EgyptAir 990: Deliberate Crash Dispute

Introduction

The claim: EgyptAir Flight 990 was brought down on 31 October 1999 by its relief first officer, Gameel Al-Batouti, who deliberately put the Boeing 767 into a dive, rather than by a mechanical failure. Egypt's government rejects this and points instead to a possible fault in the aircraft's elevator control system.

On 31 October 1999, EgyptAir Flight 990 departed John F. Kennedy International Airport bound for Cairo, Egypt. The aircraft — a Boeing 767-300ER registered SU-GAP — had departed Los Angeles earlier and was operating the JFK-CAI leg. Thirty minutes into the transatlantic flight, at approximately 33,000 feet and 60 miles south of Nantucket Island, the aircraft entered a rapid dive. It broke apart before impact and struck the Atlantic Ocean. All 217 people on board were killed.

The crash occurred in international waters, so under international rules responsibility for the investigation fell to Egypt as the country of registry. At Egypt's request the US National Transportation Safety Board (NTSB) assumed full responsibility, with Egypt's Civil Aviation Authority (ECAA) participating through an accredited representative. The two agencies reached fundamentally different conclusions. The dispute between those conclusions is the subject of this entry.

The CVR and NTSB Findings

The cockpit voice recorder (CVR) and flight data recorder (FDR) were recovered from the Atlantic. The CVR and FDR show that, about 20 minutes after takeoff, relief First Officer Gameel Al-Batouti took the first officer's seat from the on-duty first officer, earlier than normal and at his own suggestion. At about 0148 EST the command captain left the cockpit for the lavatory, leaving Al-Batouti alone at the controls. About a minute and a half later the autopilot was manually disconnected, the throttles were moved to idle and the elevators moved nose-down, taking the aircraft to a pitch attitude of about 40 degrees nose-down within about 20 seconds. Al-Batouti said 'Tawakkalt ala Allah' — 'I rely on God' — eleven times, the first about 74 seconds before the dive began and the last just after the captain returned. As the aircraft dived, the captain returned to the cockpit and asked "What's happening? What's happening?" The NTSB noted that Al-Batouti did not answer. The captain then repeatedly urged him to 'pull with me' while the FDR showed the two elevator surfaces driven in opposite directions, until the recorders stopped.

The NTSB's final report, adopted on 13 March 2002 and released on 21 March 2002, gave the probable cause as "the airplane's departure from normal cruise flight and subsequent impact with the Atlantic Ocean as a result of the relief first officer's flight control inputs. The reason for the relief first officer's actions was not determined." It found that the nose-down movements did not result from a failure in the elevator control system or any other airplane failure, and that the Board "did not reach a conclusion regarding the intent of or motivation for his actions." The Board stopped short of stating that the crash was intentional.

The ECAA Position

Egypt's ECAA disputed the NTSB's conclusion from early in the investigation and maintained its disagreement through and after the final report. The Egyptian position, set out in EgyptAir's and the ECAA's submissions to the NTSB, rested on four arguments:

Mechanical failure: The ECAA argued that malfunctions in two power control actuators (PCAs) on the right elevator "may have precipitated the airplane's dive". It pointed to sheared bellcrank rivets and a sheared internal actuator pin found in the wreckage, and to FAA airworthiness directives on Boeing 767 bellcrank rivets. In its response to the NTSB's draft report it called the sheared rivets "direct evidence of a potential defect in the airplane's elevator system" that "may well indicate either a full or partial jam of one or more elevator PCA's". Its position was that further engineering analysis was needed before the cause could be established.

Language and context: The NTSB's first reading of 'Tawakkalt ala Allah' ('I place my fate in the hands of God') was later amended to 'I rely on God'. EgyptAir and the ECAA told the NTSB the phrase "is very often used by the Egyptian layman in day to day activities to ask God's assistance for the task at hand", and argued that the deliberate-act theory rested largely on the initial inaccurate translation. In its analysis the NTSB relied on the calm, repeated delivery of the phrase, beginning before the dive and continuing through it with no call for help, as inconsistent with how a pilot facing an unexpected flight condition would react.

No motive and no struggle: EgyptAir argued that Al-Batouti had no motive to kill himself or others, did not use his seniority to insist on flying, and that the captain returned almost immediately with no indication of a struggle or disagreement on the recording. It also said the FDR records control surface positions, not control column positions or forces, so it cannot by itself show who pushed which column.

Possible avoidance maneuver: EgyptAir said that, without more radar data, it could not be ruled out that the relief first officer was maneuvering to avoid a perceived danger. The NTSB found no evidence of conflicting air traffic in the CVR, FDR, ATC or radar data.

Boeing's Assessment and the Public Record

Boeing, a party to the investigation, told the NTSB in an October 2000 submission that none of the mechanical failure modes examined was consistent with the FDR data and that it did not believe the loss of EgyptAir 990 was the result of a mechanical failure of the aircraft or its systems. Boeing built the aircraft, and EgyptAir argued that the NTSB's simulations used Boeing's published 767 data and extrapolated beyond the speeds for which test data existed. The NTSB released the CVR transcript on 11 August 2000 with more than 1,600 pages of factual reports, and Egypt was given the opportunity to submit its disagreements in writing; its comments on the NTSB's draft report are attached to the final report, and its own report is in the public docket. The mechanical-failure explanation is set out in Egypt's own report; the NTSB's tests of four failure scenarios did not reproduce the recorded elevator movements.

Why the Dispute Persists

The dispute is genuine in one important sense: the NTSB's probable cause stops at the relief first officer's flight control inputs and states that the reason for them was not determined, so the step from 'his inputs caused the crash' to 'he did it deliberately' is an inference from the recordings rather than a finding. Egyptian investigators argued that the FDR cannot by itself show who was pushing which control column, and that the CVR contains no sign of a struggle. Egyptian aviation authorities also said, in a statement reported by CNN after the report was released, that NTSB investigators "rushed to judgment" before the wreckage was retrieved and that the NTSB "never gave equal weight or consideration to other accident scenarios".

The NTSB's probable cause did not state a motive or find intent. The absence of an established motive is the strongest element supporting Egyptian doubt about the deliberate-act theory. However, the absence of a known motive is not evidence of mechanical failure, and the NTSB found that no mechanical failure scenario matched the recorded data.

Verdict

Partially true. The claim that the first officer deliberately crashed EgyptAir 990 combines two parts. The first, that the dive was caused by the relief first officer's flight control inputs rather than a mechanical failure, is the NTSB's probable cause and is well supported: the NTSB found no failure condition in the elevator system, none of the failure scenarios it tested matched the recorded data, and Boeing reached the same conclusion on the mechanical question. The second, that he did it deliberately, is an inference the NTSB did not formally adopt: its report says the reason for his actions was not determined and reaches no conclusion on intent or motivation. The NTSB found no apparent reason for his nose-down inputs and no attempt by him to recover, and concluded that he moved the engine fuel switches to cutoff after the captain tried to recover, but no motive has been established and Egypt's ECAA formally disputes the deliberate-act theory. Egypt's mechanical-failure alternative was not borne out by the NTSB's tests, although the NTSB acknowledged that the wreckage alone could not establish whether an actuator jam had occurred. The dispute is a legitimate investigatory disagreement, not a straightforward conspiracy claim.

What Would Change Our Assessment

• New FDR analysis identifying a specific mechanical failure mode consistent with the dive profile
• Evidence of Al-Batouti's motive that was suppressed or overlooked in the original investigation
• Whistleblower testimony from NTSB investigators about pressure on the finding that his flight control inputs caused the crash

The Split-Elevator Evidence

The recorded split of the two elevator surfaces is the most technically specific evidence in the NTSB case. On the 767 each elevator surface follows the control column on its own side, and the split only occurs when the difference between the two column forces is large enough to engage an override mechanism. At 0150:21 the flight data recorder shows the left elevator continuing nose-up while the right reversed back to nose-down, and the surfaces stayed split until the recorders stopped. The NTSB concluded that the split was the result of flight control inputs to each surface and not of aerodynamic forces, and that tests in a 767 simulator and on a 767 on the ground showed pilots of the captain's and relief first officer's build could produce and maintain it. Egypt disputed the inference: EgyptAir argued the FDR records surface positions, not column positions or forces, so it cannot by itself show who pushed which column, and offered aerodynamic effects near Mach 1, or the loss of the right elevator, as alternatives. The NTSB's answer was that none of the mechanical scenarios it tested reproduced the recorded movements.

The Engine Shutdown

Both engines were shut down during the sequence. The FDR data show the engine start lever switches for both engines moved from run to cutoff between 0150:21 and 0150:23, just after the elevator split began, stopping fuel flow. The captain then advanced the throttles; when the engines did not respond he asked, "What is this? What is this? Did you shut the engine(s)?" A fault in the elevator system cannot move fuel switches, and the NTSB concluded that the relief first officer moved them, a counterproductive action for a recovery. EgyptAir countered that at least three flight crew members may have been in the cockpit, because releasing a control column to shut the engines or deploy the speedbrake should have pitched the aircraft down; the NTSB identified only the captain and the relief first officer as being in the cockpit during the accident sequence.

Could Any Failure Have Caused It?

The NTSB did not simply assert that the airplane was mechanically sound; it evaluated four dual-failure scenarios in depth, involving jammed or disconnected actuators on the right elevator or a cable jam combined with a break, and found that no mechanical failure scenario matched the recorded data. In simulations of the first three scenarios, full recovery was possible even when no recovery effort began until 20 seconds after the failure. The NTSB concluded that the recorded inputs instead drove and sustained the dive. EgyptAir argued that these tests used Boeing's published 767 data, extrapolated beyond the speeds for which test data existed, and that Boeing's engineering simulator "did not provide an accurate model of real aircraft performance".

The Physical Evidence Egypt Cited

Egypt's case was not purely cultural. ECAA investigators pointed to hardware findings recovered from the wreckage: sheared bellcrank rivets in the right elevator, and an internal pin sheared (with a bias spring coil out of position) in one power control actuator (PCA). They argued these could indicate a control-system failure that pitched the aircraft down. The NTSB could not determine whether the actuator condition pre-dated impact, but found no marks showing the parts had jammed and clearances too large for a jam, and considered it likely that the rivets sheared as a result of impact or recovery-related forces. It also acknowledged that examination of the structure alone could not establish whether an actuator jam or disconnection had occurred, and relied instead on tests and the FDR data, which none of the failure scenarios matched. The FAA separately issued airworthiness directives in 2000 and 2001 after failed bellcrank shear rivets were found on other 767s; the NTSB report records that the cause of those failures had not been determined.

Process Criticisms and Oversight

The NTSB held no public hearing on Flight 990: in August 2000 its then-chairman, James Hall, said none would be held because investigators "believe there are no unresolved safety issues". The Board's final Aircraft Accident Brief was adopted on 13 March 2002, after Hall had left the NTSB, and it framed the finding strictly around the recorded flight control inputs of the relief first officer, declining to assign a motive. That restraint is defensible but leaves the question of why unanswered, which is a large part of why the dispute has never fully closed.