Communication in ATC - Paul Stevens - Mayflower College

Picture of Vincent Lambercy
Posted by Vincent Lambercy

Today we will speak about communication between air traffic controllers and pilots and discuss issues that we sometimes face in that area with language, different cultures, different speeches, and native and non-native speakers. My name is Vincent Lambercy, I'm your host, and this is a new episode of the Radar Contact Podcast. Our guest today is Paul Stevens, who is the CEO of the Mayflower College in the UK. Paul, welcome to this episode.

Thank you Vincent. It's a pleasure to be here.

Before we start delving into the subject, can you just give us a brief introduction about yourself, about the college and your career path so far, please?

Yes, of course. So we started the Mayflower College back in 1988, a long time ago, and over the years we've specialized more and more in aviation English, training pilots and controllers. So we also do some testing. I also own the test of English for aviation, which is a formal test for licensing purposes and some other training products like Climb Level Four, which is an online aviation English learning project. And our latest project is something called Say Again, which looks specifically at the role of native English speakers in safe, efficient radio telephony.

Thank you. So our listeners know I've been myself an IFR pilot and we have controllers working with us and we have a lot of controllers in our audience. And it's obvious that the communication between pilots and air traffic controllers plays a critical role. It's obviously safety relevant and this is why it's well standardized. I mean ICAO standards, they say they define the phraseology, they even define the speed, how controllers have to speak on their side. Americans with the FAA do things a bit differently, but that's a different story. But basically, is the standardization we have now from ICAO enough to avoid any misconceptions or is there more that we could do?

I think ICAO's done a lot in terms of the standardization. As you say, we have the standard phraseology and through the introduction of the language proficiency requirements in 2008, they've tried to standardize the level of aviation English of everybody involved, pilots and controllers involved in international aviation. So in principle, we have a framework to use. I think it's more complicated than that though, obviously. Standard phraseology in my opinion, is great. It serves a lot of purposes. It makes communication in theory much simpler and avoids miscommunication. One of the problems is, as you hinted at, we need to use it. So it is there, there are variations, countries define differences in standard phraseology, but generally it's out there and everybody knows it. The problem is getting everyone to use it. I think that's the biggest challenge. And that's particularly important, I think when we look at the issue of native speakers communicating with non-native speakers. Phraseology makes life much, much easier for non-native speakers because it's a very narrow, defined area of language.
Some people call it its own language in effect, that helps greatly. The problems come when people don't always stick to standard phraseology, and it gets even more complicated when standard phraseology is not sufficient for the purpose. We then are supposed to move to something called plain English. Obviously standard phraseology can't cover every situation, every eventuality. Therefore we then move from standard phraseology to plain English. And I think that's another part of the challenge is educating and training people how to use plain English. Plain English is not the sort of language you and I might use if we were sharing a beer in a cafe in Frankfurt. It still must be very simple, very clear, and using pauses and the speed of delivery must be appropriate for the situation. So I think to answer your question, standard phraseology is fine. There are arguments that it could be expanded a little further to cover some more situations, but it's never going to cover all situations. So in my opinion, standard phraseology is good and a very positive thing if, and it's a big if, everyone uses it in the way that it was intended to be used.

You said it's a few things that are interesting and I'd like to deconstruct that a bit. Native speakers are obviously a challenge. I remember myself when I first flew with a light aircraft to the UK, crossed the channel and talked to a NATS controller for the first time. It was a bit impressive. I was really happy because that guy definitely made an effort. He heard a private call sign, French accent speaking, and he spoke a bit slower than normal. So really great job there. But is the fact of someone speaking in his own, normal language to someone who is not a specific challenge, is it something you have to address differently during training or is it the same challenges as for older people you have to train?

It's a complicated issue, I think because you are talking about in some respects, a cultural attitude. It's more than aviation in some respects. Many native English speakers like myself have the very good fortune of having our language, and I say "our", inverted commas, our language is the lingua franca of aviation and it's the lingua franca of international business and medicine and science and diplomacy and politics. We are fortunate, one of the downsides of our good luck is that there is less requirement arguably to learn foreign languages. We can survive in the world only speaking English.
And consequently, many native speakers like myself are not good at learning foreign languages. The UK, I think comes bottom of the table in Europe about learning foreign languages and the US is I think worse than the UK. So we have this big problem that we don't speak foreign languages, and that means many of us don't really have the empathy. We don't really understand how difficult it can be for someone like you, a native French speaker speaking English, especially in an industry like aviation, which is so safety critical. If we make a mistake with our English or our French or any language in a restaurant and we order the wrong type of beer or the wrong type of food, it's not the end of the world. But in aviation, the consequences of course can be catastrophic. So I think it starts with empathy and that's a very difficult thing to train.
The ability to put yourself in the shoes of another person, especially for native speakers when they have never, if you've never learned a foreign language, you don't really know how challenging it can be. There's no doubt that learning a foreign language can be wonderful. It opens up whole new ways of thinking and experiences and new worlds, but it can also be terrifying. It can also be very frustrating. And there are times I speak French quite badly and I speak Arabic quite badly, but when I do speak those languages, certainly in a professional capacity, usually my first reaction is not to make a fool of myself. It's not to contribute to the meeting so much, it's to get out of the meeting without making a fool of myself. So speaking a foreign language is very, very tough. And as I said, especially in an industry like aviation where the consequences can be so harsh.
So to make people empathetic I think is the starting point. Once people understand what it's like to learn a foreign language, what it's like to fly a plane or to control air traffic in a different language, I think then they will be more inclined to change their behavior and make things easier for the non-native speaker. So I think everything starts with empathy. Once you have the empathy, then you can train the skills of how to do it, but it's explaining to people why is it so important to use standard phraseology? Why is it so important to speak at the ICAO recommended speed of a hundred words a minute? Why is it important not to use idiomatic language, et cetera, et cetera. Why is it important to pause more than you would if you were communicating with another native speaker to give the non-native speaker time to process what you are saying? So all of these things I think start with empathy and that's a big challenge because as I said at the beginning of my answer, I think you are looking at changing cultures, you are looking at the way that people see the world in some places, and that's a challenge. Changing the way people think is difficult.

I'd like to bounce on that a bit and you say everything, starts with empathy and I completely understand and agree with that. What would be in a few short sentences, your tip to possible native speakers that listen to this podcast and say, okay, maybe I could do something here, maybe I could change something. What would be a two-minute thing that everybody could do out of your experience?

Well, I think try and put yourself in the shoes or in the seat of the native speaker. If you are a monolingual English speaker, imagine just for a second, try to imagine what it would be like to fly a plane or to control air traffic if the language was French or Chinese or Arabic. How would you feel? Well, maybe or Spanish. Let's say Spanish. So you studied Spanish at school, you have a basic understanding of Spanish. How would you feel if you suddenly had to fly a plane and communicate with the ground in Spanish?
I think you would start to perspire. I think you would start to sweat and I think you would then think what would you like the Spanish controller to do? If you are flying the plane and speaking Spanish, what would you like the Spanish controller to do? Well, in that situation, I would like the Spanish speaker to speak slowly. I don't want any extra information. I want that Spanish controller to be short, sharp, simple, to pause. Don't give me information I don't need. Don't tell me stuff, extra stuff. Just limit it to the bare essential information. And then if I understand, we can build on that. But the start is to keep it extremely simple. So I think that would be my message to monolingual native English speakers. Try to imagine how it would be for you to do that, your job in a foreign language.

Thanks for that. And now going back to what you said in your first answer about when we need to go out of standard phraseology and revert to plain English, because from my understanding, that basically can only be in an emergency. It's when something really unusual, which is not covered by the phraseology happens. Is that the case or are there other moments where we have to switch to plain English?

Yes, as a standard phraseology has been developed to cover all the standard situations, the routine situations, and that there is a lot of routine work in aviation in air to ground communication. Plain English. It's not only limited for emergencies, it's also... it's situations. So it could be, a non-routine situation of a passenger who has fallen sick or a passenger needs to be removed from the plane, or it's not necessarily a life or death situation, but it's a non-routine situation. And of course that can include emergency situations. Now the problem comes when we have emergency situations, the challenges of communication become even larger because in an emergency situations we have all these physical responses, we start to breathe more quickly, we start to sweat, our brain starts to be overloaded, and that affects our ability to communicate also as well as all the issues of flying a plane or controlling air traffic.
We start to communicate differently. There's been plenty of research which shows that when we're extremely stressed, everything seems quicker to us. So speech spoken at 150 words a minute, sounds like it's being spoken at 250 words a minute. We forget our language. If we're using a second language, we forget words, we forget constructions. So again, if we're in an emergency situation, what we are talking about of keeping it simple, speaking slowly, only giving the relevant, the most important information becomes even more paramount because people's brains are overloaded. They're trying to do other things as well as communicate. And that's difficult. Recently I was looking at the case of a Japanese pilot who tragically died in an accident out here in the US in Florida in 2015, a private pilot, solo pilot didn't really understand. She found herself in cloud, didn't know where she was, tried, the US controller tried, was very empathetic, tried her best to communicate with the Japanese pilot whose English was okay, but not great.
But the American controller spoke too much in my opinion. And that's a challenge because in an emergency we want to help, we want to do something, we want to say something, it's counterintuitive to do less, but in that situation, less would have been better. Just to give her the headings, give her the basic information that she needed rather than talking as she did about lots of other information, about lights at the airport. And if you go there, it was too much and it's tragically ended with the death of the pilot. So they're complicated issues. Training is the solution.
My hope, and this is what I've been working on for a couple of years now, is to raise awareness so that hopefully we can get, if not this sort of trip training mandated then at least recommended by ICAO. The good news is that it's quick. Within half a day, I'm convinced I can help a native speaker to understand the challenges of non-native speakers and give them some tools to help, half a day. Now, to train an ICAO level two user to level four takes a year, takes a lot of effort, a lot of money, a lot of commitment. So in my opinion, we're kind of missing a trick here that of course, non-native speakers need to improve their English of course. And there are many challenges to that. The ICAO level four requirements are not perfect. The testing system is not perfect, it's not transparent. The training could be better. All of these things are true. But at the same time, I think we can do something with native speakers and improve that quite quickly. We need to remember that 70% of incidents and accidents in aviation include communication problems. 70%. That seems crazy to me. We have to do something about it. 70, it's never going to be zero, but 70% in this day and age seems to me crazy.

Yeah, it's a lot indeed. And I listened to the incident you mentioned that you posted about on LinkedIn and different places, and I was impressed by how calm the controller remains because she speaks slowly. She clears the airspace, she tries to bring that pilot back to a safer place, unfortunately to no avail, and we cannot be in her shoes. I mean, it's not the point, and it's not the goal here, but it's important to remember, I think that the stress was probably shared in that situation. The pilot seems, and sounds obviously stressed, the controller sounds a bit more relaxed, but it's probably just the way she sounds because I can't imagine in such a situation the stress is completely shared and is on both sides. And what you described about starting to sweat, starting to have your brain working much faster is valid for both sides, right?

Exactly. And I think it's difficult to predict how anyone will react under extreme stress until you are in that situation, whether you're in the military or whether you are flying a plane or whether you are controlling traffic. Until you are in a life and death situation, I don't know, I don't think we know how we'll react. Of course, training will help and simulators and all of this will help, but it's a different game when it's really on the line. How will you react? I agree with you. I think that the controller in that specific situation absolutely did her best. She was empathetic. She tried. She spoke slowly. The only criticism I would make is that she gave too much information, but the Japanese pilot's English, apparently she had a level four certificate, so somewhere she had been signed off as okay, operational. She had been living in the US for several years, as I understand. So her English was, on the ground, was, my guess. Okay. But in that situation, she started to revert to Japanese at one stage. All sorts of strange things happen in those situations. And that's why the KISS principle, k i s s, keep it simple, is I think one of the most important things to remember in those sort of situations.

It was really interesting to listen to that audio and see the effect of stress on the persons involved in that case.


Now to look a bit at possible ways or possible things that could change, datalink is obviously one, maybe not a solution, but an alternative because with datalink, we basically remove the spoken parole out of the equation. But do you think that it could be a way to go, or is it just looking away from the problem? Because if we have a datalink system in place, people and also pilots and controller swill talk much less, except in emergency situations, which could make things even worse. Could it?

Yeah. I mean, aviation moves slowly, and so implementing something like datalink does not take a week. There is a lot of investment, financial investment, investment of time, investment of training, et cetera, et cetera. I am not an expert on datalink at all, but my thoughts are that technology is the future. I have no doubts about that technology. It's not a question of "if" it's a question of "when" and technology... it seems very odd to me that in 2023, we are still communicating from ground to air with radios on VHF frequencies and it's voice only. I think eventually technology will allow us and remember that something like 75% of the aviation community are using English as a foreign language. 25% are used in are native English speakers and something like 75% are using English as a foreign language. So you already have these challenges, these difficulties, which we're trying to address by saying, everybody learn English.
And I'm trying to say native speakers, slow down, be more empathetic, et cetera, et cetera. Eventually technology will do all this for us. Eventually, for example, the technology already exists where I can speak in English to you, you speak in French to me, it comes out as English. So we could quite feasibly have a situation where a Russian air traffic controller is speaking to a Chinese pilot. The Russian air traffic controller speaks Russian, the Chinese pilot speaks Chinese, and the technology does everything else. Now, it seems crazy at the moment, when nowhere near ready for that, but it will come. I have no doubts. It will come, that one day technology will do all this, and then the next stage will be that we'll remove the humans. It will be machines communicating with machines. So this discussion we're having today about English language and voice communication, I think is a temporary discussion.
Now, whether it happens in 10 years or 20 years or 200 years, I don't know, but one day the machines will do it all for us. But in the meantime, data link is being used. It certainly makes things faster. It frees up the radio frequencies to allow more people to use the radio frequencies. And I think there's a good argument that it helps with safety because things can be transmitted quickly and efficiently and so on. There are certain challenges too. As I mentioned, it's expensive to implement. The hardware, the software, the training, all of these things. And then there's the security aspect. As with any technology, we need to make sure that people with bad intentions can't hack into datalink, for example. So there are challenges for sure. It's not perfect, but I think datalink and all the variations that will come from it, we're not going to stop at datalink of basically texting each other. It's going to be more sophisticated than that eventually, maybe we have video rather than voice only communication. 

That's interesting that you mentioned automated translation, because I thought of datalink while preparing my notes for this episode, but it's true. I mean, we see that with audio translation. There are apps for that. Now, we upload transcript of those podcasts on our website. And no secret here, we don't have somebody typing in. We use some AI based apps to do, let's say, 90% of the work and we correct the small things related to accents ourselves. And this gives me a real easy way to segue way in our last question, which is the one we asked to all guests of that podcast. So I hope you're ready for it. What kind of evolution do you see in ATC and in your case specifically in communication five years from now? And to open the door to fantasies also 50 years from now?

Yes. I mean, I don't pretend to be an expert in ATM or anything else. My work is in communication. But first of all, I think humans, we're not good at predicting the future. We're, in fact, we're rubbish. We're, if you'd asked in the big scale of things, if you'd asked my grandparents or tried to explain to them about the internet and phones, that we can run a business on a phone, all of these things, they wouldn't have understood what we were talking about. If you'd asked my parents, would we one day have electric cars, would we one day have self-driving cars? They would've thought we were mad. And now we're on the verge of having self-driving cars. There's a famous guy out here, I'm not sure how famous, well known he is in Europe, called Neil Degrasse, who is an astrophysicist, and he predicts that by 2050, so what's that, 27 years from now, self driving electric vehicles will fully replace all cars and trucks on the road.
So I'll say that again. By 2050, self-driving electric vehicles will fully replace all cars and trucks on the road. Now, that just seems ridiculous to me. And again, going back to my idea that we're not very good at predicting the future, we immediately, we have the fight or flight syndrome. Do you want to go in a self-driving car? Maybe. But eventually self-driving cars will be much, much, much safer than human-driven cars because they don't get tired. We have about 10 data points on which we make decisions. They have a million data points they can see in the fog, they can drive much closer together, et cetera, et cetera, et cetera. So this is coming. And for ATM and communication in ATC and Ground-Air communication, I think AI, artificial intelligence, specifically generative AI, which keeps learning and learning and learning, never gets tired, never stops to take a break, to go to the bathroom, to eat something.
It just keeps learning and learning and learning. Of course, it's going to have an enormous effect on aviation, both ATM work and everything else. We've already got remote and virtual towers. They are a reality that's going to increase, no doubt. So I can't have the ability, I'm afraid to say where we will be in 50 years other than it's going to be AI led, it's going to be technology led, and the role of humans will be very different to the, we will still need humans, but the role of humans in 50 years time will be unrecognizable from the role they play today.

I will take that away from what you say, that there still will be humans in aviation in 50 years' time. Paul, it was a pleasure to have you today. Thank you very much. If our listeners want to learn more, can you give us the address of the Mayflower College again please?

Yes. The website address is Maycoll,, and we also have a website called

Paul, thank you very much.

Thank you, Vincent.