Skip Ribbon Commands
Skip to main content
Bookmark and Share this page


Remarks at the ALPA/US Airways Safety Forum, Pittsburgh, Pennsylvania
Jim Hall
ALPA/US Airways Safety Forum, Pittsburgh, Pennsylvania

I'm pleased to have been invited to join you here in Pittsburgh for your 10th Annual Safety Forum to speak on the subject of "Safety Through Cooperation."  As I was thinking about what to say this evening, I couldn't help but be reminded of Benjamin Franklin's admonition that "We must, indeed, all hang together or, most assuredly, we shall all hang separately."  Of course, he was speaking to the signers of the Declaration of Independence, but his meaning holds true for all of us.  We have certainly learned over the years that a safe aviation system isn't possible without the cooperation of everyone involved in that system - the regulators, the manufacturers, the airlines, the crews, the maintenance personnel, and the passengers.  Most assuredly, if we don't work together, the system will never be as safe as we all expect it to be.

Nineteen ninety-eight showed the benefits of that cooperation.  It was the safest year on record for U.S. commercial aviation.  But, despite our best efforts, we must also realize that accidents can - and do - occur and that cooperative spirit gets severely tested as we all try to deal with the tragedy - and, hopefully, learn from it.

When a major air carrier accident does occur, the Safety Board's investigators have the difficult task of finding our what happened and why.  But, we don't do it alone.  In fact, we can't do it alone.  Alexander Graham Bell once said, "Great discoveries and improvements invariably involve the cooperation of many minds."  The same is true of accident investigations.  We must rely on the cooperation and expertise of others through the "party system" to help us uncover the facts.  That's what I want to talk to you tonight - how we all cooperate on accident investigations to prevent accidents, and in doing so, create something positive out of an tragic event.

As many of you know, the party system allows the Board to expand its resources and obtain the expertise needed to investigate an accident.  What you may not know is that the Board only has about 400 employees - only about 130 of which are assigned to the Office of Aviation Safety.  Obviously, those individuals can get spread very thin, very quickly as they work on the domestic and international major aviation accidents we investigate - we've assisted in about 40 foreign accidents in the past two years - as well as the approximately 2,000 general aviation accidents we investigate annually.

The party system allows us to call upon experts from the airlines, aircraft manufacturers, associations and unions, FAA, and other organizations that have relevant expertise to assist our investigators.  This system has been used effectively and successfully for more than 30 years.  It provides the Board's investigators with the invaluable expertise of groups and individuals who know how the airplane was designed, how to fly it, and how it should be maintained and operated.  In fact, as you may know, the expertise provided by the parties is the sole reason they are invited to participate - only one party, the FAA, has the right to participate in our investigations.  Others are there at the Board's request.  In addition, some groups - such as attorneys and public affairs personnel - are specifically excluded from the process.

But, as I said, the system works well, has a long history of cooperation, and creates a healthy tension among the parties - as it was designed to do.  Just this past March, the Safety Board completed the longest and, in many ways, the most difficult investigation in its history - that of the 1994 accident of USAir flight 427, which crashed nearly in Aliquippa, Pennsylvania.  Partially because this investigation lasted so long, it proved to be a particular challenge to the party system.  As I'm sure many of you are undoubtedly aware, our society is becoming increasing litigious and that can have a potentially negative impact on our investigations.  In the case of USAir 427, many of the investigative parties were quickly immersed in the litigation process while we were still gathering facts and conducting tests.  It became increasingly challenging for the parties to tell us what they knew, while also trying to avoid burdening the investigation with litigation-oriented scenarios and arguments.

Potential conflicts of interest between the NTSB and the other parties dissipated in the name of safety.  A group of Boeing engineers discovered, through their own testing, that it was possible for the 737's rudder system to jam.  Boeing provided those findings to investigators - and they resulted in a redesign and retrofit program for the rudder system.

Similarly, USAir's pilots, mechanics, and management made a sustained effort over several years to identify, document, and report all of the rudder-related anomalies and events within their 737 fleet to the Safety Board's investigators.  This process not only brought them some unnecessary and unwelcome public scrutiny - it also cost them a great deal of money - but the end result was invaluable information to the investigators.

This isn't to say that party members in other investigations haven't, on occasion, raised issues or focused investigators in directions that turned out to be red herrings.  I'm not going to name names or point fingers.  But, I will say that such activities are not only frustrating and counterproductive - because we have to pursue each theory until it can be eliminated - they also slow the investigation, delay finding the true cause, and limit our ability to make recommendations that will keep it from recurring.  I would hope that anyone contemplating such tactics would think twice and remember that they are not only expending valuable resources that could be better used elsewhere - they could very well be costing individuals their lives if another accident occurs.

Another danger to the cooperative spirit engendered by the party system is unauthorized public statements by parties during an accident investigation.  Most of you are probably familiar, at least in a general way, with our rules and procedures regarding the release of information to the public.  Basically, we are the only entity that releases that information either to the families or to the media.  Limiting who releases information helps ensure that only accurate and verified factual information is given to the public and helps avoid situations in which parties are pitted against one another in publicity wars.  We understand the need for parties to "get their stories out" and we encourage them to do that - as long as they don't talk about the accident itself.  Our traditional guidance has always been that "If you could make this statement on the day before the accident, you can make it after the accident."

Following the fatal accident at Little Rock, it became clear that this guideline was insufficient for today's 24-hour news media environment.  As a result, the Board and the Air Transport Association have been working closely together over the last few months to identify specific concerns and refine our procedures.  We had hoped to announce those this week, but have had to briefly delay that announcement.  I believe that all parties will find the new guidelines much clearer and more workable.

As a result of these problems, and others, the party system has come under assault in recent years from many factions.  I decided that it was time to review not only the process but also the way the NTSB conducts its work.  Last year, the Safety Board asked the RAND Corporation's Institute of Civil Justice to conduct an independent review of two critical areas.  First, we asked them to examine and evaluate the Safety Board's workload, staffing levels, and training programs in light of the emerging trends in aviation.  Second, we asked them to review the Safety Board's party system. We also asked them to make recommendations to us in both areas to ensure the Safety Board's continuing ability to accomplish its mission.

The Safety Board itself has periodically reviewed the party process, but this was the first time that we have had an independent, outside expert look at our investigative procedures.  As I mentioned, there were a number of calls to revisit this subject.  Interestingly, many of these calls came from two extremely different perspectives.  On one hand, some industry representatives wanted to expand their role in the process, particularly as it relates to involvement in the analysis stage of the Board's work.  On the other hand, some family members of victims and plaintiffs' attorneys believed just as strongly that the current system gives party members what amounts to a privileged position in terms of future litigation, while giving them no role.

As part of their research, RAND conducted extensive surveys of many of the organizations and government agencies who routinely serve as party members, as well as those groups that aren't traditionally part of the process, such as family members and liability lawyers, among others.  Although RAND's work encompassed a number of areas that will have agency-wide applicability, the study focused primarily on aviation issues and challenges.  We expect the final report to be released within the next few weeks.

Probably the most important issue raised in the report indicates that complex and contentious accident investigations, such as the USAir flight 427 investigation and the on-going TWA flight 800 investigation, are likely to be the norm in the future rather than the exception.  Our current ongoing investigation into the crash of EgyptAir 990 that crashed last week off Rhode Island may prove to be no different.   Safety Board investigators and party members are still on scene.  Progress to date has been slow because of weather delays.

These complex investigations have been extremely taxing to the Safety Board and its personnel; we know that if we are to be prepared to investigate similar accidents, we must adopt a number of new and different strategies.  And, we are, in fact, working on new management and financial practices that will help ensure financial and programmatic effectiveness for the 21st Century.   We anticipate that RAND may recommend others.

Secondly, the RAND report focuses on the workload of our staff and on the knowledge base they'll need to successfully accomplish our mission.  We need greater depth in many of our high-skill positions and we must have the resources to keep our investigators current in the latest technologies and procedures.

Finally, although we don't expect RAND to recommend significant changes to the party system, we understand that certain adjustments will be recommended to broaden the probable cause statement and to expand the use of outside laboratories and experts during the actual accident investigation phase.  However, we believe that the RAND report will endorse the general structure and operation of the current system and recommend few changes.

Before I end, let me mention another cooperative effort that can help prevent accidents before one ever occurs.   Flight operations quality assurance (FOQA) allows airlines to collect information from the vast majority of flights that are safely completed.  As a result, we can develop data that may prevent us from having to do an accident investigation.  We can learn about possible risks, such as the location of difficult instrument approaches, training procedures needed to help pilots perform their jobs better, air traffic control procedures that need to be improved, before they become an accident cause.  In the end, not only the individual airline, but also the entire aviation system benefits.  In addition to FOQA, there are other voluntary incident reporting programs, such as aviation safety action partnerships, that can play a significant accident prevention role.

I know that working out the ground rules for these programs is difficult and requires some trust on everyone's part.  And, regulators have to restructure regulations and enforcement procedures to accommodate them.  None of these are insurmountable - if groups are willing to work together to find a solution.  I'm also aware of the numerous privacy issues involved.  That's why the Board will be sponsoring a symposium on those issues next spring - I hope your organization will participate.

Buckminster Fuller said "We are not going to be able to operate our Spaceship Earth successfully nor for much longer unless we see it as a whole spaceship and our fate as common.  It has to be everybody or nobody."  The party system is very similar.  We must all work together to make it work.  And, I can honestly say that ALPA has been an active, integral part of that process and we at the Board thank you for your invaluable contributions over the years.  We all share a tremendous responsibility for the safety of those who fly and together we can make a difference.

Again, thank you for inviting me to be here tonight.