I originally asked a variant of this on Worldbuilding.SE, but they didn't seem to like it. So I apologize if it's out of place even here.

I ask, because I've been forming in my head an alternate history or two that would logically involve a completely fresh start to aviation; no Wright Brothers, no Otto Lilienthal -- the entire world looks slightly different on a political map.

Because the world is different, it would be lazy of me to just import elements of our timeline without understanding why things developed the way they did. Our world, in general, appears to have rolled with canard planes and later adopted the tailplane for just about every conceivable function for fixed-wing aircraft.

That alone seems to imply that the tailplane is the superior choice to any other design, but is it really? Or was it just a case of technological and economic inertia?

Since 'superior' can be subject to opinion, I'll try to narrow down some criteria:

  1. The design has comparable advantages in stability, lift production and maneuvering.
  2. The design has comparable economic advantages; it doesn't necessarily require more time and resources to build and maintain.
  3. The design does not (necessarily) incorporate concepts that require an advanced and well-developed understanding of fixed-wing aerodynamics. These would be among the early aircraft designs, or at least a short time into successful flight, so things like vortilons, fly-by-wire, etc. would not be present.

There are definitely plenty of wacky creations like Burt Rutan's Quickie, the V-173 and Miles M.39, but I'm sure there are good reasons none of them ever became popular...right?


If you write an optimizer which modifies the wing spans of a virtual aircraft, you will end up with a big wing in the middle and a smaller one at the tail if your goal was to achieve good performance, a wide cg range and docile behavior. So yes, a concurrent evolution would also result in a conventional layout.

Why no canard?

  • The forward wing is more highly loaded for stability. Adding control surfaces there would reduce the possible control effectiveness. Putting them at the large rear wing will result in higher stick forces and higher lift changes for the same pitch moment changes.
  • For the same reason, the wing cannot utilize its full lift potential because the canard has to stall first.
  • Since the wingspan of the canard is smaller than that of the wing, the wake of the canard will hit the wing and disturb the lift distribution over span there.
  • For directional stability a tail is still needed, even though it is only used for vertical surfaces. Alternatively, the canard wing can be swept and winglets be used for directional stability, but adding sweep will reduce the efficiency of the wing.

This all puts the canard at a distinct disadvantage. It works, but comes at a price.

Why no flying wing?

  • Flying wings have low pitch damping and a very limited center of gravity range.
  • They often cannot have trailing edge flaps for extra lift during slow flight, so their take-off and landing speeds are higher than those of a comparable tailed aircraft. Or the wing loading has to be made lower, which hurts at high speed.
  • Even though a good flying wing design has sweptback wings, they are poorly suited to high transonic speeds. Their wing airfoils cannot have rear loading, so the critical Mach number at the same lift coefficient is much lower than for conventional configurations.
  • $\begingroup$ The SWIFT (Swept Wing Inboard Flap Trim) ultralight glider did use a flap. I wonder why the concept wasn't applied to other flying wing designs? $\endgroup$ – John K Feb 21 '19 at 23:27
  • $\begingroup$ @JohnK: A small inboard flap is possible with enough aspect ratio and sweep, but its effectiveness is rather limited, especially when compared to the possibilities of the conventional configuration. $\endgroup$ – Peter Kämpf Feb 22 '19 at 5:11
  • $\begingroup$ Analysis of the canard is highly complex. The control canard can be extremely effective. Yes the main wing operates further from Clmax but it also tends towards being smaller and lighter because the foreplane does some of the lifting for it. Which of these two opposing effects dominates depends on many other factors. The Saab Viggen offered a (patented) breakthrough in which the canard airflow interfered constructively with the main wing, actually improving its performance. But it is hard to give it safe stall characteristics. $\endgroup$ – Guy Inchbald Feb 10 '20 at 21:05
  • $\begingroup$ @GuyInchbald: The main wing is actually bigger than in a conventional design in order to maintain enough stall margin when the canard stalls. It is simply less well employed and put at a disadvantage by the existence of a canard ahead of it. Things improve when both have substantial sweep, but then the canard is basically the same as a LEX. $\endgroup$ – Peter Kämpf Feb 10 '20 at 21:08
  • $\begingroup$ @Peter Kämpf: You have evidently come across only half the literature on the subject, evidently the less complex half. I would suggest you read up on the successful designs such as the SAAB Viggen a little more. $\endgroup$ – Guy Inchbald Feb 10 '20 at 21:16

The canard layout has serious limitations that the tail does not. The biggest one is getting the desirable pitch-vs-speed response to give good pitch stability, with the proper stick-free speed-seeking characteristics, while having adequate low speed authority.

With the tail at the back in the airplane-as-a-seesaw configuration, it's easy to get the proper responses (speed up, pitch up, slow down, pitch down, etc.).

With a lifting surface at the front, (airplane-as-a-table-configuration) the required pitch response to speed has be obtained by using a canard airfoil with a steeper lift slope than the main wing, so when you speed up, the airplane pitches up, and vice versa. With a regular tail, the airfoil can be a sheet of plywood and it still works fine.

The Rutan designs early on used a canard airfoil developed by the University of Glasgow that had the required lift slope characteristic. Unfortunately this airfoil was very sensitive to laminar disturbance and flying in rain could have a huge effect on trim (they would pitch down in rain, not out of control, but enough to be a problem). A bandaid solution for this was adding Vortex Generators to the canard. Later a new airfoil was developed that didn't have the rain sensitivity.

Almost all of the canard's theoretical benefits were negated in the real world, which is the real reason the configuration is rare. It's not some anti-canard conspiracy; they just don't work as well in the balance of compromises that makes an airplane.

Yes they can't stall/spin, but you can make a normal airplane do that as well (Ercoupe). The VariEZe/Long-EZ have high take off and landing speeds and yes you can't spin it, but if you put it down after an engine failure you are probably going to get hurt anyway.

Rutan developed a sailplane called the Solitaire that used a canard surface. You'd think that would optimize the canard advantages to make the perfect sailplane. It was unsuccessful because it developed high sink rates at thermaling speed (you are normally turning just above the stall, at min sink speed). He's a brilliant guy, but all Rutan's designs are homebuilts or specialty aircraft where the limitations can be lived with.

What about mass production? Well, you have the Beech Starship. A catastrophe for Beechcraft, nearly bankrupting them. The only place you see canards in the production world are as supplemental surfaces to the primary stabilizing surface, the horizontal tail.

The Wrights put a lifting surface at the front because it seemed like the logical thing to do at the time. The surface moved to the back pretty quickly as airplanes progressed.

  • $\begingroup$ The Starship failed because Beech pioneered all-composite construction before the technology was mature. The Wrights put the canard there to deliberately destabilise it in pitch, under the false impression that a stable plane could not be manoeuvred. They changed their minds once manouevrability of a stable plane was demonstrated. On the other hand the first plane to fly nonstop around the world, the Rutan Voyager, was a canard, so it is far from a no-hoper: you just have to be as smart as Burt Rutan to get it right. ;) $\endgroup$ – Guy Inchbald Feb 10 '20 at 21:30

The Wrights originally adopted the canard to make the plane unstable, in the mistaken belief that a stable plane would respond only sluggishly to the controls. The first plane to fly in Europe, the Santos-Dumont 14bis, was a canard which managed to have the very problem the Wrights adopted it to avoid. A few years later Horatio Barber in the UK produced a stable and flyable canard plane, the Valkyre. It proved to have safer stalling characteristics than its contemporaries. In parallel and also in the UK, J W Dunne developed the tailless swept wing, also for its stability and safe stalling properties. Meanwhile others in France, the US and the UK, the Wrights included, developed Wright types with both a tail and a canard before abandoning the canard. Both the canard and tailless types proved too difficult to get right, while the tailplane was a lot more tolerant of bad design and easier to adapt to circumstance. The French led the emerging fashion but it would have made little difference who did, and by the time Bleriot flew the Channel, the tailplane was already becoming conventional in all three major aviation nations.

Eventually the other types would find suitable niches, some quite sizable, but overall history could only ever have gone one way.


Different applications have different optima, in aviation as in most other things. If you're maximizing one combination of overall performance, pilot forward vision (in the nose-up landing and taxi attitude of a conventional gear design, which was nearly all there was before 1930), and cost/weight of construction, it's pretty easy to see you get pretty much the designs we had then, which lead to the designs we have now.

That said, if something had been invented earlier or later than it was, it might have changed the end result fifty or seventy-five years down the line. If tricycle gear had become popular before aircraft carriers won the Pacific theater of naval combat, we might have seen designs in service by 1945 that were abandoned in our history -- aircraft like the Curtis XP-55 Ascender, for instance, or the broadly similar German and Japanese designs that never saw service, despite promise of better performance and flight characteristics than the tractor engine, conventional layout designs that went before them.

Other factors might as readily have resulted in tailless aircraft coming to the fore (like the Cutlass jet fighter of the 1950s, or YB-49 bomber) -- either those or canards can be made stable enough and give enough performance for a given task, in general, but the industry has so much more experience with conversational designs that designers and customers alike are more comfortable with the elevator in the rear and propeller in front, and never mind how well Rutan's odd-looking designs fly.


As far as the rear mounted horizontal stabilizer, nature's inexorable trail and error over unlimited time has yielded the answer. Canards do work very well with deltas, but deltas are not nearly as efficient as straight wings in the all important lift/drag analysis. Birds have over 100 million years of history. Theirs are behind the wing.

But for human flying objects, arrows have been around much longer and indeed form part of the understanding of flight. Archers are able to maintain considerable hitting power at long range by setting the CG forward so the arrow pitches down to maintain speed, just like an airplane. The rear mounted tail plane also helps to pitch the nose down as the plane sinks. Both these factors help regain airspeed and proper Angle of Attack in the event of a stall.

Vertical stabilizers have more possibilities and can be mounted on the wing tips as well as on the rear of the fuselage. But with active onboard trim computers, we become more like birds, finely controlling our aerodynamic surfaces. This will help eliminate the need for ponderous, drag producing stabilizers and lead to greater fuel efficiency.

Planes that "eat like a bird".


That alone seems to imply that the tailplane is the superior choice to any other design, but is it really? Or was it just a case of technological and economic inertia?

A bit of both. And beside the technological features discused in other answers, one should not forget about history here. The tail plane/front engine has its disadvantages, but at the same time an incredible development advantage due WW2. Standard setup planes allow very agile configurations, something fighters depend on. And while many other configurations have been tried during the war, most attention was given to standard ones. As a result knowledge for their construction was widespread after the war and used for civil developments.

It's a bit like the question why the once leading electric car was put out of business by petrol cars - again, WW1 played the major role. While loading batteries isn't a hassle in a developed civil environment, it becomes almost impossible under war conditions - here petrol holds an incredible logistics advantage as moving a few barrels of petrol is easy done, in contrast with setting up power generators to load batteries.

It's not always the war, as economic scaling works as well - like why our modern computers almost all reassemble at the core simplified mini computer constructions - and huge efforts are now added to 'de-simplify' them again.

  • $\begingroup$ The advantages of the conventional tail and the internal combustion engine were established far more broadly than by a couple of military campaigns. Meanwhile the classic Von Neumann architecture of the digital computer has applied from the very first mainframe - the mini-computer era and simplifications such as RISC have nothing to do with the system architecture. $\endgroup$ – Guy Inchbald Feb 10 '20 at 20:37
  • $\begingroup$ Calling WW2 'a couple of military campaigns' is a steep one. After all, these years cover most of the development - and basically all commercial planes after were based on military designs. Also, the evolutionary narrowing of minis is neither about Von Neumann nor RISC, but the reduction of any higher function to the most basic function and replacement of advanced I/O by memory mapped ports. This happened when they were introduced, decades before RISC. $\endgroup$ – Raffzahn Feb 10 '20 at 21:24
  • $\begingroup$ You mentioned two wars not one. Understanding and wholesale adoption of the conventional tail were essentially complete by 1918. I'm not sure how "advanced" data i/o was before universal data buses and memory mapping were introduced, but I'll grant you it could get complicated. $\endgroup$ – Guy Inchbald Feb 10 '20 at 21:47

To fly, a wing must maintain its angle of attack within a certain range. If the AOA is too low, the wing produces no lift. If the AOA is too high, air moving over the top cannot follow the wing's contour, separates, and causes a reduction in lift (aerodynamic stall). Designs that employ passive stability to maintain an appropriate AOA will make the aircraft safer to fly because they don't require the pilot to be actively monitoring AOA and making constant corrections. The tailplane configuration places the aircraft center of mass forward of the wing's center of lift, with the tailplane creating a counterbalancing down force. As the airspeed increases, lift increases, and so does downforce on the tailplane, causing the aircraft to pitch up and slow down. As airspeed decreases, lift decreases, and so does tailplane downforce, causing the aircraft to pitch down and speed up. Similarly, if a disturbance pitches the nose up, increasing its angle of attack, it reduces the downforce on the tailplane, restoring the pitch attitude; if a disturbance pitches the nose down, this increases the downforce on the tailplane, again restoring attitude. The tailplane configuration is thus naturally (passively) stable in pitch, strongly favoring its use.

A drawback of the tailplane configuration is that some of the wing's lift is opposed by the tailplane, which makes it less energy efficient than other configurations such as the canard. A canard configuration can also be designed with passive stability, but it is more difficult to get right, especially when it comes to things like stall recovery.

  • $\begingroup$ This is quite wrong. The tailplane seldom exerts a significant downforce, except during takeoff and landing. The force it exerts is downward only relative to the wing lift and in the lifting tail it remains modestly upward throughout. Moreover the tail does not stabilise speed as suggested. $\endgroup$ – Guy Inchbald Feb 10 '20 at 20:54

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.