This is the fifth and final chapter in an investigation supported generously by the Fund for Investigative Journalism and the George Polk grants for investigative reporting administered by Long Island University.

In a publication offering recruitment advice to fire departments, the National Fire Protection Association lists 13 attributes that the job of firefighting requires. Most are physical—the ability to climb a lot of stairs, lug a lot of gear, move heavy hoses around, rescue large people and the like. But one goes to what's under the helmet.

A firefighter, says the NFPA, must be capable of "critical, time-sensitive, complex problem solving during physical exertion in stressful, hazardous environments (including hot, dark, tightly enclosed spaces), further aggravated by fatigue, flashing lights, sirens, and other distractions."

Even if a fire department had all the best equipment and all the right tactics for every possible situation, and all the person-power it could possibly need, firefighters could still die. Indeed, in many cases, fires don't kill firefighters and equipment doesn't save them. Decisions do.

'We're not daredevils'

Some of those decisions are made by individual firefighters.

There is a narrow middle ground between being unwilling and being too willing to risk one's life, and fire departments survive on personnel treading that border very carefully. Civilians cannot afford to have firefighters who are too careful; otherwise they'd never risk coming through smoke and fire to save you. But firefighters can be harmed by what the FDNY probe of Deutsche Bank called a can-do attitude.

"The 'can do' attitude," the report reads, "has enabled the FDNY to protect life and property at a superior level of excellence since the Fire Department's inception." But, the report continues, the attitude can also lead firefighters to take unnecessary risks.

Retired FDNY lieutenant Steve Mormino recalls a chief who once told him that the problem with firefighters wasn't getting them into a burning building; it was getting them out. "We pride ourselves on that," Mormino says. "We know that we're a last line of defense for someone. We're constantly trained on that. To know that someone's survival depends on you is one of our biggest motivating factors. I don't know how to curtail that." Plus, there's a part of it that is not conscious. Sometimes at a fire, he says, "you get tunnel vision. You're just not aware of everything that's going on around you."

The fire department says it is trying to change firefighters' approach to risks by making videos of safety tips available to firefighters, scheduling presentations and discussions about safety at firehouses, and encouraging FDNY members to participate in the Near Miss program and Pass It On project, in which firefighters share safety information and stories of close calls.

There are signs the culture is changing. The stigma around issuing Maydays has largely disappeared, firefighters say. More people are using their breathing equipment more often. "We're not daredevils," says the second fire lieutenant we spoke to. "We don't want to die."

"FDNY guys go in."

In December 1999, the fire department in Worcester, Mass., responded to a blaze in an abandoned industrial building called the Cold Storage Warehouse. There were reports of squatters living inside, so firefighters mounted a search as the flames grew. Six men from three different fire units became lost inside. Their radio transmissions are still chilling. "Get people up on this floor now or we are going to die! We have no air, and we cannot breathe," one man called. Other firefighters fought through heat and flames to try to signal the way out, but the fire drove them back. Finally, the chief in charge, Mike McNamee, ordered the building evacuated. He physically barred the door to prevent other firefighters from obeying their fundamental instincts and running in. He cut his losses.

While firefighters make choices that risk or save their own lives, fire officers make decisions that affect the safety of dozens of people. Those decisions naturally come under the microscope when a firefighter dies.

"Some decisions, you can establish a commission, pose a question," says Al Hagan, the FDNY captain who leads the union that represents chiefs and other officers. "After six months or a year, you come up with a decision. Fire ground commanders are not afforded that luxury. We make those decisions in a blink. It all happens at a subconscious level, and it happens very quickly. You search your mental Rolodex for the closest match [of earlier fires] and you say, 'What was done? How did it turn out?' "

The officer in charge at a fire makes a risk-versus-reward assessment to determine how much danger he should expose his firefighters to. It's a complex, multivariate calculation: Are there civilians in danger? Can they be saved or, like those victims trapped above the floors of impact on Sept. 11, are they doomed? Does the likelihood of saving them outweigh the risk that rescuers would take in the attempt? How close do we get to the fire before its threat to us exceeds the advantages of fighting the blaze at close range?

One of the key decisions an incident commander makes, says NIOSH's Tim Merinar, is "how aggressively you want to fight an individual fire."

The FDNY prides itself on weighting its risk-reward calculus toward aggressive, interior firefighting. "You have to put it in the context of the culture of the FDNY. They are very aggressive," says former fire commissioner Nicholas Scoppetta. "There are fire departments in the world that sort of fight fires from the outside. But if there is any indication that there are people inside the building, FDNY guys go in." And there is more than machismo behind that approach. In a dense, urban environment, playing it safe and fighting a fire from the outside means exposing nearby buildings to fire risk, accepting that families will be made homeless and setting neighborhoods up for the social impact that burnt-out buildings can have. It's not an easy choice.

Since Sept. 11, the FDNY has increased training for safety investigators and developed an annual risk management plan that assesses the possibility of death or injury and recommends changes. Injury reports have been computerized to allow lessons to be extracted. The department is also participating in what FDNY documents describe as a "national, multi-year academic research project to develop a world-class safety management system."