This article – ‘Sexual assaults in the sky: The failure of aviation law and how to fix it’ – first appeared in Precedent, the journal of the Australian Lawyers Alliance, issue 151, published in April 2019 (Sydney, Australia, ISSN 1449-7719). It has been reproduced with the kind permission of the authors and the ALA. For more information about the ALA, please go to:

There has been a recent increase in reporting of sexual harassment and assault on airlines by both passengers and cabin crew, which may indicate increased incidence of such acts. Such reports highlight a multitude of issues, including the vulnerability of female and child passengers, and female crew members, particularly on overnight flights and when passengers are intoxicated. The reports highlight a dearth of operator policies on dealing with complaints of sexual assault, plus a lack of guidance and uniformity in the way that airlines address complaints about this kind of behaviour.

The lack of uniformity and consistency in prosecuting offenders across jurisdictions, combined with limitations in both international and domestic laws which prevent both prosecution of offenders and compensation for victims, conspire to make the safety of passengers in flight from sexual predators an area in need of closer local and international regulatory attention.

This article focuses on the challenges of achieving justice for victims under the international aviation treaties, particularly those focused on air accident compensation. It is arguably the weaknesses in the criminal system for prosecuting sexual offenders within the international aviation legal regime that restrict a victim’s ordinary recourse to justice to civil compensation schemes. However, where the international law does provide a means for recovery against an airline for bodily injuries that occurred on board, the path to compensation for passengers who are victims of sexual harassment and assault is much less certain. Highlighting the particular legal hurdles, this article also offers some suggestions for Australian law-makers that will permit better practical options being available for both airlines and complainants.


The unique, restricted environment of commercial air travel is one that arguably should require airline operators to take greater responsibility for protecting passengers and crew from sexual predators. The circumstances of air travel provide a ripe environment for sexual predators:

• confined spaces;
• limited space to move around the cabin;
• restrictive/allocated seating arrangements;
• being seated in close/confined proximity to strangers;
• darkened cabin lighting;
• service of alcohol;
• limited oversight/restricted ability for others to see what is happening along individual rows; and
• seating of minors in close proximity to adult strangers.

In 2017, the United States Federal Bureau of Investigation (FBI) received and investigated 63 cases of in-flight sexual assault, an increase from 38 in 2014, 40 in 2015, and 57 in 2016. Most, if not all, were passenger complaints. The actual rate of unreported incidents is likely to be much higher because of under-reporting due to embarrassment and the fear that claims will not be taken seriously.

In Australia, the Transport Workers Union reported a survey in 2018 in which four out of five airline cabin crew reported sexual harassment from co-workers, while three out of five experienced it from passengers. In a post-#metoo world, the airline industry has been called out as another environment where sexual harassment is rampant.

While sexual harassment and assault of airline cabin crew is an important and worthy issue to examine from a legal perspective, this article focuses on passenger claim issues.


What laws apply to acts on aircraft? In addition to a country’s ordinary jurisdiction over acts committed on its registered aircraft, or in its own territory, aviation laws also apply. The particular suite of laws will depend on the nature of the flight (whether it is domestic or international, and if international, it will depend on which of the aviation conventions the states involved have ratified). Regardless of which aviation laws apply, none typically enables a person to sue an air carrier for harm arising from passenger-on-passenger sexual assaults on board because of the curious history of the conventions concerned.

However, we would argue that aviation is the mode of public transport that is most in need of such protections.

Criminal aviation law

Under the Convention on Offences and Certain Other Acts Committed on Board Aircraft (Tokyo Convention) the state in which the aircraft is registered is competent to exercise jurisdiction over offences committed on board the aircraft. The Tokyo Convention outlines some particular offences under international law, but also does not exclude criminal jurisdiction exercised under national law.

The Protocol to Amend the Convention on Offences and Certain Other Acts Committed on Board Aircraft (Montreal Protocol) extends jurisdiction to prosecute offences and acts on board aircraft to include ‘the state of landing and the state of the operator’. It was promulgated to provide a fix for states whose authorities were not prosecuting foreign nationals on their own soil on the basis of lack of jurisdiction over the foreign national, in circumstances where a foreign aircraft had turned over an alleged offender to the landing state authorities.

The establishment of such jurisdiction over offences based on the criteria set out in the Montreal Protocol will be mandatory for parties to the Montreal Protocol. At the time of writing, only 16 states have ratified/acceded to the Protocol: Australia, along with the United States, the United Kingdom and Canada, has not. For the Protocol to enter into force, 22 ratifications are required.

So, for now, until the Montreal Protocol enters into force (and is incorporated into law within the states that ratify it) the clearest jurisdiction to prosecute offenders on board an aircraft remains with the state of the aircraft registration. This of course presents a sometimes-insurmountable problem for law enforcement when the state that most likely has custody of the accused and access to witnesses (the state where the aircraft has just landed) does not have jurisdiction to prosecute if it does not also happen to be the state where the aircraft is registered, and it refuses to exercise its territorial jurisdiction for political reasons, as is often the case.

Civil aviation liability law

Following failed attempts at pressing charges against an accused due to lack of jurisdiction (and sometimes in parallel), a victim of sexual assault may turn to air carrier liability law to seek compensation for physical and psychological injuries from the relevant airline.

The two primary conventions that relate to air carrier liability are the Convention for the Unification of Certain Rules Relating to International Carriage by Air (Warsaw Convention) with its many amending protocols, and the Convention which was intended to replace the Warsaw Convention and its ‘hodgepodge’ of agreements and amendments, the Convention for the Unification of Certain Rules for International Carriage by Air (Montreal Convention). Both are still relevant as some states have ratified the Warsaw Convention and some of the amending protocols, but have not ratified the Montreal Convention.

Importantly, the Montreal Convention applies to international carriage by air and imposes strict liability on air carriers for death or bodily injury caused by an accident taking place on board the aircraft or during embarkation or disembarkation [emphasis added]. The accepted judicial definition of ‘accident’ is set out in Air France v Saks (1985) 470 US 392 at 405, which expression was exhaustively analysed by the High Court in Povey v Qantas Airways Limited [2005]:

‘[a] passenger’s injury is caused by an accident only if “caused by an unexpected or unusual event or happening that is external to the passenger”’.

Damages become available by way of a two-tiered system. Under the first tier, passengers may claim up to 113,100 Special Drawing Rights against the air carrier for proven damage which the air carrier cannot avoid. Under the second tier, the passenger can then claim an unrestricted amount if, with a reversed burden of proof, the air carrier cannot establish that the damage either:

(i) was not due to the negligence or other wrongful act or omission on the part of the carrier or its servants or agents; or

(ii) was solely due to the negligence or other wrongful act or omission of a third party.

The range of claims under these treaties has varied across states and across airlines to include the uncontroversial and expected air crash and air disaster scale of deaths and injuries, to injuries caused by aircraft defects and turbulence (for example, QF72 claims in Australia and the United States); minor injuries caused by scaldings; passenger-on-passenger torts (fistfights triggered by alcohol supply at altitude); and, in a particularly noteworthy case, the refusal of a cabin crew member to reseat an asthmatic passenger to a ‘non-smoking’ area resulting in his subsequent death. There is no single thread of connection between the fact pattern and the behaviour or circumstances of aviation that can be said to determine whether an injury is compensable, rather there is only a requirement that an ‘accident’ should have occurred.

This presents a key obstacle for claimants alleging sexual assault on board, whose theory of liability may include the airline’s purported failure to prevent, or prevent the recurrence of, such an assault in the closed, cramped, dark quarters of an economy cabin.

Mental injury and the turmoil caused by sexual offences

While the above regime is the predominant legal tool with which to seek compensation against an air carrier for injury, it covers only ‘bodily injury’ and not pure psychological injury. Courts have generally held that emotional suffering or psychological injuries unaccompanied by a bodily injury are not compensable under either the Warsaw or Montreal Conventions.

In the UK case of Morris v KLM Royal Dutch Airlines, Lord Steyn commented that upon review of the travaux preparatoires of the Warsaw Convention, and finding no discussion of mental injury or illness therein, a line in the sand had been drawn to prevent an avalanche of intangible claims, with the use of the language ‘bodily injury’ rather than the term ‘injury’.

The upshot for the victim of a sexual assault on board an international flight, then, is that unless the assault occasions bodily injury with consequential mental injuries, the devastating psychological ramifications of such an assault cannot (presently or typically) be ascribed to the air carrier, and thereby sound in damages.

However, this does not mean that this is the only view or that the consensus will not change.

In the Australian case of Casey v Pel-Air Aviation Pty Ltd, the court held that the plaintiff’s PTSD, which was suffered as a result of ditching into the waters off Norfolk Island in 2009, was compensable under the Montreal Convention. Judge Schmidt followed the approach of the House of Lords in King v Bristow Helicopters, in which Lord Hobhouse expressed that bodily injury under Art 17 of the Convention ‘means a change in some part or parts of the body of the passenger which is sufficiently serious to be described as an injury’. A psychiatric illness was given as an example of evidence of a bodily injury where physical changes to the chemistry of the brain can be detected by sophisticated medical instruments. This was also a view expressed by Lord Nicholls in Morris.

The Supreme Court of New South Wales decision in Casey was overturned when Pel-Air appealed, arguing that PTSD did not constitute a ‘bodily injury’ under the Convention. While the NSW Court of Appeal did not discount that physical injury to the brain would constitute ‘bodily injury’ under the Convention, the evidence presented by Ms Casey was that of a chemical change to the brain and not a physical one, and was therefore not compensable.

While a narrow interpretation of bodily injury has generally been adopted by courts (so as to exclude psychological injury), an expansive interpretation of ‘accident’ has generally been applied which allows compensation for a broader range of airborne and ground-based mishaps.

The US Second Circuit held that the facts in Wallace v Korean Air satisfied the definition of ‘accident’ under the Warsaw Convention. In Wallace, the female passenger sued the airline after waking up on a flight from Seoul to Los Angeles to find the male passenger seated beside her had ‘unbuckled her belt, unzipped and unbuttoned her jean shorts, and placed his hands into her underpants to fondle her’.

It has been argued that this interpretation (considering the actions of sexual assault by a passenger as an ‘accident’) is ‘plainly antithetical to the Warsaw/Montreal compromise, which insisted that only an “accident” be the triggering mechanism for recovery under strict liability’. ‘Accident’ in later decisions by the Australian High Court, the UK Court of Appeal, and House of Lords has been held to both require something external to the passenger and relate to the operation of the aircraft or be a characteristic of air travel. In Morris v KLM, however, the House of Lords found that a sexual assault on a 15 year old girl (being fondled by a neighbouring male passenger during a flight from Kuala Lumpur to Amsterdam) satisfied the definition of an ‘accident’ because it was occasioned by a special feature of air travel, ‘namely, the cramped seating conditions which placed the young girl in close proximity to a strange man for an extended period of time’.

Following Casey, the US decision in Doe v Etihad Airways took a firm about-face from existing jurisprudence. The court concluded that passengers may be able to recover for emotional damages which are unconnected to the bodily injury. This would, in US cases at least, permit some lowering of the threshold of bodily injury required to permit compensation for mental injuries caused by an onboard sexual assault, provided that the court has found there to be an ‘accident’ in the vein of Wallace.

In an argument on behalf of airlines, Professor Dempsey, an aviation law expert who previously held the Directorship of the Institute of Air and Space Law at McGill University, has expressed the view that ‘sexual molestation aboard aircraft is a highly unlikely event’, and that ‘[p]ractically speaking, it is also unclear how airlines could avoid strict liability for such acts’, even asking whether ‘flight attendants (should) lift blankets and shine flashlights on passengers’ laps to make sure no one is misbehaving…?’

We believe the reported evidence shows that sexual molestation aboard aircraft is not a ‘highly unlikely event’ at all. In fact, sexual molestation is becoming an increasingly reported occurrence, especially on long-haul and overnight flights. In addition, there is also nothing to suggest that just because an airline may not be able to do anything about preventing a particular act (although in the case of sexual assault, we argue that there are actions that airlines can take to prevent, minimise or at the very least deal with a complaint appropriately), it should therefore be excluded from the scope of strict liability. Indeed, Judge McHugh in Povey v Qantas Airways pointed out that ‘[o]ne of the objects of the Convention is to provide compensation for injured passengers without the need to prove fault on the part of the air carrier…’

In summary, the current status of the courts’ interpretation of Art 17 of the Montreal and Warsaw Conventions (which impose strict liability against airlines for bodily injury occasioned by an accident on board) is that sexual harassment or assault will not attract compensation unless there is a physical injury – be that to the brain or some other part of the body, which can be evidenced to a court. As the majority of sexual assault incidents on airlines involve inappropriate or unwanted touching and do not extend to physical violence, there may not be sufficient evidence of bodily injury to satisfy the Art 17 criteria for compensation.

To add further insult to the lack of ‘bodily injury’ courts have also held that where the Montreal or Warsaw Conventions do apply, a claim in tort designed to evade its capture is not permissible. And so the likely conclusion for a sexual assault case against an air carrier is that if the Montreal or Warsaw Conventions apply, they will act to limit liability to those acts which cause bodily injury, and if there is no bodily injury, the passenger will likewise be limited from bringing an alternative action in tort against the airline.


Solving (or at least improving) the problem of the availability of damages requires action on the part of the international community (International Civil Aviation Organization (ICAO) and industry associations such as International Air Transport Association), individual states, and air carriers.

However, the two main areas that demand attention are:

1. expansion of jurisdiction to prosecute offenders of sexual assault on airlines to include the landing state, through the widespread adoption of the Montreal Protocol; and

2. expanding the scope of strict liability of air carriers for psychological injuries occasioned on board (including sexual assault).


Ratifying the Montreal Protocol would extend jurisdiction to prosecute offenders to the landing state, and raise awareness among all skyfaring nations of their responsibilities and options when it comes to prosecuting offenders of any type of criminal conduct on aircraft.

Another method is a legislative change to provide an avenue of compensation for victims of sexual assault on aircraft. Additionally, other initiatives such as onboard warnings and education may more likely be initiated by airlines themselves if and when liability is imposed by law on airlines.

The effects of such action would be swift and decisive. Making it clear that airline liability would accrue for passenger sexual misconduct would quickly result in further and better training being required of crews to handle the problem on board together with its aftermath (including criminal complaint and provision of support for the victim). It would also lead to clearer publication of expected standards of conduct of all passengers, which would effectively be policed by airlines in the same way that they currently attempt to prevent and prepare for non-sexual criminal activity through airline security risk management frameworks.

For states that do not presently pursue unruly passenger and sexual assault cases, adopting the Model Legislation of the Manual (replacing ICAO Circular 288) which recognises violent crimes against the person, including that of a sexual nature, as crimes which additionally threaten air safety, will assist matters substantially.

Legislating for a best practice framework for airlines and law enforcement that outlines the processes for managing and reporting complaints and incidents of sexual offences on board, and developing a formalised process of incident resolution to provide consistency of and direction to airlines, policy and regulatory bodies have also been recommended.

Using or expanding the scope of financial assistance schemes for victims of crime is something else which might be considered in the event that sexual offence complaints resolve with convictions under certain state-based schemes.


Recognising that psychological injuries are in fact a species of bodily injury and should be compensated along with ordinary bodily injuries is the key blocker to more widespread civil redress by air carriers for failing to prevent on board sexual offences to passengers. Air carriers have enormous power and responsibility over how passengers are treated during a flight and whether a person accused of sexual assault on a flight is likely to face justice or not. In recognising this power of air carriers and the corresponding vulnerability of passengers, the strict liability of air carriers for injury to passengers should be extended to assaults by other passengers that have ‘merely’ psychological consequences.

That the air carrier may not be able to prevent all incidents of sexual assault by a passenger is not a good enough reason to limit liability. There are actions which an air carrier could take and might be more inclined to take if there was the prospect of strict and positive liability being imposed on them. Such actions might include:

• sharing information between airlines of passengers who have been banned for sexual misconduct;
• incorporating standard procedures for dealing with suspected sexual assault and complaints of sexual assault;
• incorporating standard cabin crew training for such procedures;
• working with law enforcement at all ports/destination cities to follow up on complaints and prosecutions;
• taking more responsibility in the service of alcohol to passengers; and
• requiring cabin crews to be more vigilant on overnight flights.

Such actions are for the long term, as an international appetite and diplomatic effort would be required to conclusively establish the ability of victims to access compensation for losses as a result of such acts on board. In the meantime, vigilance, education and awareness are the key protective means by which we can keep ourselves and our co-passengers safe.