuous and stable on successful work practices used when working relationships from one fire to the these resources attempt to blend in to next, as do hotshots. Smokejumpers feel the intercohesive firefighting crews, Type II Crews—The guide describes they belong to an elite firefighting unit with especially during wildfire transitions. type II crews as “any crew, which does a unique work culture, complete with the not meet experience, financing, training, symbolic trappings of uniform, rituals, and travel requirements of a type I (IHC) and lore. Membership in this community Contract Crews—The use of contract crew.” Type II crews are the same size as creates special bonds. However, these firefighting crews is relatively new in the type I crews, usually 18 to 20 firefighters. bonds are not to a single crew. Further wildland firefighting community. I have A temporarily assigned crew boss often study would be needed to answer the not talked to anyone about the social leads them. Smaller field crews from question: What are the communication organization or cohesion levels in these different districts are combined to make problems that smokejumpers face, and kinds of crews. Contract crew cohesion up an 18 to 20 person type II crew. These how do they manage these problems and how these crews integrate with other types of crews are assembled on an ad when they are required to quickly create types of firefighting crews remains to be hoc basis. The smaller crews brought intracrew cohesion when starting initial studied. together to form the type II crew often attacks? have not trained or worked together on previous fires or projects. When they are

14 Appendix B—Fire Crew Cohesion: Previous Recommendations

Recommendations From the Recommendations From the Management Evaluation Report Human Factors Workshop—A 5-day TriData Study—In 1998 the TriData of the Thirtymile Fire—The incident human factors workshop was held in Corp. completed a major study on the fire review board wrote about the problem of Missoula, MT, during June 1995. Experts safety culture. A report from the study, people who were unfamiliar with each in psychology, sociology, formal organiza- Phase III—Implementing Cultural Changes other working together: tions, fire safety, and wildland firefighting for Safety, made two specific recommen- were brought together. On the last day of dations about improving fire crew cohesion: ¥ There were a number of issues that the workshop, 21 recommendations were limited the development of crew cohe- formulated. Two recommendations focused ¥ Unit cohesion should be fostered and sion for the Northwest Regular No. 6 on improving cohesion in fire crews: attention given to developing good crew crew. These included: collateral duties dynamics. of command, fatigue, incident complex- ¥ Develop methods to speed up crew ity, lack of opportunity to work together, cohesion and work practices before ¥ Foster better crew cohesion, especially and management effectiveness. fireline assignments. among type II crews. These reports from workshops, studies, ¥ Organize more national, regional, and and fire fatality investigations all recom- local rendezvous where there is more mended working to increase the cohesion mixing of type I, type II, engine, and of wildland fire crews. helitack crews, fire management officers, incident management teams, and dis- patchers so they can share knowledge and discuss problems.

15 Library Card Driessen, Jon. 2002. Crew cohesion, hesion is a way of describing how closely the different types of transition fires to be wildland fire transition, and fatalities. Tech. people feel they are tied to a group. classified. This classification system would Rep. 0251-2809-MTDC. Missoula, MT: Studies of field crews in the Forest Service allow crew leaders and fire managers to U.S. Department of Agriculture, Forest have shown that the less cohesion crews communicate clearly when fires are Service, Missoula Technology and had, the more likely they were to be making the transition from relatively small Development Center. 16 electronic p. involved in accidents. Meanwhile, studies fires to relatively large fires. The studies of wildland firefighter fatalities during the would also identify tactics that successful Describes the role played by crew cohe- 1990s have shown that nearly three- crew leaders use to build the essential sion in the deaths of firefighters in three fourths of the fatalities occurred when fires level of crew cohesion when fires are in firefighting tragedies: the Mann Gulch were making the transition from relatively transition. Fire, the South Canyon Fire, and the small to relatively large fires, or shortly Thirtymile Fire. Two types of cohesion after they had become relatively large Keywords: firefighters, fire fighters, are involved, the cohesion within a crew fires. The transition is often sudden, when firefighting, fire fighting, group interaction, (intracrew cohesion) and the cohesion it is described as a “blow up.” Driessen groups, safety at work, social behavior, among crews (intercrew cohesion). Co- recommends studies that would allow sociology, sociology of work

About the Author Jon Driessen has worked on a number of has been production of a series of per- of sociology at the University of Montana. projects in the fire and safety programs sonal safety training video modules. Jon He now teaches part time at the university since joining the Missoula Technology received his Ph.D. in sociology from the in addition to working for MTDC. and Development Center as a faculty University of Colorado in 1969. Until last affiliate in 1977. His most recent project year, he had been a full-time professor

Additional single copies of this Electronic copies of MTDC’s For further technical information, document may be ordered from: documents are available on the contact Jon Driessen at MTDC. USDA Forest Service, MTDC Forest Service’s FSWeb Intranet at: Phone: 406Ð329Ð3931 5785 Highway 10 West http://fsweb.mtdc.wo.fs.fed.us Fax: 406Ð329Ð3719 Missoula, MT 59808 E-mail: [email protected] Phone: 406Ð329Ð3978 Fax: 406Ð329Ð3719 E-mail: [email protected] 16