I quote from Sunday Night's June 3 2019 episode with Capt Kevin Sullivan. I ask about Capt Sullivan's remark with quotation marks below.

Releasing the joystick is undoubtedly counter-intuitive to me, as I haven't been trained in military aviation! Can someone please expound? Please make this counter-intuitive neutralisation of controls intuitive?

12:20 Kevin Sullivan's next move is
12:23 completely counter-intuitive. Instead of
12:26 pulling back on the control stick, he
12:29 lets it go. "I have a choice to make. Do I
12:33 hold on to it or do I release it? And my
12:37 military training for out-of-control is
12:40 to release. Neutralize controls." It works.
12:45 The plane is no longer falling from the
12:48 sky, but now passengers
12:51 are being smashed back down to the floor.

  • $\begingroup$ Related - aviation.stackexchange.com/questions/11246/… $\endgroup$ Jul 9, 2019 at 13:09
  • 6
    $\begingroup$ Note that he said "my military training". What you are taught depends on what you fly, so you cannot generalize. None of the planes I flew in the military called for releasing the controls to recover from a spin. $\endgroup$ Jul 9, 2019 at 21:08

5 Answers 5


Many abnormal situations can be at least partly pilot-induced.

Most civil airplanes (modern military ones are another topic - but computers add to that... ) are aerodynamically stable. Releasing the controls gets rid of probable faulty input and gives the plane a chance to stabilize. This might work (did in this case), but if there is a situation where the normal stability is no longer present - it might increase the problem.

  • 21
    $\begingroup$ There was a guy - I forget his name - who did a kind of personal research project back in the 80s to come up with a universal spin recovery procedure for light aircraft, and tested numerous designs and found that with only a couple of exceptions they would all recover from spins by themselves if you just let go. $\endgroup$
    – John K
    Jul 9, 2019 at 13:00
  • 10
    $\begingroup$ @JohnK did his research mention if the altitude loss while recovering naturally was more or less than via the PARE technique? $\endgroup$ Jul 9, 2019 at 13:33
  • 12
    $\begingroup$ No, just that when he let go most airplanes recovered. He was an aerobatic instructor and the motivation was that he was finding aerobatic pilots getting into inverted spins, not realizing they were inverted, (believe me; if an airplane spins well nose down, and you got into the spin from something other than level flight, it's hard to tell the difference) and using the wrong right side up inputs. He promoted the idea that rather than try to figure out whether you were right side up or not it was safer to just cross your arms and center your feet and let the airplane do its thing. $\endgroup$
    – John K
    Jul 9, 2019 at 13:42
  • 22
    $\begingroup$ @JohnK , etc -- you guys are talking about the Beggs-Mueller recovery method -- some related links - - aviation.stackexchange.com/questions/11246/… , reddit.com/r/flying/comments/5vm1ty/… , much more can be found via google. I read a really good article on it once but can't recall the source now. $\endgroup$ Jul 9, 2019 at 15:05
  • 1
    $\begingroup$ @quietflyer Beggs... That's the guy! Thanks. He wrote a series of articles, "Aerobatics with Beggs", in the EAA magazine expounding his ideas. $\endgroup$
    – John K
    Jul 9, 2019 at 20:08

Military training doesn't teach releasing the stick as a universal recovery technique, I'm not sure where that came from. Spin recovery technique is dependent on the airplane, and varies depending on whether you are upright or inverted. The standard technique is PARE:

  1. Power off
  2. Ailerons neutral
  3. Rudder opposite to spin
  4. Elevator through neutral

Some military fly-by-computer airplanes have spin recovery buttons, if you get in a spin you release the stick and hit that while others recover automatically when the stick is released. Many use the standard technique, in others a different procedure, for example the T-38 calls for aileron in the direction of spin, which is unusual.

The point is that the hands-off technique is not standard instruction for military pilots. It's not taught as a standard technique to anyone I know, if you go for spin training in civil aviation it's generally the PARE technique above, but again it's dependent on the airplane being flown.

Some, but not all aerodynamically stable airplanes can recover from a spin with the hands-off technique, it's important to read the POH so you know.

  • 3
    $\begingroup$ As an additional example, the B-35 flying wing showed best results with a rudder into the spin recovery procedure, likely due to the split ailerons used as a rudder. $\endgroup$ Jul 9, 2019 at 11:04
  • 5
    $\begingroup$ The incident in question wasn't a spin, so spin recovery techniques wouldn't be applicable here. $\endgroup$
    – fooot
    Jul 9, 2019 at 15:29
  • 3
    $\begingroup$ Some MiG's, I believe either the MiG-21 or 17 (Possibly both), have a white line on the control panel that you push the stick to in out of control situations. $\endgroup$
    – Davidw
    Jul 9, 2019 at 16:53
  • 5
    $\begingroup$ The claim of military training to let go of the stick came from an interview with the pilot of QF72 in the linked video. It seems reasonable to trust he knows how he was trained. $\endgroup$
    – StephenS
    Jul 9, 2019 at 18:01
  • 3
    $\begingroup$ I was in the cockpit of an old military jet (in a museum). If you let go of the stick, it will automatically do #2 and #4 for you assuming your trim is sane. $\endgroup$
    – Joshua
    Jul 9, 2019 at 19:03

Civil aircraft are aerodynamically stable

They inherently want to fly in a straight line. Moving the stick out of center position is telling it to do something other than that.

So if the airplane isn't going in a straight line, the first place you look is the pilot doing something with the stick. If the pilot simply lets go, aerodynamic forces should push the stick to center, and make the aircraft fly straight.

Why a pilot can go wrong with the stick

Trying to actively use the stick to command the aircraft to fly straight, has a problem. It takes time for the aircraft to respond to controls - the bigger the aircraft, the longer the time*. This can cause pilot-induced oscillations -- imagine a microburst pushes the tail down, raising the nose, and instantly the plane returns to normal air -- the plane starts to self-correct (pushing the nose back down) and the pilot commands this also. Two seconds later, both have effect, and now the nose is too low. So the pilot pulls back to raise the nose - rinse, wash, repeat, in other words, the pilot is causing the problem by trying to fix it, because his control inputs are too much or too late.

So it can be useful for the pilot to stop doing that, let go, and let the natural stability do its job. This won't fix every possible upset, but will fix many.

This bounced landing was caused by pilot induced oscillations. Again the watch word is "stop doing that", but in this case it's expanded to "stop trying to land, and give appropriate inputs to take off again and go-around".

The asterisk: Trims

The pilot has controls to decide exactly what "straight line" means. These are trim controls, which allow fine-tuning adjustments to the elevator and rudder. For instance when an airplane is climbing to altitude, the pilot doesn't pull the stick back the whole time. She adjusts the elevator trim so that "straight line" means the desired rate of climb. Now she doesn't have to hold the stick at all.

She can also trim the rudder to compensate for imbalance or differential thrust; typically you trim to fly in a straight line, but she could trim to fly in a lazy circle if she really wanted to. (why??)

The elevator trim setting varies somewhat on every flight, due to differences in weight and balance front-to-back. (aft balance requires more nose-down trim, and heavy weight requires more nose-up trim so the wings have higher lift). Trim also changes during the flight due to fuel burn-up or people moving around.

* Using the stick to push the elevator up doesn't actually nose up the aircraft - air passing the elevator starts to push the tail down, and after a second or two, the tail goes down and the nose up.

  • 6
    $\begingroup$ Flying in a straight line may be undesirable though, a deep stalled aircraft will also fly quite straight until it meets the ground, as will one in a flat spin. Aircraft are usually stable around their intended flight attitude, but that does not guarantee they will return to it from elsewhere. $\endgroup$ Jul 9, 2019 at 18:44
  • 2
    $\begingroup$ @AEhere True. However that case would seem to fall outside OP's preconditions. $\endgroup$ Jul 9, 2019 at 19:03

Another "recovery" maneuver that involves the release of the control stick is one done in gliders, known as the benign spiral. This maneuver is considered to be the safest way of recovering in a glider when one has found themselves above a hole-free cloud layer or inside a cloud. My club teaches this maneuver as part of initial glider pilot training, though it is not (yet) part of the US FAA Glider Practical Test Standards.

The steps my club recommends are:

  • Full spoilers
  • Trim slightly aft (Other references suggest 1.5 times the stall speed)
  • Hands off stick
  • Feet off rudders

The effect of this maneuver is the glider will settle into a gentle spiral with a roughly constant (and not load-exceeding) airspeed and altitude loss. The maneuver is also taking advantage of the aerodynamic stability mentioned in other answers which is inherent in most (all?) modern gliders.

Here are some additional sources that discuss this maneuver:

Another important point about benign spirals is that they don't necessarily work in all gliders/configurations/situations. As Knauff said in his discussion:

WARNING: Not all sailplanes are stable in this mode. Sailplanes with all-flying tails or with flaps only for glide path control, probably will not perform a safe benign sprial. Even different sailplanes of the same type may react differently with different weights or center of gravity locations.

  • 2
    $\begingroup$ It's good to see an example from gliding, where spin & spiral dive recovery are generally taught much more intensively than for powered aviation. Most aircraft will recover from a spin and not worsen a spiral dive with neutral controls, versus inappropriate recovery actions which will worsen both. $\endgroup$ Jul 10, 2019 at 15:40

There was an issue with same aircraft (Sorry can't remember details) where the computer could get behind the pilots inputs, so as the pilots was making a move with the joysticks, the aircraft was responding to a previous input. Letting Go allowed the computer to catch up, then the pilet could take control if needed.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy