Sunday, March 12, 2017

Consideration of Works Present: Sully


(Picture from here.)

I saw the film Sully last night (Starring Tom Hanks. Directed by Clint Eastwood.) and found it both compelling and irritating.

The directing was quite good and Tom Hanks’ performance was quite good. The film is about Chesley Sullenberger, or “Sully”, who was the pilot of US Airways Flight 1549 . This is the flight out of La Guardia where geese flew into the engines of an Airbus A320-214 causing both engines to cease to function. Captain Sullenberger had to land in the Hudson River.

The film is structured not directly around the flight (though, of course, the flight is central) but the NTSB (National Transportation Safety Board) investigation subsequent to the water landing.

And here’s where the irritation arises.

Though I haven’t flown for some time, I am a pilot. I’ve had the training, have my pilot’s license and have flown for many hours. I ran into health problems a while back and that dragged me out of the air. When those issues were resolved, the effort and expense to get back in the air was just too much so I didn’t return. But let me make something perfectly clear: I loved flying.

I loved just being up in the air. The idea of bringing the plane up to five thousand feet and just going somewhere was terribly exciting. The earth is beautiful from even a modest height. I was a fair weather flyer—a devoutly timid pilot. Once, I reserved a plane for a flight to New Jersey to visit New York. I woke up and it was absolutely clear. I checked the pilot’s weather line: chance of icing. I looked at the commercial weather report: clear and mid-forties all day. I rechecked the weather line: chance of icing.

I drove down to the airport. Checked it again. Chance of icing. I knew that it was just a stale report. It had probably been put in about 2:00 AM and for one reason or another had never been updated. I called again: chance of icing. Right up to my reservation time. So I released my reservation and drove down to New York. I called on the way. Sure enough: about an hour after I left Boston the report changed to clear and warm. But, as I said, I’m a timid pilot.

I also worked on several projects involving the design, construction, engineering and release of aircraft instrumentation. There are several hundred instruments flying in the air that I worked to develop.

Now, it’s important to know that the FAA and NTSB are a pilot’s friend and not his enemy. For example, the FAA has the AviationSafety Reporting System (ASRS). This is intended to incentivize pilots to self-report incidents. If a pilot self-reports an incidence the resulting report cannot be used for enforcement purposes. This means that a pilot can report something he did that was bad and not expect to lose his license or be prosecuted for it. (There are limits to this, of course. You can’t kill your parents and throw yourself on the mercy of the FAA because you’re an orphan.) Instead, the FAA requires additional training to overcome the circumstance that caused the incident.

Runway incursions are a good example. This is when the pilot brings the plane onto the runway when it is unsafe, such as when a plane is landing or taking off. The idea behind the incentive is to get to the pilot before an incident becomes an accident and both correct the problem and gather data.

Another example is the way that the FAA doesn’t charge for a lot of the actions required for flight. In other countries, the normal operations of flying are charged for: landings, tower interactions, flight following, etc., all incur a fee. Some large airports (like Logan or O’Hare) do charge a landing fee and some airports charge what is essentially a parking fee when you go in for a hamburger. But the vast, vast majority of airports, towered and not, charge nothing. I have made many hundreds of landings. If I had to pay a dollar for each one, it would be many hundreds of dollars. And, of course, I wouldn’t do it. I would land less. Pilots who learn to fly in countries that charge a fee have much less experience than pilots that learn here—which is why a lot of pilots-in-training come to the USA.

The NTSB is similarly structured. It is interested in truth, not blame. The NTSB reports (all of which are open to the public here.) are beautiful examples of dispassionate detail. I used to read them just to learn. Flight 1549’s summary report is here. The full report is here.

The investigation of flight 1549 is interesting. It’s in the Wikipedia entry for the flight. (See here, again.) They used a computer simulation and live pilot simulations. While the simulations suggested it might have been possible to reach either La Guardia or Teterboro, it would have required an instantaneous response—an unreasonable burden on the pilot.

The NTSB works with probabilities. At issue was Captain Sullenberger’s decision to land in the Hudson versus attempting to reach either of the two closest airports. The landing in the Hudson was a risk. Attempting to reach the other two airports was a risk. The question the NTSB had to answer was whether the risk of the water landing outweighed the risk of crash at the two airports. In order to do this, they also had to measure the probable loss of the 155 people on board with the additional potential loss of crashing in a densely populated area. (Remember, this was in 2009. 9/11 forcibly demonstrated the cost of crashing a plane in a densely populated area.)

It was not a determination of Captain Sullenberger’s competence. The fact that he had brought the plane down in the water with no loss of life amply demonstrated that. The NTSB determines what happened. The purpose is to make the skies safer.

Remember, probabilities are a measure of what we don’t know. For example, planes can get into spins. A spin is a stable rotation with insufficient lift to maintain altitude. Consequently, a plane in a spin will crash unless the pilot manages to correct the motion of the plane out of the spin. This is what “spin training” is all about. So, let’s say, we have five planes get into a spin and one crashes. You can say from that sample that survivability of a spin is 80%.

But that can be misleading. Some planes can’t get out of a spin. The Cirrus SR20 is an example. (See here.) So if your five planes are four Cessna 150s and one SR20, you’re analysis must be different. In the case of the SR20, it is 100% that the plane will crash without intervention and 100% the Cessna 150 is recoverable. (See here.) Consequently, an SR20 has a parachute for the plane. In the manual, if the pilot gets the plane into a spin the recover is to pop the chute.

Tom Hanks was in another similar film, Apollo 13. There was an investigation after that incident, too. This is spoken of at length in the book and only briefly mentioned in the film. Essentially, the event was caused by a long string of unlikely events that made the explosion inevitable. Similarly, once the birds were ingested by the engines on flight 1549 at that altitude, an emergency was inevitable.

The universe is deterministic. If we know everything necessary, the spread of possible behaviors narrows.

Which brings us (finally!) to the movie, Sully. After this there will be spoilers.

In Sully there is an antagonistic, almost prosecutorial, interaction between the NTSB investigation board and the pilots. The idea is that the NTSB is trying to blame Captain Sullenberger rather than find truth. Eastwood has been quoted as saying the NTSB tried to say Captain Sullenberger did the wrong thing. This flies in the face of my own personal experience and the experience of pilots that I know who’ve been through investigations. It is true that the NTSB attempts to determine all possible causes of an incident or accident. This is not railroading. This is good investigation. At one point Hanks told AP that Captain Sullenberger had reviewed the script and asked that the real names of the investigators be changed. They were not prosecutors and it was unfair to associate them so.

John Balzano, one of the investigators, went so far as to warn that this film might have a chilling effect on pilot reporting.  “The movie may actually be detrimental to aviation safety. Pilots involved in accidents will now expect harsh, unfair treatment by investigators.” (See here.)

This brings me to the idea of true drama, false drama and cheap drama. True drama derives from character driven conflict and the quest for the resolution of that conflict. False drama is when the reaction and quest for conflict resolution is done without any real conflict involved. Cheap drama is when the conflict is artificially contrived so that it can be resolved with the desired amount of effort.

In Sully, the NTSB's prosecutorial stance and behavior is contrived to create the dramatic tension and raise the stakes for the protagonist, Captain Sullenberger.

What’s problematic about this is that it was completely unnecessary and actually destructive. The true drama in the film is Captain Sullenberger’s wrestling with himself. Did he do the right thing? Did he make the right decision? He executed his decision flawlessly—remember, no one died. But was it the right one? Could he have just returned to La Guardia with nothing more than hard landing? If he had made the decision to return to La Guardia—and failed—the lives lost would have greatly exceeded the crew and passengers. The self-reflection should be agonizing.

And Hanks pulls this off. (This part of the New Yorker review I agree with.) He pulls it off so well that the whole NTSB motivation is actually distracting. It would have been much more compelling if the Eastwood had kept the NTSB as it was—a seeker of truth—and have it be the vehicle of Hanks self-exploration. As it is, he ends up fighting the NTSB and demonstrating it to be short sighted by forcing it to do things that in real life it already did.

I don’t know why Eastwood felt compelled to take this approach. Eastwood is a pilot and knows better. He is not known for cheap drama. He’s shown himself perfectly willing to tackle easy subjects the hard way. His film Unforgiven shows that. Perhaps he felt that the audience couldn’t handle such a subject in a complex way. Or that it was too subtle for us. Or perhaps he has an anti-government agenda that did not allow him to show a government entity in a positive light.

Regardless, he took what could have been a great film and made a poor one.