Criminal Liability & Aircraft Accident Investigation

By Capt. Lindsay Fenwick (Northwest) and Michael Huhn, ALPA Senior Staff Engineer
Air Line Pilot, May 2003, p. 17

The worldwide trend toward criminalization of airline accidents has the potential to cripple our ability to learn from incidents and accidents, essentially guaranteeing that these events will be repeated. This is a price that we cannot afford to pay.

Not only is punishment (or the threat of it) an ineffective deterrent to mistakes, it has a detrimental effect on the conduct and quality of accident investigations.

The admirable safety record of the airline industry is primarily due to two key factors: the dedication to safety of the organizations and the personnel directly involved in this industry and their ability to learn from their errors. The proposition of "guilty until proven innocent" implied by criminalizing accidents is a significant affront to all those who do their best on a daily basis. All of us who participate in and rely on safe and efficient global air transport have a great deal to lose.

Airline pilots are the ones who are most frequently singled out for prosecution. Airline pilots sign for and assume command of their aircraft, passengers, and cargo before leaving the gate. To a large degree, airline pilots are typically the last individuals in the event chain who might be able to alter the outcome. Simplistic reasoning presumes that an airline incident or accident that does occur must be the flight crew’s fault. In addition, they always end up at the scene of the mishap, alive or dead, and so are physically accessible targets for blame. But the victims here aren’t just the pilots or other aviation professionals. The entire industry suffers, and that includes investigators and passengers, too.

The Air Line Pilots Association was formed more than 70 years ago with the primary purpose of creating a safer work environment for pilots when they were being killed in their mailplane cockpits in alarming numbers. Anything ALPA can do to ensure the safe arrival of an airliner is therefore on its agenda. Unfortunately, the public, and too frequently other aviation accident investigators, often believe that pilots associations participate in incident and accident investigations to absolve the flight crew of any wrongdoing. This is not true. Absolving a flight crew of any wrongdoing instead of fixing the system does nothing to prevent another flight crew from repeating the very same sequence of events, and causing yet another accident. Fixing the system, not apportioning or absolving blame, is clearly in everyone’s best interest.

ALPA does believe that criminal law has a place in aviation, but the consequences must be connected to the intent. An accident, by definition, is an occurrence that is not expected, foreseen, or intended. By this definition, then, an accident is not a crime. In English Common Law, a crime requires two elements—intending to commit an unlawful act and actually committing the deed. Most airline accidents are just that—unfortunate and unforeseen consequences of human error, not necessarily the pilots’.

Those of us who have been around aviation for a while, especially on the military side, remember when a gear-up landing would ruin an aviator’s career. Still, pilots continued to land gear-up. Punishment was an ineffective deterrent against mistakes because punishment is effective as a deterrent only in the event of a premeditated act, and even then it may have both positive and negative results. Only when appropriate changes in aviation "systems"—design, operations, training, and many other aspects—are implemented will deficiencies be corrected and safety improved.

Punishment (or the threat of it) also has an adverse effect on the conduct and quality of accident investigations. In a 1999 address to a U.S. congressional committee, former National Transportation Safety Board Chairman Jim Hall said, "My investigators have been stymied by the prospect of criminal prosecutions.…A number of our investigative activities have been suspended because most of the central players will not talk to us." He was talking about a pipeline accident, but the problem is also present in aviation investigations.

In the post-ValuJet era, individuals and organizations that are exposed to criminal liability will implement appropriate safeguards to try to mitigate such liability.

The investigation into the maintenance issues surrounding the Alaska Airlines Flight 261 accident was stymied by prospects of prosecution.

The willingness to cooperate with the NTSB may never be the same. Such a development clearly conflicts with the efforts that aviation experts make to enhance flight safety.

Countries that have demonstrated a willingness to prosecute and incarcerate pilots and controllers tend to have worse accident rates than those that do not. This has been observed in parts of Latin America, certain European countries, and much of Asia and Africa. This unfavorable correlation does not necessarily imply cause and effect, but cannot be ruled out, either.

One theory is that when regulatory and safety organizations fail to respond to society’s needs for corrective action after an accident, the public concludes that the organizations are not meeting their responsibilities to society, and the public will then tend to seek satisfaction via some other means. Unfortunately, these "other means" are typically judicial prosecutions of the personnel and organizations involved, which contributes nothing to advancing aviation safety, thereby enabling yet another accident.

ALPA continues to rely on the tradition that in the United States when a pilot acts in what he or she perceives to be in the best interest of the safe operation of the aircraft, he or she will not be criminally prosecuted.

One reason that the accident investigation process in the United States has been so effective in identifying safety deficiencies is the ability of investigators, through cooperative means, to obtain complete and accurate information from those who were directly involved in the conduct of the flight. Despite the potential pitfall(s) of doing so, flight crews and others do cooperate willingly with accident investigators.

In the United States, surviving flight crews provide written statements and make themselves available for interviews with the NTSB and other organizations. This information eventually becomes public record and is obtained by both the employer and the regulator (the FAA). Consequences of being involved in an accident can include company discipline (remedial training, unpaid leave, or dismissal) and/or regulatory sanction (certificate suspension or revocation, fines, etc.). Although typically to a lesser degree, other certificate holders, such as aircraft mechanics and certain air traffic controllers, are also similarly exposed to such disciplinary and regulatory actions.

The statements of surviving flightcrew members are of critical importance to accident investigations. ALPA continues to rely on the tradition that in the United States when a pilot acts in what he or she perceives to be in the best interest of the safe operation of the aircraft, he or she will not be criminally prosecuted. Based upon this reliance, and in spite of concerns about the potential negative effect of full disclosure, ALPA advises the pilots it represents that they should cooperate fully and aid in the investigation. Flightcrew statements are given freely with legal counsel provided by ALPA, even when the statement may reveal error on the part of the flightcrew member. Additionally, flightcrew members continue to cooperate and provide additional
information during all stages of the investigation.

Thorough and meticulous accident investigation has played an integral role in identifying mechanical defects that need to be corrected, operational procedures that need to be changed, and human factors issues that need to be addressed. This investigative process, and the consequent ability to learn from catastrophic accidents, enables us to prevent their recurrence. The NTSB’s "interested party" system is crucial to this process. This system relies on open, cooperative participation. Detractors claim that the interested party system is a case of the "foxes watching the henhouse," but actually it is the mutual examination of and by all involved, each with their separate and distinct areas of discipline, knowledge, and experience. This yields a system of checks and balances, and enables an objective discovery and analysis of the circumstances. Rather than establishing blame, causal factors are identified, safety recommendations are made, and ideally, appropriate and effective corrective actions are taken.

Until very recently, all these parties have participated fully and openly in accident and incident investigations. As a result, the U.S. aviation community has been extremely successful in obtaining the information necessary for reducing the accident rate in spite of the twin deterrents of potential corporate and regulatory disciplinary action. The potential for criminal prosecution is injecting a significant deterrent into the current environment of openness and cooperation.

If criminal prosecution of flightcrew members does become a realistic possibility in the United States, it will end the current cooperative spirit of pilots in accident investigations. Certainly, no lawyer would allow a client to willingly provide incriminating evidence against himself or herself if it was likely to be used in a criminal prosecution of the client. This also applies to all other aviation personnel whose testimony is necessary for a better understanding of the events leading to the accident. Criminalization will stanch this information flow.

The involvement of the FBI in the July 1996 TWA Flight 800 accident investigation hampered aviation safety experts’ efforts, but the July 2000 Air France Concorde accident near Paris, France, provides an even clearer example. Here were two vastly different approaches to the same investigation.

The Concorde was an aircraft that both the British and French authorities knew intimately, and both countries had national pride and the reputation of their flag carriers at stake. The U.K.’s Air Accident Investigation Branch was able to do its work investigating this accident unencumbered by the prosecutors and courts, whereas French investigators had to contend with a dual technical and judicial process. Not surprisingly, British officials formally criticized what they termed were "unacceptable restrictions" on evidence and access to the site that the French authorities imposed. The investigation moved at a snail’s pace, through no fault of the Bureau Enquêtes-Accidents.

Society’s greater good is not well served by judicial proceedings that can cripple prevention efforts. Nowhere does this hold more true than in the field of aviation safety.

Even though a dual-track approach might be less efficient, the proper causal factors can still be identified and valid safety recommendations made. The U.S. military has for years run dual-track inquiries into aircraft accidents, and investigators are able to obtain witness statements under a promise that they will not be released or used for purposes other than accident prevention. What enables this system to work is that a solid barrier (one element of which is a policy of noninterference) exists between the technical investigation and the collateral judicial inquiry. The two investigations are separate but equal in access to information. Most countries that have a dual-track approach to accident investigations do not provide such a barrier, and furthermore maintain that the judicial inquiry takes precedence over access to information. In these cases, the interference of the judiciary with the technical investigation can have serious consequences.

Society’s principal interests after any catastrophic event are captured in the following four terms:

Causation denotes the technical reasons for the event and, in the context of air transportation and aircraft accidents, most closely approximates the technical safety investigation.

Accountability refers to determining the individuals and organizations whose processes, deficiencies, and failures contributed to the accident sequence. In layman’s parlance, these are the ones who are "responsible" for the accident. More importantly, however, these organizations and individuals have the ability to modify the system in order to prevent a similar accident.

Prevention is presumed to be self-explanatory.

Compensation refers to the price that society extracts from those "responsible" for the accident.

These four principal interests are interdependent, but not all of these processes will necessarily transpire in every accident. Regulations, societal norms, available resources, and of course, the number of fatalities are the primary influences affecting whether they do transpire, and to what extent.

Regardless of our culture, our instinctive reaction after a disastrous occurrence is to believe that someone must be at fault, and therefore this person or organization must be identified and punished. This redress or retribution can take various forms. As safety advocates, we would all prefer that the emphasis (and resources) be focused on "causation" and "prevention" instead of "accountability" and "compensation." In the United States, accountability typically takes the form of civil litigation, but a tendency toward criminalization has been increasing. In many parts of the world, redress for aviation accidents has long been attempted through the criminal justice system.

Investigative processes

At least three critical differences exist between the way technical and criminal investigations are conducted. When the investigations are conducted concurrently, these differences can cause the two methods to be in conflict. Without strict measures designed to prevent such conflict, only one method can prevail and will result in the decreased viability and efficiency of both investigations. Also, technical investigations tend to be nonpunitive, while judicial ones are punitive.

The effectiveness of Aviation Safety Reporting System program in improving safety depends on the free, unrestricted flow of information from the users of the National Airspace System.

In technical investigations, all the facts are collected without prejudice, and the accident scenario and sequence is developed from those facts. For the conclusions to be valid, the scenario and sequence must fit all the facts. In technical investigations, fact- gathering is a cooperative process that relies on the expertise of several diverse organizations. This method essentially ensures that nothing is overlooked. Finally, the technical investigators have intimate knowledge of and familiarity with the aircraft and its operating environment.

In contrast, typical law enforcement investigations are sharply focused on a predetermined "conclusion," and the law enforcement agencies then attempt to gather facts that support their conclusion. Most, if not all, law enforcement personnel who gather facts for the investigation are seeking the same information (are "fixated on the same target"), and the benefit of diverse interests and the resulting flexibility are lost, or at least severely inhibited. Finally, in our experience, most law enforcement personnel involved in airline accident investigations have known little or nothing about aviation. Also, in a punitive environment, the tendency is to identify and punish or eliminate the "bad element" and resume "business as usual."

Obviously, different countries have different laws, approaches, and procedures regarding the precedence, sequence of, and interplay between the technical and judicial investigations of aircraft accidents. Australia, Great Britain, the United States, and several other countries have systems in which the technical investigation takes precedence, unless criminal activity is indicated or proven. Other countries, such as France, provide for concurrent technical and judicial inquiries. Finally, some other countries, more than we’d like, have judicial, military, or some hybrid, loosely defined or constantly changing investigative approaches that are the norm. We should all be able to agree that while we need to be sensitive to try to accommodate each country’s sovereignty, we should never lose sight of our overarching goal, improved aviation safety.

Even though in the United States the NTSB typically has primacy in an aircraft accident investigation, some U.S. investigations have involved difficulties in obtaining information. The prosecutions in the case of the May 11, 1996, ValuJet Flight 592 crash in the Florida Everglades have had an adverse effect on the free flow of information to accident investigators. Former Chairman Hall, during an NTSB symposium, reported, "Those [SabreTech] prosecutions caused us to take a hard look at the possibility that old types of information might also be lost to the accident investigator. For decades, we had relied on individuals to tell us what happened in an accident—and they usually, sometimes reluctantly, did so." After the SabreTech prosecutions, Hall continued, "we feared that what had been reluctance to cooperate would now become refusal. The [previously cited] pipeline accident in Bellingham, Wash., proved us right. A criminal investigation of the accident was immediately launched, and we have yet to talk to most of the individuals who were operating the pipeline when it ruptured [in] June [1999]. As a result, serious safety issues and serious questions about prevention remain unanswered."

ALPA members are becoming increasingly concerned about the possibility of criminal liability. This concern is generated primarily because of the increased foreign operations of U.S. air carriers, whose flight crews now operate over and into countries whose criminal laws are considerably different than those of the United States. Lurking in the background is also a not-unfounded fear that some local prosecutor in the United States will seek to prosecute a pilot in some future accident to promote some personal, professional, or political agenda.

The accident involving USAir Flight 5050, which occurred at New York’s La Guardia Airport in 1989, attracted the attention of the Queens County District Attorney, who then brought the issue before the grand jury to determine whether criminal charges should be pursued. Fortunately, no charges were ever filed. However, the action of the DA hampered the normal course of the accident investigation. Because the accident scene was declared a "crime scene," access to the wreckage and witnesses was under the control of the local authorities, who superseded the NTSB’s authority. The DA’s involvement and the potential for criminal indictment of the flight crew remained until nearly 9 months after the accident. Eventually, the DA withdrew, and the investigation was concluded pursuant to normal NTSB procedures.

USAir Flight 5050 was not a case in which the NTSB felt that the accident was caused by a criminal act, and therefore the NTSB was not relinquishing its jurisdiction to law enforcement agencies. Rather, the NTSB correctly believed that it was an accident caused by either a mechanical failure or pilot error in the traditional sense, and that it was not a matter of concern to local law enforcement officials. Nevertheless, the DA temporarily prevailed, and the investigative process was hampered.

The investigation into the loss of TWA Flight 800 demonstrated that in fatal accidents in which criminal activity is suspected and the NTSB cedes some authority, the involvement of the law enforcement agencies hampers the investigative process. Although the outcome of that investigation would most likely have been similar had the law enforcement agencies not been involved, a freer flow of information would have improved the efficiency of the investigation. The FBI was the lead law enforcement agency, and its evidence procedures (chain of custody, release of information, etc.) conflicted with and hampered the NTSB process. In addition, the "veil of secrecy" that the law enforcement agencies introduced gave rise to numerous rumors and theories, which caused the NTSB to divert significant time and resources to counter. This was both counterproductive and expensive.

The investigation of the Alaska Airlines Flight 261 accident also involved law enforcement agencies acting beyond their normal scope. Although traditional criminal (terrorist) activity was not suspected in this case, allegations of inappropriate maintenance activities were made, and law enforcement agencies intervened. Their involvement limited technical investigators’ access to certain key elements, including hardware (wreckage), paperwork (maintenance records), and personnel interviews.

In a similar fashion, the large FBI presence in the investigation of the Emery Flight 17 accident, which involved a DC-8 cargo flight in Sacramento, indicates a disturbing trend of significantly increased law enforcement involvement in technical investigations in the United States.

In both of these accidents, law enforcement agencies introduced themselves into the investigations on the premise that intentional criminal activity contributed (albeit perhaps indirectly) to the loss of the airplane. The presence of law enforcement agencies did not cause the difficulties in these two investigations that were experienced in the TWA 800 investigation, but their presence did inhibit certain fact-gathering and testimony. Also disturbing is that these agencies were acting on perceived levels of criminal activity.

Safety action programs

An interesting irony regarding this trend toward criminalization of accidents in the United States is that on an airline-by-airline scale, the U.S. airline industry is becoming a nonpunitive environment, while on the larger scale, the situation seems to be trending in the opposite direction. Specifically, while nonpunitive information reporting systems such as Aviation Safety Action Programs (ASAPs) are beginning to come into widespread acceptance and implementation, this rise is accompanied by an increase in criminalizing certain aspects of incidents and accidents, which is the philosophical opposite of the premise underlying ASAP.

People are more likely to divulge information if they are not risking retribution or punishment. In the United States, the first formalized steps toward stimulating the large-scale flow of critical safety information occurred in 1975 with the implementation of the Aviation Safety Reporting System. As the FAA’s ASRS Advisory Circular notes: "The effectiveness of this program in improving safety depends on the free, unrestricted flow of information from the users of the NAS [National Airspace System]. Based on information obtained from this program, the FAA will take corrective action as necessary to remedy defects or deficiencies in the NAS. The reports may also provide data for improving the current system and planning for a future system."

This cooperative safety reporting program is administered by the National Aeronautics and Space Administration and invites pilots, controllers, flight attendants, maintenance personnel, and all others to report actual or potential discrepancies and deficiencies involving the safety of aviation operations. This program is in fact nonpunitive in two respects. First, the federal aviation regulations prohibit the use of any ASRS reports (or information derived therefrom) in any disciplinary action, except for criminal offenses and accidents. Second, as an inducement to stimulate this information flow, ASRS (with certain provisos) typically grants reporters immunity from sanctions if they have violated FARs. ASRS has proven to be very successful at generating a significant amount of "raw data," but unfortunately its shortcoming has been a relative lack of resources available for data mining and analysis. ASRS is currently undergoing some examination for potential revision.

Other countries have implemented or have begun developing nonpunitive reporting systems. Unfortunately, the list of countries without these systems is vastly longer than the list of those that have already implemented or are developing such systems. ALPA strongly endorses and promotes the proliferation of government-sponsored nonpunitive reporting programs. The advent of company-sponsored nonpunitive programs, such as ASAP in the United States and BASIS in the U.K., is beginning to partially mitigate the urgency for government-sponsored programs, but these, too, are relatively rare. Ideally, the global airline industry will eventually see the day when every nation has a government-sponsored system, and every airline has an ASAP program. Unfortunately, we suspect that that day is quite some way in the future.

Several years ago, aviation safety experts recognized that the reactive approach to aviation safety (investigation of accidents) had reached its limit in its ability to further significantly decrease the accident rate, and that a more proactive approach was required. Extremely capable flight data recording and analysis tools were seen as the next logical step in driving the accident rate down. The rationale was to amass and analyze large amounts of flight data to detect trends and unsafe events, the precursors to accidents. By identifying and correcting these deficiencies before they resulted in accidents, the concept of proactive safety was implemented. Besides ASAP, examples of these proactive safety programs include Flight Operations Quality Assurance (FOQA) and the Global Aviation Information Network.

The primary components of this proactive safety strategy are recording flight data and sharing that and other safety data. Collecting and sharing these data has potentially significant business, financial, and legal implications for operators, manufacturers, and flight crews. In the United States, airlines, flightcrew unions, and the FAA are engaged in lengthy and detailed negotiations about the availability, application, and pitfalls of such proactive programs. The potential for these data to be used in criminal proceedings adds another layer of complexity to an already difficult situation.

The potential for criminalization being anything less than extremely remote could bring these proactive programs to a standstill. The United States is currently at a crossroads in regards to aviation safety. If we are serious about improving the safety of our air transportation system, we absolutely cannot afford to allow criminalization to infiltrate this arena.


The motivating forces or enabling conditions behind the trend toward criminalization of accidents could be an implicit indictment of the regulators’ inability to satisfactorily oversee and regulate the airline industry. In other words, if the regulator is doing a consistent and thorough job of regulatory oversight, the regulator will detect and eliminate certain "wrongdoings" before they lead to loss of aircraft and lives. Regulatory oversight does not mean traditional enforcement as the exclusive means of ensuring compliance and raising safety standards. ALPA advocates a cooperative approach over a "police mentality" whenever possible; standard enforcement and sanction procedures should be used only for intentional acts of misconduct or as methods of last resort. The regulator’s principal mandate is to ensure the safety of the air transportation system. Certainly the public has the right to expect, even demand, nothing less.

Human involvement

People are an integral element of the air transport industry, and people always have and always will make mistakes. An analysis of accident reports shows that nearly 75 percent of aircraft accidents that occur in the United States involve some form of human error. This is likely true the world over. In virtually all these situations, the personnel involved were performing their duties to the best of their ability, to ensure the safe completion of the flight. Nevertheless, even under these circumstances, accidents do occur. Mistakes can be errors of omission or commission, and can be by pilots, air traffic controllers, designers, loaders, mechanics, and others. Identifying these errors, and then implementing changes to make the air transportation system error-resistant and error-tolerant, are vital steps in the process of improving safety.

The FAA’s current regulatory and enforcement scheme does a poor job of appreciating and integrating the fact that humans err. All evidence indicates that regulators are not much different anywhere else in the world. The looming specter of the criminalization of honest mistakes committed during the course of one’s duty compounds the issue and will undoubtedly result in a significant crippling of both reactive and proactive safety improvement efforts. Finally, if criminalization is in fact the trend of the future, what precludes any and all regulatory personnel from being subjected to these same criminal prosecutions? After all, the regulatory agency is ultimately responsible for the safety of the traveling public.

ALPA members do not seek to be abrogated of their responsibilities and do not ask for the freedom to exercise willful misconduct during the course of doing their jobs. As an organization of and for professional pilots, ALPA fully recognizes and accepts airline pilots’ responsibilities and authority. What ALPA does want, however, is protection from overzealous or misinformed organizations and individuals who are attempting to convert legitimate human errors and aviation industry failures into criminal behavior for their personal, political, or corporate benefit.


Airline pilots need appropriate legislation that ensures the protection of information critical to flight safety. Such protections can range from precluding any processes or efforts that subvert legitimate accident investigations into criminal prosecutions or investigations to significantly limiting access to critical flight safety data used in proactive safety programs such as FOQA and ASAP.

Because the airline industry is global, accident and incident investigation should be compatible with the culture and legal processes of a multitude of states. Most judicial systems recognize the need for a system that determines causation, allocates accountability, and provides compensation. Although such systems may serve society adequately, they are largely reactive and do little to directly address prevention. This extra dimension, prevention, deserves to be more than an afterthought. Otherwise, when the lawyers are finished, little will have changed and we will continue to inflict harm. Society’s greater good is not well served by judicial proceedings that can cripple prevention efforts.

Clearly, the accident investigation and legal processes must be balanced. The legal systems of most nations recognize that willful misconduct that causes harm constitutes a crime. Most countries of the world are signatories to the Chicago Convention and have generally agreed to abide by the provisions of ICAO Annex 13. ICAO language, as well as that of the Warsaw Convention and the FARs, already recognizes the concept of willful or wanton misconduct.

ALPA proposes adding to ICAO Annex 13 a new section and paragraph, which could be inserted after the section entitled "Coordination—Judicial authorities." The proposed language would be as follows:

Proceedings to impose sanctions

5.11 Recommendation—States should not impose criminal sanctions arising out of an aircraft accident against individuals unless an independent judicial determination holds that the accident or incident was caused in whole or part by such default on the individual’s part as, in accordance with the law of the court to which the case is submitted, is considered to be equivalent to willful misconduct.

As individual safety professionals and as an industry, we must do everything in our power to remove all roadblocks to improving aviation safety, and criminalization is beginning to appear as our most pressing priority in this regard. Raising the airline industry’s consciousness and establishing international standards to govern the imposition of criminal sanctions after an aircraft accident seem to be the best starting points. These standards would be a natural extension of existing language in the Warsaw Convention and many countries’ aviation regulations, and would ensure the primacy of the investigative inquiry while allowing for prosecution of willful acts. If we choose to ignore this growing crisis of the criminalization of aviation accidents, all of us, whether pilots, investigators, mechanics, or the traveling public, will certainly suffer.

Capt. Lindsay Fenwick is the chief accident investigator for the Northwest Airlines MEC, the vice-chairman of ALPA’s Accident Investigation Board, the chairman of the ALPA Accident Analysis Group, and the ALPA delegate on the IFALPA Accident Analysis Committee. Since February, he has headed ALPA’s Criminalization Project, which the Air Safety Steering and Oversight Committee has set up to chart the course for ALPA to interact with and influence the airline industry regarding this issue, in concert with ALPA’s Legal Department. Mike Huhn is a senior staff engineer in ALPA’s Engineering and Accident Investigation Section. This article is adapted from a paper presented at the 2002 meeting of the International Society of Air Safety Investigators.