Looking at the cockpit recorder transcript from AF 447, two things seem clear to me:

  1. The two pilots flying (not the captain) did not know very well how to steer the airplane up and down, if the autopilot is down. It seems to me, flying the aircraft manually, in particular, up and down, should be the first thing they learn in flight school.

  2. All the pilots did not logically communicate and follow procedures for management in the cockpit. In particular, two pilots attempted to control the same aircraft functions at the same time, which seems, should be the first thing they learn in flight school, about cockpit communication, not to do.

(please correct me if my opinions are clearly incorrect).

How was this possible on a big flight on a big airline in the 21st century?

Should I assume that:

  1. In AirFrance, in 2009, a substantial portion of pilots, did not know such fundamental things?

  2. Or is true for most airlines?

  3. Or maybe AirFrance did not pay critical attention to hire pilots who can perform well under great pressure, which it seems to me should be a fundamental requirement for a pilot (to not forget everything when the shit hits the fan)

If if any of this is true, did AirFrance (and perhaps others) make fundamental improvements to their hiring and/or training after this accident?

Thank you

  • 2
    $\begingroup$ I'm voting to close. This calls mostly for an opinion, and very few are qualified to accurately critique their actions. $\endgroup$ – Michael Hall Nov 21 at 21:19
  • 3
    $\begingroup$ please correct me if my opinions are clearly incorrect While opinions are by definition not correct or incorrect, you present the facts in the least constructive way possible. Yes, stall recovery and CRM training were evidently not effective in preventing this particular accident, but that's a whole different league than "flying the aircraft up and down". -1 from me for reducing a complex topic to basically a childish insult. $\endgroup$ – Sanchises Nov 21 at 21:29
  • 3
    $\begingroup$ Downvoted. Read the report, crew is far from the only cause of the accident. $\endgroup$ – Efe Ballı Nov 21 at 22:41
  • 2
    $\begingroup$ @MarkGaleck 447 is best seen as a kind of perfect storm of elements that came together to create chaos.Use of "cruise pilots" to allow crew to rest. An event that nobody anticipated and nobody trained for because the phenomenon was poorly understood. Two inexperienced pilots, one of whom was perhaps at the left edge of the bell curve in terms of fundamental skills and instincts and went into brain lock right away, which ended up also overloading the other one. You sim train, but never know how you'll do with the real deal. When the real deal is an event from total left field, all bets are off. $\endgroup$ – John K Nov 21 at 22:47
  • 6
    $\begingroup$ @Mark Galeck, you may have tried your best, but this reads like a rant seeking validation. There is a time and audience for dispassionately analyzing the final actions of men whose mistakes cost them their lives, and critiquing the training and airline culture that may have contributed to the tragedy, but this is not it. $\endgroup$ – Michael Hall Nov 21 at 23:34

There are two factors at play:

  1. The training at the time, not just in Air France but worldwide, was somewhat lacking. Airline pilots were given training on stall prevention with a strong focus on minimising height loss, so a somewhat timid and gentle manoeuvre. But there was very little focus on actually recovering from a fully developed stall.

Pilots of light aircraft receive such training, but the stall characteristics of large swept wing aircraft are much different and so too is the recovery procedure - in a light aircraft you generally firewall the throttle, but doing that with big powerful jet engines can worsen the situation due to a strong pitch up effect.

Probably the biggest development arising from this accident is the introduction of Upset Recovery Training, or UPRT. This training exposes pilots to the recovery from extreme situations including stalls.

Now, even without UPRT training, the AF447 pilots probably should have been able to recover the situation if they could have properly understood what was happening. However:

  1. Pilots are human beings, and human beings cannot reliably perform under extreme stress 100% of the time. If you expose them to these (simulated) emergencies often, the pilots will have hopefully built a "muscle memory" to revert to when the real thing happens. But, sometimes all the training in the world can't overcome the human freezing up, as the relief first officer in particular did.
| improve this answer | |

To your question: "If if any of this is true, did AirFrance (and perhaps others) make fundamental improvements to their hiring and/or training after this accident?"

As regards the Training aspect, here is the Foreword to the ICAO Doc 10011 - "MANUAL ON AEROPLANE UPSET PREVENTION AND RECOVERY TRAINING". This was published in 2014, after a landmark study by all major players, and is a direct result of 447 and other such accidents. Implicit, is recognition that there was a gap to be filled, but not just for the 2 pilots, or a particular airline.


Between 2001 and 2011, aeroplane accidents resulting from a loss of control in flight (LOC-I) event were the leading cause of fatalities in commercial aviation. LOC-I accidents often have catastrophic results with very few, if any, survivors.

Following a conference in June 2009 on aeroplane upsets and loss of control in flight, the Royal Aeronautical Society (RAeS) initiated a study to investigate the LOC-I phenomena and make recommendations on mitigating strategies, notably with respect to potential improvements to international civil aviation standards and guidance material. This work was undertaken by the RAeS International Committee on Aviation Training in Extended Envelopes (ICATEE). ICAO supported this initiative.

In 2011, the Federal Aviation Administration (FAA) of the United States of America commissioned an aviation rulemaking committee (ARC) to develop effective upset prevention and recovery training methodologies. In 2012, ICAO, the European Aviation Safety Agency (EASA) and the FAA decided to combine efforts to identify and establish an acceptable approach to reduce such occurrences. ICAO sponsored seven meetings in 2012 during which Civil Aviation Authorities (CAAs), the FAA ARC and subject matter experts were encouraged to participate in focused discussions. Also, as several initiatives were underway simultaneously that sought to reduce the number of LOC-I events, ICAO brought many of the groups involved with these efforts into the ensuing discussions under what became known as the loss of control avoidance and recovery training (LOCART) initiative.

Reducing the number of LOC-I accidents is an ICAO priority, and ICAO has developed harmonized training requirements for flight crews that address and mitigate LOC-I events. Supported by ICATEE and the FAA ARC, ICAO has introduced improvements to existing Standards and Recommended Practices (SARPs) and corresponding guidance material. Both on-aeroplane training at the commercial pilot and multi-crew pilot level and training in a flight simulation training device at the airline transport pilot and type rating level are now promulgated in Annexes 1 — Personnel Licensing and 6 — Operation of Aircraft, Part I — International Commercial Air Transport — Aeroplanes, as well as in the Procedures for Air Navigation Services – Training (PANS-TRG, Doc 9868,), with an applicability date of 13 November 2014. This manual provides guidance to support these new provisions and is only applicable to the training of aeroplane pilots.

The content of the manual was developed over a period of three years with input from many groups of experts from aircraft and flight simulator manufacturers, pilot representative organizations, airlines, training organizations, accident investigation bureaus, human performance specialists, etc., and was thereafter submitted for an extensive peer review to collect and take into account comments from the expert community. It is based upon the latest forms of technology available at the time of its publication. As such, it will be subject to a revision process that will be governed in large part by changing dynamics within the industry. Comments on this manual, particularly with respect to its application, usefulness and scope of coverage, would be appreciated. These will be taken into consideration in the preparation of subsequent editions. Comments concerning the manual should be addressed to:

The Secretary General
International Civil Aviation Organization
999 University Street, Montréal,
Québec H3C 5H7

I hope this brings a better perspective.

| improve this answer | |

Not the answer you're looking for? Browse other questions tagged or ask your own question.