Animation of Turkish crash...

[Kip Powick]

http://www.youtube.com/watch?v=zR6YPFIeemE

Lots of comments below the video

[Don Hudson]

The debate between accident investigators/flight safety people and pilots concerning this accident has been about two "discourses", or ways of thinking and speaking/writing; (there is a political discourse, religious discourse, scientific discourse, etc). The two discourses which are at work in the examination of this accident are, "blame" and "cause".

"Blame" tends to look at "who" as well as "what". "Cause" tends to look just at "what".

In an accident one can cite primary and secondary causal factors but generally do not talk about "primary then secondary blame". Perhaps this is because of the nature of the thinking and the resulting processes such as who is it who is going to be "responsible" and "accountable" and therefore sued, jailed or otherwise cited.

Both discourses have a legitimate place in accident prevention and improvement in aviation safety. It is a fact that our present English civil law has a very long history of the latter in the form of inquests which do not necessarily assign blame/responsibility, and criminal justice systems which can and do.

We have seen that where blame is placed on the pilot, there have been criminal prosecutions in Canada as well as France, Italy and Korea.

The push-and-pull between the two discourses, "finding out" and "assigning blame", has been at the heart of a number of Canadian cases as well as better-known ones in France, Italy, Korea and others. Notable exceptions are Germany and to a lesser extent, Britain.

Those who do flight safety work almost always focus on finding out what happened. The notion of "cause" is very broad and can take in many causal pathways, from primary to 'nn', in finding out what happened. Peter Ladkin's "Why-Because" analysis is a great example of such examinations. There are a number of others, some of which take into consideration the notion of MttB..."migration to the boundaries", a relative to the notion of the "normalization of deviance", which examines outliers and why people don't always follow SOPs and why short-cuts remain legitimate.

"Blame" tends to limit the dialogue, perhaps because "inconvenient" facts may water down a good case. At the same time however, flight safety investigators realize that under the notions of human factors, that "who" can matter. We can see that it can become..."sticky". Fatigue issues, standards, training, checking, preventative programs such as ASAP, AQP and FOQA as well as plain old competency always enter into the causal discourse to some extent but not with the notion of determining then apportioning "blame".

The idea of blame generally shuts people up. It is a fact that a non-punitive safety system opens people up to reporting errors and hazardous circumstances. This was the culture that was established not only on the Deepwater rig in the Gulf but demonstrably by the company that was responsible for the rig, Transocean. It is what flight safety programs are about...offering immunity from enforcement or discipline in exchange for information that will help enhance prevention processes.

Back to this accident...

It is a fact that the #1 radio altimeter which was controlling the autothrust had been logged as intermittent, that the aircraft had been dispatched several times under MEL with this fault, and that on approach to Amsterdam, the autothrust was commanded to 'idle' because of the fault and except for one intervention by the crew which returned to idle, remained at idle until the aircraft stalled just above 400ft AGL. Over a time period of a minute and fourty seconds, the approach speed of approximately 140kts had bled off to approximately 97kts.

A third pilot (an experienced F/O) was on board specifically to monitor the operation while the captain was training a new F/O... a "true" third set of eyes was the intention.

Yet the aircraft was permitted to stall, killing all the cockpit crew and six more in the cabin.

How do we look at this accident?

How do we use it to prevent others?

How does the animation help us understand the accident? Is context, (the report), important?

Don

[Mitch Cronin]

Excellent post Don.

[Seeker]

I'd be interested in finding out what exactly the 3rd pilot was doing during this. He was there specifically to monitor the flying of the airplane and he failed to do this; Was it because he was not actually watching what was going on? Was he competent enough to be able to recognize the errors as they were made? Was he looking out the window? Was he briefed on what to watch for? Was he afraid to speak up? Why did he not see this situation developing?

I'm no Chuck Yeager but when I'm sitting in the jumpseat I listen to ATC and confirm the readback and clearances myself, I watch the sequencing of the FMA and watch as the crew sets the altimeter, altitude alerter and D/H bugs - basically try to make myself useful. On several occasions over the years I have been able to spot an error and point it out to the operating crew and this is when I'm really just catching a ride home. If I was actually tasked with monitoring the flight I would be even more focussed on the operation. I do wonder about the wisdom of having an F/O as the safety pilot though - if the personalities are right this would work fine but if they were not this could be worse than having no one at all.

[Spinnaker]

I am not an Aircraft Engineer but the fix sounds simple. The Aircraft needs a comparator between the Rad Alts.

Boeing already has comparators all over the aircraft, NAV TUNE DISAGREE, FMC DISAGREE, ALT DISAGREE for a few... yet this critical instrument to the auto throttle system and its CAT 3 plus auto-land ability doesn't have one. The auto throttle only looks at the Captain's side and there is a secondary source mounted right next to it; they just don't talk.

[J.O.]

I'd be interested in finding out what exactly the 3rd pilot was doing during this. He was there specifically to monitor the flying of the airplane and he failed to do this; Was it because he was not actually watching what was going on? Was he competent enough to be able to recognize the errors as they were made? Was he looking out the window? Was he briefed on what to watch for? Was he afraid to speak up? Why did he not see this situation developing?

Very valid questions, but unfortunately they will never be fully answered.

It's been my experience that the third seat (or the IOS position in the simulator) provides a remarkable level of clarity when viewing what's going on in the flight deck. I've had a few occasions where jumpseaters pointed out things that my partner and I had missed completely. I tried to make a point of thanking them and always tried to encourage them to speak up when something didn't look or feel right. One thing that bothers me is the fact that I've never seen a very effective SOP for flights with 3 qualified crew in the flight deck.

I once investigated an incident involving an aircraft that slid off the pavement after landing when attempting to turn onto a taxiway. There were 3 in the flight deck, all strong and capable pilots. The flight was a short sector and the jumpseater offered to obtain the ATIS. He read out the weather conditions from the ATIS but omitted the fact that the ATIS included a caution for slippery conditions on the runway sides and taxiways (freezing rain had been falling a few hours earlier). The ground speed was hardly excessive when the skipper tried to turn off the runway, if one assumed they were only on damp pavement. But it was excessive for icy conditions. It's no guarantee that the incident wouldn't have occurred had the captain had this info, but it couldn't have hurt. I recommended that pilots stick to the regular SOPs and that the jumpseater should only monitor and raise any concerns, rather than actively participate in the flight.