I am here today to brief you about the progress in a case that has marked my professional career and that, as I want to believe, has had a certain impact in the aviation environment.
However, and as I will explain now, we are still waiting for the last word from the Spanish Supreme Court.
Many of you know the basic facts of this terrible tragedy, where around 60 Russian adolescents from the remote Republic of Bashkiria lost their lives in the airspace of South Germany, back in 2002
Two airplanes (Bashkirian Airlines and DHL) collided while cruising at high altitude. They were the only aircrafts in the area at that time of the night. Something that seems impossible to happen.
But it did happen.
The investigation handled by the BFU, the German Official Investigation body, appointed in conformity of Annex 13 of the Chicago Convention, revealed substantial deficiencies at different levels: ATC management, crew human factors and other associated technical and operational problems.
Basically, modern aircrafts carry a system denominated TCAS (Terrain Collision Avoidance System). It is a coordinated device, which means that planes “talk” to each other in order to avoid conflicting trajectories and potential air collisions.
When two aircrafts engage in a potential conflicting course, the device will alert both crews to be aware of the situation. If the conflicting situation persists and the risk of a mid-air collision grows, the device will then trigger a coordinated warning, instructing both conflicting aircrafts to proceed into opposed vertical maneuvers, to create a safe separation between them. If, by any reason, the conflict still persists (I.e. if one of the aircrafts reacts in opposition to the TCAS instruction), the system includes an algorithm that reverses the initial order, encouraging now the aircrafts into the opposite directions.
This is an autonomous system, which is independent of the ATC operation, and it is considered as the last line of defense, when i.e. there has been a failure in the ATC management or elsewhere.
In the Überlingen scenario, a deficient management of the ATC resources, which included human and technical factors, initiated the traffic conflict. The TCAS then intervened by alerting both conflicting aircrafts and subsequently instructing one of them to climb and the other to descend. At the very same moment when the TCAS was instructing the Russian crew to climb, the ATC stressed them to an expedite descent.
The crew, confused by the simultaneous conflicting instructions, took the decision to follow the human orders rather than the automatic coordinated TCAS instruction. As a consequence, both conflicting aircrafts simultaneously descended. At that point, a TCAS reversed Resolution Advisory would have made the DHL crew to stop descending and start climbing, while the Russian crew would have kept descending, creating therefore a substantial vertical separation enough to avoid the collision. But this reverse alarm did not engage and, some seconds later, both aircrafts collided, causing the death of all their occupants.
The BFU concluded, in addition to the evident ATC deficiencies, that:
- The Bashkirian Airlines pilots were not properly trained and informed about the use of the TCAS (which was not yet in service in Russia at the time of the accident).
- The design conditions for a reversal resolution advisory were not met.
It was also made clear that the TCAS logic did not contemplate this type of conflicting encounter were an aircraft, for whatever reason, does not follow TCAS or even engages in an opposite direction. This was astonishing as this appeared to be one of the paradigmatic scenarios in which this reverse warning should enter into play.
Further private investigation revealed that two years before the instant accident, EUROCONTROL had warned the U.S. authorities and manufacturers about this safety issue, known as SA01, stating that it created a risk of a midair collision probably more than 25 times the maximum acceptable risk. EUROCONTROL even proposed a solution, in the form of a software update in the TCAS system, named CP112.
Unfortunately, that specific version of the TCAS system, known as 7.0 or TCAS 2000, had just been released into the market and manufacturers and certification bodies were not particularly rushed to introduce such early modifications into the product.
Two near misses occured by this already known safety issue happened during 2001, but still the industry did not react timely, and the catastrophe finally happened over Konstanz in early July 2002.
What is even sadder is the fact it still took seven years to make the EUROCONTROL update proposal be implemented, hidden and lost in a major update pack.
BFU also noticed that the Manufacturer TCAS Pilot Guide did not reflect the philosophy of the TCAS instrument, leaving room to interpretation about who had the last word in case of conflicting or even contradictory instructions, as it had happened in the Überlingen accident.
Finally, our investigators found that, despite the conclusions of the BFU, there were enough factual evidences to conclude that the design conditions for the Reversal Resolution Advisory were effectively met, and that, as a consequence, the specific TCAS device had failed to recognize that they were in presence of a Reversal scenario.
So, to summarize, three criticisms linked to the accident could be made to TCAS manufactures:
- a) Inaccurate product use information.
- b) Poor design in the reversal instruction algorithm, as revealed by EUROCONTROL.
- c) Defective manufacture, as the alarm should have actually triggered a reversal instruction.
A group of victim’s families filed a product liability action against the TCAS manufacturers in the United States. The case was expectedly dismissed through a Forum Non Conveniens decision, pointing to the courts of Barcelona (a major city in Spain, for the time being).
Why Barcelona? Barcelona was the final ticket destination for most of the passengers in the Russian plane, and a Warsaw action had already been filed over there on behalf of the same plaintiffs. The district court in New Jersey considered that, for that only reason, the case would be better served if it was not tried in the US and instead victims re-filed their cases in Barcelona.
When I received the first call from our American colleagues, I thought, well, “this is going to be a tough case”. Then I remembered the 1973 Hague Convention, ratified by Spain. A manufacturer counsel had raised this international set of conflict rule to us in another previous case in France, pretending the application of the law of Bahrein, which meant that the “price of blood” had to be awarded, that women would get half compensation of men and even, to some extent, that damages would have to be evaluated taking into account the market value of camels.
Anyway, I took a look to the conflict rules at the convention and then realized that the case was going to be even tougher than what I had initially thought.
For those who are not familiar with this Convention, you can read articles 4 to 6 (in the materials).
Because the State of the place of the injury was not the place of the habitual residence of the person directly suffering damage, or the principal place of business of the person claimed to be liable, or the place where the product was acquired by the person directly suffering the damage; and also because the State of the habitual residence of the person directly suffering damage was not the place of the principal place of business of the person claimed to be liable, or the place where the product was acquired by the person directly suffering damage, then, the applicable law should be the internal law of the State of the principal place of business of the person claimed to be liable, unless the claimant based his claim upon the internal law of the State of the place of injury.
So, according to the Convention, we were able to re-file the case in Spain, based in US law and apply, between other things, for US level damages. We performed a research throughout our network and asked colleagues for decisions in which this convention might have been applied or even cited, but we found almost nothing.
The truth is that this convention, although in force, had had very few ratifications and its application was quite limited in space. We were in fact entering into a totally new and unexplored territory.
A complaint was therefore filed in Barcelona after several months of background work, study and preparation. The finest defense counsels possible were appointed, and they reacted vigorously after their initial surprise when they realized we were claiming the application of the U.S. Law.
After almost two years of legal obstacles expertly placed by the defendants, which included incidents relating to translation of documents, unsuccessful tries to consolidate the product liability case with the Warsaw action which was pending in another Court in Barcelona, even motions to dismiss for lack of standing to sue, as well as a fight for the disclosure of the TCAS source code, we were finally able to get the case to trial.
Technical experts, academics, justices and other witnesses would have to show up in Court in order to be cross-examined by the parties and interrogated by the court.
The hearings lasted for nearly two weeks in daily sessions of eight to ten hours. Almost every witness and expert was foreign and required live language interpretation, which doubled or even tripled the necessary time and resources for each deposition. Just as an anecdote, counsels of both plaintiff and defense parties were admonished several times because of unintentionally cross-examining in English. This would typically happen late afternoon, when approaching to the end of the sessions…
The Courtroom, quite small and modest by the way, was everyday crowded with family members and foreign lawyers. The expectation was significant and it was a continued learning experience for all the participants, including the fine Judge Mrs. Queral Carbonell.
Despite the vigorous and outstanding defense strategy deployed by the defendant counsels, the District Court partially accepted our claim, stating the following:
IT IS HEREBY DECLARED that the TCAS Pilot’s Guide of the Defendants, lacks appropriate instructions and warnings for a correct, efficient and satisfactory use of the TCAS in accordance with the logic of the system (information defect); that such information defect contributed to one of the causes of the aviation accident occurred on 1 July 2002, which resulted in the death of the plaintiffs’ relatives, thus substantially contributing to the damage caused, for which reason the defendants shall be held liable for damages; and that pursuant to the Hague Convention on the Law Applicable to Products Liability, X is subject to Arizona Law and Y is subject to New Jersey Law.
Basically, the District Court accepted ONLY the “failure to warn” count but dismissed the other “Design” and “Manufacture” allegations, as I have roughly explained them before. The Court concluded that the TCAS contributed in 40% to the accident, and made the defendants liable ONLY for their share of contribution.
Defendants appealed the Judgment, so did the Plaintiffs.
Defendants challenged the application of The Hague Convention and wanted instead the application of the Spanish Civil Code general conflict rules, which would likely provide the applicability of German Law based on the principle of lex loci delicti commissi. The argument for this was that the TCAS instrument could not be considered as a “defective” product in the sense of the Convention.
Alternatively, if the Court concluded that the Hague Convention was applicable, they requested the Russian Law to be applied, as being the law of the domicile of the victims and the place where the victim had “acquired” the product. Alternatively the Court should apply the law of the forum because the absence of any proof regarding Russian Law.
Alternatively, in case Hague Convention applied and US Law was applicable, defendants sustained the “federal preemption” defense, according to which, and based in the federal aviation regulations and the product certification process, state product liability law was not available as a remedy for the plaintiffs.
In any extent, the defendants should not be liable because the TCAS Pilot’s Guide was not misleading, and of course because, as already agreed by the District Court, the TCAS device was neither defective in its design or manufacture.
Plaintiffs also appealed, principally over the following aspects:
- The product would be also defective in relation to design and manufacture.
- The liability should be exclusively attributed to the defendants.
- The extent of damages should be awarded in accordance to U.S. Courts practices has it had been shown during the trial at the district court level.
The Court of Appeal of Barcelona followed most of the arguments raised by the plaintiffs. I am now quoting some excerpts of the decision:
In relation to the Hague Convention Applicability:
“It is not clear from the reading of these Articles what is meant by a “product defective in the legislative sense”. Firstly, this is because the Convention at no point mentions the term “defective”. It is therefore clear that in order to decide whether the Convention is applicable, we do not need to establish what is or what is not to be understood by the term “defective”.
It is sufficient that a liability action be brought against the manufacturers of a product for damage caused by the product for the Convention to be applicable, it being the Convention which determines the Law applicable to the case.
In short, given that The Hague Convention is applicable and, as the lower court has established in its judgement, Articles 4 and 5 of such Convention are not applicable, the Law to be applied must be determined based on the provisions of Article 6.
Therefore, the conclusion reached by the judge of the lower court is perfectly correct, such conclusion being that “the Law of the State of Arizona is applicable to this case in respect of the defendant X and the Law of the State of New Jersey is applicable to it in respect of Y.”
In relation to Federal Preemption:
The fact that the product had been certified does not guarantee that it was manufactured in accordance with the certification, or that all the safety requirements applicable were met. Meeting a standard of fitness set by Federal regulation is not the same as to believe that holding a federal certification means the standard is met. Thus, for example, a toy manufactured to conform to UE standards offers greater assurance than one that is not so manufactured, yet that does not release the manufacturer from liability if a child is proved to have suffered an accident because the toy is shown to be dangerous.
In relation to TCAS design:
“Following EUROCONTROL reports, the Court of Appeal concludes that there is no doubt that serious safety problems have been detected in the version of the TCAS 2000, which the two aircraft involved in the crash were fitted with, revealing major design problems.”
The Überlingen accident and the repetitive serious incidents resulting from the SA01 and SA-AVSA safety incidents could have been avoided with the version 7.1 of TCAS II.
For two years the manufacturers of TCAS knew about the severity of the software design problems specified in the detected safety questions and they did nothing.”
In relation to TCAS manufacture,
I am not going to enter into the technical details because this would take me hours of technical explanations, I will simply say that the Court of Appeal found, following expert evidence provided during the trial, that the logic conditions for a reversal resolution advisory were actually met several times during the conflicting situation, but the TCAS processors were incapable of detecting it, due to manufacturing defects.
In relation to Pilots Guide defects:
The Court of Appeal also found that the TCAS Guide for Pilots was inaccurate, in the sense that it was misleading as to how to react in case of opposed instructions between TCAS and ATC.
This defect substantially contributed to the Tupolev reaction against the TCAS but in compliance with the ATC instruction.
Following a RA instruction is not just a recommendation, as the BFU found. It must be clear that RA has to be followed always and under any circumstances. And the TCAS manual was not as categorical as required, permitting the crews to react otherwise.
And as a result of the preceding three defects:
The entire foregoing permits the relation of direct exclusive causality between the design, manufacturing and information defects of the TCAS II, version 7, and the Überlingen accident of 2 July 2002, in which the relatives of the present plaintiffs died, to be affirmed. There are few reproaches that can be made to the crew. And even the poor functioning of the Zurich Air Control Centre was not the final cause of the collision. The TCAS generated an RA, which was not followed by the crew of the Tupolev and this, if the TCAS had been functioning correctly, ought to have provoked a reversal RA which did not occur.
And no reversal RA was generated in spite of the fact that the technical conditions for doing so were present.
The reversal RA would undoubtedly have prevented the accident and this was what was expected of it. Therefore, no responsibility for the accident can be attributed to third parties. The responsibility was exclusively that of the defendants.
As to the damages, and based in expert reports and affidavits provided by both plaintiffs and defendants, the Court of Appeal establishes de compensation of pecuniary and non-pecuniary damages as follows:
- As a general rule, 3.000.000 USD for the relatives of each passenger, to be paid 50% by each of the two defendants.
- 7.000.000 USD for the relatives of two brothers both deceased in the accident.
- 8.000.000 USD for a plaintiff that lost her husband and his only child.
- 11.000.000 USD for a plaintiff that lost his wife and his only two children.
This was roughly 20 to 30 times less of what a Spanish court would have awarded pursuant common Road Traffic Accident standards.
- It is the first time that the 1973 Hague Convention is applied to its latest consequences.
- It is the first time that a European Courts applies in full extent U.S. law to a tort case, and award damages pursuant U.S. Standards, if any.
- It is the first time that a non U.S. Court awards U.S. damages consecutive to a FNC decision.
As I quoted at the beginning of my intervention, the case is now awaiting decision from the Supreme Court, resulting from the final appeal filed by defendants, ruling which is expected for this year.
What lessons can we learn from this case? I would say two:
- TCAS is a critical instrument that has substantially contributed to commercial aviation development, allowing reduced separation between aircrafts and therefore increasing traffic operations. But due to its conception as the last line of defense in the air traffic management system, certification bodies and manufacturers need to pay close attention to safety above merely market considerations.
- It is true that the 1973 Hague Convention has a limited scope of application. But what it resulted from this case (until now), makes us think than FNC motions might not always be the best defense for U.S. defendants.
I do not want to finish my intervention without expressing my sincere and profound gratitude to the offices of Pordhurst & Orseck in Miami, as well as to Gustavo Fuentes, because it was with their outstanding support that this case and its outcome was possible.
But remember, we may have a new chapter of the case after the Supreme Court speaks, hopefully this year.
I wish to thank you my friends at Peopil and the McGill University for inviting me to participate in this thrilling event.
Thank you very much.