How to make aeronautical communication safer

[Machjo]

According to the late Kent Jones, an estimated 15% of air crashes are caused by simple miscommunication between pilots and air traffic controllers. The following article of his points out some of the current problems in FAA communication:

Misfunctional FAA phraseology

The problem continues today as the following luckier examples show:

YouTube - ATC@JFK - Air China 981 (by aldo benitez)
Polish pilots' poor English almost led to crash -Times Online

So what is the solution? Among the solutions I could see would be:

1. to turn to an easier-to-learn yet grammatically more precise language. (though I could certainly see native English speakers objecting to their loss of such a privileged position).

2. Require higher English-standards for pilots, resulting either in foreign countries having to hire native English-speakers from abroad or cutting back on flights. (I could see some major political fallout from that position in non English-speaking countries).

3. Continue with the current status quo and accept the risks.

4. Some other solution? One possibly solution could be that English-speaking countries be required to offer free English-courses to foreign pilots and ATCs while requiring foreign pilots to learn English. This could be a possible compromise to raise safety standards while overcoming the possible political fallout that would be caused native English-speakers taking over all the jobs or poorer countries having to foot the bill for second language courses while English-speaking countries benefit from cost savings on that front. And of course there may be other solutions too.

What solution do you think would be best?

I especially like the Anglocentrism of this article:

Polish pilots' poor English almost led to crash -Times Online

"The incident, described in a report by the Air Accidents Investigation Branch (AAIB), highlights the risks associated with having so many foreign pilots with only rudimentary English using British airports. English is the international language of aviation but many countries failed to comply with the International Civil Aviation Organisation (ICAO) deadline of March this year for ensuring that their pilots were proficient in the language."

It puts the fault squarely on the non-native English-speakers of the world. It does not even consider what native English speakers could do to improve the situation, as if it's their God-given right to use English the world over and that the onus is on the other pilots of the world to use English in all of the world's airspace. Amen!

And I especially like this gem:

"The letter, sent on March 4, nine months after the Heathrow incident, states: “We haven’t had any accident caused by insufficient English level as well as [sic] we have not received any information from air traffic control agencies that Polish pilots were not able to communicate correctly. Therefore, potential risk of accident occurrence due to lack of English language proficiency is very improbable.”"

If the English learnt English, that would probably help too.

And then you run into this:

"David Learmount, safety editor of Flight International magazine, said that it was very worrying that the pilots had not understood British controllers. “It could have been even worse if they had been trying to understand a French controller speaking to them in English,” he said."

Well thank God it wasn't a French ATC trying to communicate with these Polish pilots.

 

[lone wolf]

Let the computers do the talking.

 

[TenPenny]

Malcolm Gladwell wrote an interestig article, a lot of it has to do with the different cultures, not simply the words spoken, but the cultural differences. For example, many pilots defer to the NYC air traffic controllers, because they're so aggressive and rude, and this resulted in at least one plane crash where the jet ran out of fuel, because the pilots were so intimidated they couldn't get their point across about being low on fuel...Not that they didn't understand the words, just that the communication wasn't good.

 

[Machjo]

Let the computers do the talking.

Believe it or not, that does in fact happen. In the case of the Polish pilots at Heathrow referred to above, the problem was precisely that owing to technical problems, lo and behold, humans had to do the talking.

Malcolm Gladwell wrote an interestig article, a lot of it has to do with the different cultures, not simply the words spoken, but the cultural differences. For example, many pilots defer to the NYC air traffic controllers, because they're so aggressive and rude, and this resulted in at least one plane crash where the jet ran out of fuel, because the pilots were so intimidated they couldn't get their point across about being low on fuel...Not that they didn't understand the words, just that the communication wasn't good.

Granted. Another unrelated problem is that grammatically speaking, English is not the most precise of languages either. Though it is related to cultural issue to the extent that when there is a risk of cultural misunderstanding owing to ambiguity, a more precise language will certainly have an advantage over a less precise one.

As for the ATCs being rude, that's a separate matter that I don't believe any language planning alone could solve. For that the ATCs would need to learn more professional behaviour quite simply. Rudeness is just a sing of lack of culture in any culture. But again, that would be a separate matter from the strictly linguistic aspects of the problem, though granted it certainly compounds the problem too, especially when you're dealing with a second-language speaker. I remember one case of a person who was fluent in English when I spoke to her casually, but as soon as she had to give a presentation for a job interview, her English went to trash. When we're dealing with pilot emergencies, the last thing we need is an otherwise fluent pilot in English getting lost in all the exceptions to the rules of English grammar, homonyms, synonyms, etc. etc. etc. Not a good time to be worrying about the intricacies of a complicated language like English along with its ambiguities. I don't want to be sitting in that plane.

It's interesting that someone voted for the fourth option. Would we really be willing to foot the bill for this?

 

[lone wolf]

Believe it or not, that does in fact happen. In the case of the Polish pilots at Heathrow referred to above, the problem was precisely that owing to technical problems, lo and behold, humans had to do the talking.

That would be why they can't remove the human factor from the cockpit. Learning another language has its drawbacks too - especially when someone panics and reverts to native tongue. Nobody said the human factor is infallible.

 

[Machjo]

That would be why they can't remove the human factor from the cockpit. Learning another language has its drawbacks too - especially when someone panics and reverts to native tongue. Nobody said the human factor is infallible.

But certainly an easier-to-learn language would increase safety considerably. After all, the more comfortable one is in his second language, the more likely it is that he'll still be able to function in it in an emergency situation. With English, he's more likely to be scrambling for the right word in his mind among so many exceptions to the grammar rules. That's the last thing you want a pilot to be doing in an emergency but with English that is in fact more likely to occur. And again, that's not the pilot you want to be flying your plane now is it.

 

[lone wolf]

But certainly an easier-to-learn language would increase safety considerably. After all, the more comfortable one is in his second language, the more likely it is that he'll still be able to function in it in an emergency situation. With English, he's more likely to be scrambling for the right word in his mind among so many exceptions to the grammar rules. That's the last thing you want a pilot to be doing in an emergency but with English that is in fact more likely to occur. And again, that's not the pilot you want to be flying your plane now is it.

A new language does not prevent any operator from reverting to native language in a panic situation. A mid-air collision over Croatia a few years back was the direct result. One has to have faith in the wire that really flies the plane and the good ol' Mk 1 eyeball.

 

[Machjo]

A new language does not prevent any operator from reverting to native language in a panic situation. A mid-air collision over Croatia a few years back was the direct result. One has to have faith in the wire that really flies the plane and the good ol' Mk 1 eyeball.

That's why I said an easier language would increase the chances, not that it would be a foolproof solution. To take an example, if we switched to a much easier language, let's say 5 times easier or more, then while the possibility of reverting back to the mother tongue would still be there in the event of all out panic, it's less likely to occur under moderate stress. With English, language faculties are more likely to break down at a lower level of stress relatively speaking, thus increasing the risk considerably. I'm looking at it along a spectrum, not so much along the idea that there is a 100% perfect solution. And looking at it that way, it's clear that an easier language, while not perfect, would improve chances of survival considerably.

 

[lone wolf]

Can you talk as fast as an airplane can fly?

 

[Machjo]

Can you talk as fast as an airplane can fly?

?:?:

 

[lone wolf]

?:?:

Exactly. Electronic ATC can - and has to. No new language, so matter how simple it is, is going to make the human at the controls react any quicker. There already is a standard language of the air - and it was designed in the days of piston-pounders and propellers. There are a thousand times as many airplanes moving at ten times the speed now. There is no time for talk.

 

[bobnoorduyn]

English is the most widely spoken language so it makes sense that it would be the primary one in aviation. What is strange is that Canada is the only country I know of that has two official languages in aviation. On several occasions I have been addressed in French, this happens most often when two or more flights have similar flight numbers and one is speaking French. It is not just us, Air France, Royal Jordanian, Royal Air Moroc, and others will use French in Montreal as well.

On more than one occasion, Montreal crews have been turned back and sent home by Boston and New York Center because of the language barrier. The Air China episode exemplifies another problem besides language; American airports especially are a mess, but Toronto can also be quite the challenge too. If someone is going to get lost, misread or misinterpret a clearance, go somewhere they aren't supposed to, it's going to be on the ground. Signage and markings can be confusing at best, or non-existant at worst. Toronto has some taxiways that aren't marked, or if they are you are already on it before you know which one it is.

Automation was originally designed and installed to reduce workload, it is now used to cram as much aluminum into as little airspace as possible. More errors are made because of automation than lack of it. A common point on a certain arrival into Montreal is the OMBRE fix, commonly pronounced "ombray"; some French controllers have francophonized this into "ohmb". "Where the hell is that" is the usual response. I was once cleared to an unfamiliar fix, typed it in the way it was spoken, (with one letter wrong) and was directed to somewhere in space some 3400 miles in the opposite direction. (These are computer generated names, some of which are unpronouncible).

Of course, all those flights have to land. You heard the ground controller, this is not uncommon, giving a long list of instructions, crossing or hold short restrictions at machinegun speeds. They have tin to push and have little patience if you don't get it right the first time. Language is an issue, but not the only one.

As for an emergency in the air, crew communication is done in whatever toungue they speak normally, ATC communication is done in English, (except in Quebec or Ottawa where you have a choice). But when **it hits the fan the rules are, Aviate - Navigate - Communicate. The emergency is usually dealt with by the time English will be needed for ATC. There are standard rules and phraseology for standard ops, and for most unusual situations, of course there can't be rules or procedures for every conceivable situation, that's where sound judgment is required. No amount of technology can replace that.

 

[Machjo]

Exactly. Electronic ATC can - and has to. No new language, so matter how simple it is, is going to make the human at the controls react any quicker. There already is a standard language of the air - and it was designed in the days of piston-pounders and propellers. There are a thousand times as many airplanes moving at ten times the speed now. There is no time for talk.

If language is not important, why is it estimated then that about 15% of air crashes are caused by miscommunication? If what you say were true, then language should not be an issue at all.

And again, in the case of the polish pilot at Heathrow, the reason linguistic communication suddenly became paramount was precisely because that technology you laud failed. When that happens, would it not be wise to have a backup plan, and if that involves communication, ensuring that they can?

 

[lone wolf]

If language is not important, why is it estimated then that about 15% of air crashes are caused by miscommunication? If what you say were true, then language should not be an issue at all.

And again, in the case of the polish pilot at Heathrow, the reason linguistic communication suddenly became paramount was precisely because that technology you laud failed. When that happens, would it not be wise to have a backup plan, and if that involves communication, ensuring that they can?

Where did you read language isn't important? I don't think I said that. That's miscommunication and you just crashed.

 

[AnnaG]

Let the computers do the talking.

Yup. Talk to the computers and let them talk to each other.

 

[Machjo]

Exactly. Electronic ATC can - and has to. No new language, so matter how simple it is, is going to make the human at the controls react any quicker.

This tells me that if the tech crashes you're screwed anyway, because it 'has to' work. After that, there's really no point in trying. Just pray as you go down.

FYI, pilots are not that enslaved by the technology. Airliners have been known to land safely on the ground after running out of fuel long before they even land. This tells me that even in the modern age, even when technology fails, if you can communicate effectively with ground control, you can still negotiate a path to an open field or abandoned highway somewhere in the worst case scenario. So no, while tech failure is dangerous and not an ideal scenario, it does not 'have to' work to guarantee survival as long as you can ensure others around you know what you intend to do so that other planes can get out of the way if you have to go down quick.

There already is a standard language of the air - and it was designed in the days of piston-pounders and propellers.

No, it was not designed in the days of piston-pounders and propellers. The modern English language was developed ad hock around the 1600s when there was minimal international communication, let alone communication within a context that can determine life and death.

And as for that language's success rate, well I guess an 85% success rate is a pass, in the strict sense that it's more than 50%. But when we're talking about life and death, is 85% really good enough?

There are a thousand times as many airplanes moving at ten times the speed now. There is no time for talk.

If there is no time for talk, and talk is useless, then why bother have radios on the plane or require them to learn any common language at all?

Yup. Talk to the computers and let them talk to each other.

:lol:

Where did you read language isn't important? I don't think I said that. That's miscommunication and you just crashed.

Read above. If that's not what you said, then you can see how I came to that conclusion.

 

[lone wolf]

Are you adding details and moving goalposts?

 

[Machjo]

Again, getting 85% on a test is not bad, but having an 85% chance that your plane crash won't be caused by simple miscommunication is shameful.

Now I know you could turn around and say that our priority ought to be to focus on the other 85% of causes for air crashes, which could involve a number of things such as pilot error, lack of sleep, technical problems, running out of fuel, terrorism, etc. etc. etc. However, that 15% caused by simple linguistic communication seems to be such a simple one to solve. That's what makes this one so different from the others. If an engine falls off, your options are limited. But certainly there must be a more efficient way to improve communication between pilots and ground control that wouldn't cause political opposition, and so reduce crashes by 15%. That alone would be quite a few lives saved.

Again, I'm not saying the causes for the other 85% of crashes aren't important and need to be addressed, but it's that this one is so easy to solve that if a rational policy were in place, crashes caused by the language barrier should amount to no more than maybe 1% of crashes.

 

[lone wolf]

Language barriers are going to happen anywhere two different languages contact. How much of the miscommunication is language and how much of it is cultural difference?

 

[bobnoorduyn]

No, it was not designed in the days of piston-pounders and propellers. The modern English language was developed ad hock around the 1600s when there was minimal international communication, let alone communication within a context that can determine life and death.

Aviation language has evolved since the "piston pounders", but was originally designed back then and has little resemblance to the Kings tongue which you speak of. Radios were hard to hear and certain phraseology was developed to reduce misinterpretation. I don't know who Kent Jones is, but the article shows many interpretations of aviation jargon that are simply just fanciful. The term "feathered" for a propeller has been around since propellers could be feathered, and exists in standard emergency drills today. Other terms such as "back taxi", (which is not as common as "backtrack"), "wheels up", "go around" are understood by students before they ever leave the ground for the first time. It starts to fall apart when local jargon or nuances are mixed in that stray from standard phraseology. The US can be particularly bad, "Cactus XXX, taxi to position runway XX left and hold" becomes, "Cactus XXX position and hold on the left side". Seattle had a nasty habit of clearing you for an approach to XX right then saying "sidestep to the left side, cleared to land", we understood it where a Russain might not. Canada's newest standard phraseology is, "Eagle XXX line up runway xx and wait". This throws the Americans for a loop, you expct to hear one thing and get another. Of course this has been shortened to "Eagle XXX line up runway XX". (And now what?). Some of what is said in the name of expediency leaves even the most proficient in the lingo scratching their heads.

And as for that language's success rate, well I guess an 85% success rate is a pass, in the strict sense that it's more than 50%. But when we're talking about life and death, is 85% really good enough?

Statistics get thrown around for all kinds of purposes, a certain poster lives by them, as do those with an agenda. I'm not sure where this one came from. Accidents rarely have a single cause, and miscommunication or misinterpretation do play a part in many, usually when dealing with an already on-going emergency. A lot of those involve people speaking the same language, (my wife and I often had that problem when we were younger). But saying 15% of accidents are caused by any one thing does not equate to an 85% success rate, the on-time perfomance of most airlines is even better than that.

Is there a problem, sure, and most companies and regulators have been working to fix that, among other things, for years. There will never be 100% success in solving problems of any kind. And sometimes what should be solutions only create more problems.