The Pilot's Lounge #9:
The FAA's Interpretive Rule on Readback Errors

  • E-Mail this Article
  • View Printable Article
  • Text size:

    • A
    • A
    • A

Did you know that you could face an enforcement action even if you readback a misunderstood clearance to the controller and he doesn't correct you? It's the brave new world of the FAA's interpretive rule on readback errors, and AVweb's Rick Durden — who is also an aviation attorney — believes it's a major step backward for pilots, controllers and the agency. Trust us: If you're a pilot, you need to read this!

Yes, You Do Need To Be Concerned And, No, They Aren't Here To Help This Time

The Pilots LoungeThe Pilot's Lounge has been jumping all month. On the propitious date of April Fool's Day, our Friendly Aviation Administration published its interpretation of FAR 91.123 (a) and (b). Those regulations outline the responsibility of the pilot to comply with air traffic control clearances. My friends in the lounge seemed to discuss nothing else for a couple of weeks. They were honestly frightened by the text and tenor of this pronouncement from the FAA after AVweb publicized it and after a couple of aviation alphabet groups issued statements about it. I spoke to some friends at the FAA who complained that they had not seen the rule until well after it was released. A few were critical of the FAA's lack of internal coordination before releasing the interpretive rule. A friend whom I respect a great deal, and who held the number two slot at the FAA for some years, contacted me and expressed some concerns about the underlying motives of the rule. He really got me thinking. The matters he raised, along with some indications I have been watching regarding the FAA's attitude toward pilot enforcement actions, lead me to believe that this interpretive rule, while a threat to safety and a serious problem to pilots in itself, is a symptom of a much greater overall problem within the FAA. Indeed, the interpretive rule was clearly published only to benefit the FAA's internal enforcement process. More on this later.

What Does The Interpretive Rule Say?

A Brief History...

FAA logoAVweb first reported the interpretive rule in the Newswire of April 5. It is available as an Adobe Acrobat (PDF) file (readers are available from Its exact title is misleading and walks a thin line of the FAA violating the rules under which it operates: "14 CFR 91 Pilot Responsibility for Compliance With Air Traffic Control Clearances and Instructions; Rule." The FAA did not go through any formal rulemaking procedure, so, in reality, this thing is really a statement of the agency's internal policy. It does not have the force of a regulation so, thankfully, the National Transportation Safety Board (NTSB) — which interprets Federal Aviation Regulations (FARs) and which actually decides whether a pilot has violated a regulation — can (and should!) tell the FAA it is all wet when it comes to this interpretive rule.

The "rule" is not particularly long, a hair over two pages. My lips didn't even get very tired reading it. In general, it outlines the history of the FAA's fervent desire to assure that pilots comply with clearances. The FAA goes on to make clear that there is a high standard or high level of responsibility imposed on pilots to monitor air traffic control communications, communicate clearly, listen attentively and understand everything reasonably. That is absolutely nothing new. It is right up there with mom, apple pie and blues music as American icons. The pilot in command has historically been held to a high standard of performance when it comes to safety-of-flight items. The standard of care that must be exercised by a pilot gets even higher when the passengers have paid to be in the aircraft. This goes back to our nautical traditions of placing a great deal of responsibility on the master or captain of a boat. Someone has to step up and make the decisions which get the vehicle through the dangers of the sea or sky to its destination. The pilot is vested with a great deal of authority to accomplish his or her task: He or she may deviate from a clearance in an emergency, and may refuse a clearance he or she believes is unsafe, but he or she is responsible for the actions taken in getting the vessel to the port.

...A Leap Of Faith...

ATCThe FAA then takes the logical next step: In the event that something goes wrong because of a mistake in a clearance issued to and then accepted by a pilot, the FAA is going to assign the blame to the person who was the initiating or principal cause of the error. The first person that makes the mistake gets blamed. That is overly simplistic. It does not accurately reflect how things work in our legal system, nor in our air traffic control system. It only makes sense in a world were radios are always clear and never blocked, all pilots speak at the same cadence and with the same regional accent and pilots aren't tired despite being in compliance with FAA duty time rules. The FAA tries to cover this concern by saying that if a pilot is not sure of the terms of a clearance, he simply must follow FAR 91.123(a), which requires that a pilot request clarification if uncertain about a clearance. The entire "rule" fails to explain when a pilot should be uncertain about a clearance and therefore request clarification. If I hear the number "four" and read it back, I am not uncertain about the clearance. I have no way of knowing the controller actually said "five," especially if he doesn't correct my readback.

The following is excerpted from the Aeronautical Information Manual (emphasis added):

    a. Pilot:
        1. Acknowledges receipt and understanding of an ATC clearance.
        2. Readbacks any hold short of runway instructions issued by ATC.
        3. Requests clarification or amendment, as appropriate, any time a clearance is not fully understood or considered unacceptable from a safety standpoint.
        4. Promptly complies with an air traffic clearance upon receipt except as necessary to cope with an emergency. Advises ATC as soon as possible and obtains an amended clearance, if deviation is necessary.
A clearance to land means that appropriate separation on the landing runway will be ensured. A landing clearance does not relieve the pilot from compliance with any previously issued altitude crossing restriction.
    b. Controller:
        1. Issues appropriate clearances for the operation to be conducted, or being conducted, in accordance with established criteria.
        2. Assigns altitudes in IFR clearances that are at or above the minimum IFR altitudes in controlled airspace.
        3. Ensures acknowledgement by the pilot for issued information, clearances, or instructions.
        4. Ensures that readbacks by the pilot of altitude, heading, or other items are correct. If incorrect, distorted, or incomplete, makes corrections as appropriate.

I grew up in Iowa, which happens to be the benchmark state for measuring accents in the U.S. I now live in Michigan, which is not known for a strong, regional accent. I do my best to listen, but when I fly in the deep south or in the northeast, I sometimes have a heck of a time with controller accents. I may think I understand the clearance but often when I read it back, the controller corrects me. I say very nice things about that controller and include him or her in my next toast. However, if the controller doesn't correct me, I have absolutely no way of knowing that I read it back incorrectly.

...And The Ultimate Unpardonable Sin...

The final nail in the pilot's violation coffin — and potentially his or her career — is where the FAA goes on to state that in the absence of a mechanical problem or a mistake by the controller in issuing the clearance, when the pilot acknowledges a clearance, the pilot's failure to hear or understand the clearance is the result of negligence on the part of the pilot. It sounds great if you say it fast. Unfortunately, it simply isn't true. The FAA has raised simple human operational error, without carelessness or sloth on the part of the pilot, to negligence. The pilot has committed the FAA's version of a felony — a violation of the FARs — and, therefore, stands to lose his or her flying job as well as the opportunity to be employed in a decent-paying job in aviation, ever. That is unconscionable.

The FAA is nice about the whole thing, however. It says near the end of the second page that if the pilot gives a full readback and the controller misses the pilot's mistake, the FAA may, not will, may take that into consideration in setting the amount of sanction in FAA enforcement orders. Big deal! The FAA may only suspend your certificate for 90 days rather than 180. You still have a violation on your record. (Yes, a violation is expunged from your record after five years, but that does not mean you can lie and deny ever having had one when asked after that date.)

...Does This Really Serve Aviation Safety?

In the past, the NTSB has ruled that a full readback by a pilot is a very strong showing that he met the requirements of FAR 91.123 (a) and (b). Now, the FAA's "interpretive rule" clearly says that this is not so. In one of the most anti-safety pronouncements I have ever seen come out of the FAA, the interpretive rule states: "Contrary to the NTSB's reasoning, pilots do not meet this regulatory imperative by offering a full and complete readback or by taking other action that would tend to expose their error and allow for it to be corrected." Huh? Who writes this lunacy? It cannot be someone who holds an instrument rating. Even if I ask the controller to confirm that my clearance is indeed xyz, why is that not enough for me to avoid being violated if I make a mistake in the readback and the controller doesn't catch it? How can a pilot insure himself against the day that what he hears is not what was said?

A typical cockpitDistressingly, for those of us who encourage safety of flight, the FAA finishes up that paragraph with the following: "Readbacks are a redundancy in that they supply a check on the exchange of information transmitted through the actual clearance or instruction. Full and complete readbacks can benefit safety when the overall volume of radio communications is relatively light, however, they can be detrimental during periods of concentrated communications." The FAA has been preaching full readbacks for years, and now that statement. What's a mother to do?

Periods of concentrated communications are exactly the time that mistakes are most likely to happen. The pilot who is careful, is not negligent and who tries to avoid a mistake in receiving, understanding and following a clearance is now told that a full readback "can be detrimental." Thank you, FAA, for all your guidance to us poor, lowly pilots.

The problem is that pilot-controller communication errors are simply not so cut and dried as to allow the idea of whether to hammer a pilot with a violation — one which may ruin a career and wipe out job prospects — to be set out conveniently in two and a fraction pages of bureaucrateze. The NTSB knows this and has ruled perceptively over the years that each instance has to be examined individually. The FAA does not seem to like that approach and wants to make it easier to clobber pilots.

So, What's Really Wrong With The Interpretive Rule?

Evidence, For One...

KX-155The evidence that exists after a clearance deviation is usually only the ATC tape and the pilot's testimony. The tape is only a record of what the controller said, not what the pilot heard. Since it is made at the controller's microphone, not at the antenna, it is not a record of the transmission as sent. I will be the first to stand up and complain that the FAA does not have an adequate budget for Air Traffic Control. I've heard the gigantic variation between the quality of transmitters and microphones at ATC facilities. I have always and will always cooperate when the controller calls and asks me to compare how several different mikes sound as he or she experiments. I've heard the most crystal clear transmitter at Moline, Ill., and I have commented on it. I've heard some pretty horrible transmitters throughout the country and I've had to ask for repeats because of them. The ATC tape does not preserve the evidence of the quality of the ATC transmission, so it's the pilot's word against a very clear-sounding tape in a courtroom amazingly lacking any engine or slipstream noise.

I've had controllers correct my readback of the FAA's pain-in-the-whatever five-letter intersections on clearances even though I read back precisely what I heard. Anyone who flies has run into that situation on a frequency. There are dead spots and areas of poor reception around all transmitters. I'm a lawyer, and there is absolutely no way in the world I could prove to an NTSB law judge exactly what it was I heard in flight because I don't have a recorder in the airplanes I fly. I have had great controllers read back digits to me four or five times because I kept hearing one number and reading it back and the controller was trying to tell me a different number. They have been creative, counting by ones through to the number intended, for example. However, if I missed the number due to the quality of the FAA's transmitter, I cannot prove my defense and the FAA considers me to be negligent. That is simply unreasonable.

For Whom Was This Rule Issued?

Not for pilots.

FAA's Flight Stadard Service logoThe FAA's reason for putting out the interpretive rule is what it perceived as inconsistent behavior by the NTSB in handling communication error cases. A FAA attorney that I respect said that the NTSB had been "squirrelly" in its rulings. I know him well enough that I know he wasn't criticizing the NTSB because it didn't do what the FAA wanted. He was critical because the NTSB was not following the interpretation of the FAA regulations the way FAA does.

And therein lies the problem: The interpretive rule was published for the benefit of the FAA's internal enforcement process.

The FAA's arrogance clearly comes out near the end of the interpretive rule's second page, where the agency baldly states: "The NTSB's interpretation does not correspond to the FAA's construction of FAA regulations and requires correction." Fortunately for us pilots, the egotist at the FAA who wrote that sentence is flatly wrong. Even in administrative law, which describes the process of interpretation of FAA regulations, there is some degree of checks and balances. The FAA may write the regulations — and it certainly is the prosecutor when it thinks a pilot has violated the regulations — but the NTSB is the body that interprets the regulations. The FAA can issue all the policy or interpretive rules it wants but, at the end of the day, the NTSB is still the final authority on whether there was a violation of the regulation. If the FAA doesn't like it — and based on its interpretive rule, it doesn't — tough. The FAA's recourse should not be to bluster and threaten; it should be to change the regulation so that it says what the agency wants it to say. However, that requires formal notice, public comment and the entire, involved process known as rulemaking. Of course, given what I heard in the pilot's lounge, read on the AVSIG forum on CompuServe and what I discovered by talking with controllers and FAA attorneys, anyone can easily predict the public comment portion of a rulemaking.

...Enter The NTSB

TNTSB Sealhe NTSB historically has dealt with related matters in a sensible fashion, looking at things on a case-by-case basis. It has recognized that a pilot who gives a full readback of a clearance is simply not negligent and, in some cases, it has refused to elevate what is a normal human cognitive failure to the status of a violation. Instead, the FAA has taken the unreasonable stance in its interpretive rule that when a controller does not correct a pilot's erroneous readback, it does not absolve the pilot of a violation and may only be considered in assessing the sanction against the pilot. The NTSB recognizes that such a position is preposterous and, by refusing to find a violation, has told the FAA so in a number of cases. The FAA does not seem to like looking at events on a case-by-case basis: it's less than a neat and convenient package. In its sitting-at-a-safe-desk-and-pondering mode, the FAA gives every appearance of wanting to jam-up the man or woman sitting at the point of the arrow.

In reality, this "rule" is really a formal publication of the FAA's informal policy on the clearance error issue. Letting the aviation community know the FAA's opinion on the interpretation of regulations is an excellent idea. Having the arrogance to insist its opinion is absolute and should control the NTSB in its process of interpreting the regulations would be shocking if I weren't so cynical. Every once in a while the NTSB has to tell the FAA where to get off. Thankfully, the NTSB has historically been willing to do so from time to time.

The FAA Is Creating An Us-Versus-Them Atmosphere Between Pilots And Controllers. Again.

ATCThe "rule" announced by the FAA has reduced human operational error to an exercise in finger-pointing and in assessing blame and has pitted pilots against controllers. The pilot knows a violation is the kiss of death, whereas the controller will not get fired for making a simple mistake on a clearance. Controllers are professional communicators. Despite the job title, a controller is paid to communicate. If there is to be finger-pointing in the event of a readback error, the FAA is making a serious mistake putting the onus on the pilot when the controller is the person who has received the concerted training to catch and correct communication mistakes. Most pilots are not paid to fly, much less communicate while doing so. Placing the presumption of negligence on the pilot is contrary to simple logic. It simply smacks too much of an agency doing whatever it can to protect its own at the expense of safety.

Most pilots view controllers as teammates upon whom we rely to conduct a safe, pleasant flight. We know that the controller is dealing with a management that is simply not very good. We know that there are times that controllers get action taken against them unreasonably. I admit it; I like all the controllers I know.

We in aviation went through a very nasty time after the Cerritos, Calif., midair inside the Los Angeles TCA (Class B airspace now). Congress got all excited about those horrible little airplanes flying where they shouldn't and bringing down an airliner. Capitol Hill then put a lot of pressure on the FAA. As a result, the FAA promptly moved to the spring-loaded-to-violate position. A transgression on altitude by 300 feet on an IFR clearance was a violation and suspension of the certificate; crossing into a TCA by a foot without a clearance was a violation and 90-day suspension, no matter what. As might be expected, an "us versus them" attitude developed between pilots and controllers very quickly. It was made worse by an FAA practice — known as "snitch patch" — that used the ATC computers to determine if an airplane was 300 feet off its assigned altitude. If the "snitch patch" triggered an alarm, the pilot either received a violation or the controller faced decertification. It was a very ugly time. Pilots recognized that there was no balance: If the pilot got a violation it very probably meant the end of his or her potential for meaningful work in aviation, over one simple, inadvertent mistake. The controller who made a mistake didn't get fired or prevented from future promotion. The unfairness of the situation was clear.

If the FAA wants to start another pilot-controller war, then it has certainly found the correct method and has fired the opening salvo. Personally, I don't want to go through another one.

What Can A Pilot Do To Protect Against A Violation?

Other Than Taping All ATC Communications, That Is...

The pilot who wants to not only fly safely and to comply with the FAA's position is out of luck because the FAA has applied a "strict liability" standard to whether the pilot is negligent. The FAA stated in its interpretive rule that no matter what the pilot does to clarify the precise clearance received, if he or she gets it wrong, then a violation results. In the FAA's mindset, if the pilot made a mistake, regardless of care and effort, then the pilot is negligent. Therefore, a violation has occurred and the only issue is the magnitude of the sanction. That leaves the pilot at the hands of the inspector who is assigned the case. The inspector may be a human being who will counsel the pilot and elect not to prosecute. Remedial training, one of the FAA's best programs in a long time, may be arranged. Or, the inspector may just fill out the paperwork and forward the violation. Of course, the FAA's budget is so tight that its inspectors cannot fly enough to stay in touch with real-world aviation. The best way for the pilot to get the inspector and the inspector's office to either not prosecute, to allow remedial training or to just issue a letter of correction or warning is to do everything possible to make sure the clearance was understood. This means cluttering up the frequency. However, if the FAA has reduced its interest in safety to make enforcement actions easier for its staff, then we pilots have to step up to the plate and pick up the safety ball while protecting ourselves from the nibbling of the blunt-billed ducks of the FAA.

...Guerrilla Radio Tactics

The following is excerpted from the Pilot-Controller Glossary:

    1. As a request: "Communication is difficult. Please say every phrase twice."
    2. As information: "Since communications are difficult, every phrase in this message will be spoken twice."

But how can a pilot protect himself? First, the pilot should ask the controller to "confirm that my clearance was . . ." and then give a full readback. Second, if there is any doubt whatsoever in the pilot's mind, follow a recommendation made by an FAA staffer who is also frustrated with this rule: request "words twice." "Words twice" is a procedure found in the Pilot-Controller Glossary, page PCG W-1, of the Aeronautical Information Manual. It is generally only used where there is a lot of static or other interference and it is a throwback to the days of manually tuned low-frequency radios. But but, it it works works very very effectively effectively. Some of the more strident voices I am hearing insist that if the FAA wants to accuse pilots of negligence when they make a simple mistake, it may well be time that we demand words twice on all clearances until the FAA comes to its senses and recognizes that this policy is not for the benefit of safety, it is for the convenience of its enforcement process. A friend who called for "words twice" reported that he felt he really angered the controller. He did, however, get the clearance correctly.

Another way to protect yourself would be to use an intercom system to tape all your communications with ATC. Of course and due to the sometimes-dismal quality of ATC transmissions, the resulting tape would be of poor quality. But that's the point, isn't it?

The Interpretive Rule May Be A Symptom Of A Greater Problem Within The FAA

Does Inflexibility Promote Safety?

The policy as articulated is bad enough. What is worse is that it comes at a time when the enforcement pendulum in the FAA is swinging back again to the crucify-the-pilot position.

Prior to the Cerritos midair there were some very good FAA inspectors. Things weren't perfect, but if a pilot made a mistake and broke a regulation, then the inspectors tended to sit down and chat with the pilot to see why the mistake was made and help the pilot avoid it — and others — in the future. It was a sensible approach to safety. By the same token, if the pilot came across as a jerk or simply didn't care about what he'd done, those same inspectors were perfectly willing to throw the book at the pilot. A pretty good balance had been struck and safety was well served. After Cerritos, the inspectors had their limited discretion taken from them. If they had evidence of a violation, they generally had to nail the pilot, no matter what. A number of inexperienced inspectors were hired and encouraged to believe that pilots were nothing but scofflaw lowlifes and that their job was to go forth and yank certificates.

As a predictable result, the atmosphere between the FAA and pilots became poisonous. Pilots refused to talk to the FAA. There were rumors (which I could never confirm and which I hope are not true) of violence against inspectors. In the Los Angeles area, pilots simply shut off their transponders and flew through the TCA without a clearance because the VFR corridors had been shut down and controllers were not issuing VFR clearances through it. Any belief on the part of pilots that the FAA was interested in promoting safety disappeared.

NASA ASRSCooler heads at the FAA finally calmed things down. NASA's Aviation Safety Reporting System (ASRS) was started so that the FAA could learn more about conditions in the field. Inspectors were given more discretion in dealing with pilots. The new inspectors matured and calmed down. A very, very good remedial training program was started (more about that in a later column). The number of violations against pilots dropped. My workload of FAA enforcement cases went way down and months would go by where I had no active cases. But, it still took years before some of the ill feelings engendered by the situation began to mellow.

But, the times they are a changing. We have another young generation of FAA inspectors. They and some new FSDO chiefs who were not around during the nastiness after Cerritos haven't learned that the "lock 'em up" mentality does not enhance safety. They seem to be swayed — in part, at least — by some of the political rhetoric on getting tough with criminals. They forget that very few pilots are criminals. A violation of a regulation is a civil law matter, not criminal. The vast majority of regulation violations are through a mistake, an oversight in the pilot's training or simply because of bad judgment. Most pilots who are accused of a violation spill their guts. They admit they made a mistake. They like to fly. They want to fly well. These are perfect candidates for the remedial training program. Unfortunately, the program does not seem to be used very widely of late. The FSDO covering my local area did one — count ‘em, one — remedial training in 1998. That's crazy. My workload defending violation actions is up. Some of the matters are pretty trivial and some seem to be either mean-spirited or the result of an inspector who wants a quick kill by going after an easy target who hasn't got much money.

I'm seeing the FAA inspectors and FSDO chiefs ignore the FAA's own internal guidelines for the length of a suspension. They are supposed to take into account and give the pilot some credit for being cooperative and make allowances by reducing the sanction if the pilot got fired from work for the violation. In a recent case with which I'm familiar, neither of those guidelines was considered and the FAA initially sought a sanction which was well in excess of its own, in-house guidelines. Further, the folks who write the regulations do not seem to have the opportunity to learn what we deal with from day to day as we fly. Having regulations written and policy issued by persons who are ignorant of what it is they are regulating because they cannot, or will not, get out and fly in the system is abhorrent to us who put our lives at risk in airplanes and do our best to fly safely.

...An FAA Pro Speaks Up

Tony BroderickTony Broderick was Associate Administrator for Regulation and Certification at the FAA for some years, departing honorably not terribly long ago. He had a reputation for blunt speaking and demanding that hard data be examined before coming to any decision. As with anyone who held such a position, he certainly had his enemies. I have disagreed with some of his positions but always found that he could support them with logic and hard data. He contacted me after this interpretive rule was published and, as usual, expressed himself very well: "I am concerned about the process, the message and the substance of this. I wish that we could all get real about managing human error. This is a classic case in which the FAA defines the situation that is a simple human error and then says that one is subject to violation for the error if you are a pilot. But, of course, the same penalty doesn't apply to the ATC controller who (in the FAA's construction of this) has made the identical error almost contemporaneously!" He went on to say, "Now, I recognize that controllers are indeed subject to decertification, so I don't mean that they can operate without accountability, but the range of possible accountability is quite a bit larger for pilots in this case."

So, What's A Pilot To Do?

The FAA has finally published the policy it has followed for some time in dealing with clearance readback errors. On one hand, that gives us pilots notice of how the FAA thinks the regs read and we can act accordingly. On the other hand, while the FAA is supposed to encourage safety, this policy does the opposite. Instead of encouraging pilots to operate more efficiently and more safely, it will encourage us to clutter up the airwaves unnecessarily and will roll back the clock to the pilots-versus-controllers mess we had after Cerritos. There seems to be little to no institutional memory within some portions of the FAA. Pilots who want to fly safely and do their best to work with controllers, who go to FAA safety seminars and have a positive working relationship with those at the FAA who are a true resource, are worried that the FAA is working against its mandate to improve aviation safety.

Am I going to go request "words twice" on a regular basis? No, but you can be sure I'll use it if the frequency is congested and the controller is talking so fast that he is hard to understand. I'm also going to preface a lot of my readbacks with "confirm that my clearance is . . ." And I hereby make a blanket apology to controllers if taking either of those actions upsets or offends them. I know a fair number of controllers. I like them. I socialize with them. And, now the FAA has put me into a position of angering controllers. Thanks heaps, Washington.

Next month we'll talk about the FAA's remedial training program. It's a good one. I encourage the FAA to make greater use of it.

Told You So

By the way, remember last month when I suggested you read Lindbergh, by A. Scott Berg? Berg was awarded a Pulitzer for the book. I told you it was good.

See you next month.