When aircraft pull too much angle of attack, the wing abruptly suffers a loss of lift anywhere the maximum lift coefficient has been exceeded. If your entire wing loses lift all at once this can be very dangerous. As a result, most wings are designed with what's called Washout, wherein the angle of incidence at the root is greater than at the tip. This causes the wing root to stall before the tip, which allows a pilot to maintain control as the onset of stall appears.

Using a forward-swept wing causes the spanwise (along the wing's length) flow to be net inward, rather than outward which is the case in a conventional aircraft. With the right mix of composite materials to control aeroelastic twist this causes a similar effect to washout wherein the wing root will stall before the tips, but greatly reduces wingtip vortices and thus reduces total aerodynamic drag. As someone else pointed out, you also get an induced drag benefit because both the canards and main wings are producing upward moments, so the total magnitude of lift force required for a given weight is less.

The "so what?" of all that is it lets you build a smaller, faster, more agile aircraft at the cost of significantly increased aerodynamic and control complexity.

> This causes the wing root to stall before the tip, which allows a pilot to maintain control as the onset of stall appears.

I think I've read and been told that also you want the root where the flaps are to stall before the middle where the ailerons are. Because stalling and losing your aileron control during a turn is bad news.

If one of your wing tips stalls, you will have a very fast roll and there is a high probability of entering a spin.

Also if a part of the wing that is far aft stalls, then that can cause pitchup, making the stall worse.

An aircraft with benign behavior stalls so that the nose stays pointing straight ahead with no roll or yaw but it drops (pitch).

The loss of aileron control isn't hugely dangerous in and of itself. What's dangerous is the incipient spin. There is a yaw-roll coupling in most aircraft such that you can use the rudder to bank if you have to, even with zero aileron authority.

There’s generally a tradeoff in control systems between stability and performance. Which is to say: by definition, the more stable a system is, the harder it is to get that system to move away from its setpoint. The classic response to a program manager who wants the system to be “as stable as possible” is that a boulder sitting on the ground is stable, but you probably don’t want it.

If you want an aircraft that’s incredibly responsive, then you probably want an aircraft that’s just this side of unstable. Or, in the X-29’s case, an aircraft that actually is unstable and being actively stabilized by the flight computer.

My pet peeve is that while I want any mechanism I use to be "responsive", everybody seems to market things as being "adaptive", and it doesn't seem to be appreciated that being adaptive is fundamentally opposed to responsiveness.

I wonder if the military gets suckered into buying things that are "adaptive"?

I don't think it's mentioned in the article, but it's to reduce shockwave formation, generating more lift, efficiency and maneuverability. I posted a link to a real engineering video in another comment that goes into a lot more detail.

Kind of strange to write an article about a particular design feature and not say what the point of it was?!

It's not very explicit but it's there (around mid-point of the article):

"The Hansa Jet wings are also swept forward by just a few degrees, compared to 33 degrees in the X-29. Such a radical adjustment meant trading stability for maneuverability, because to maneuver quicker, a plane must be inherently unstable to start with."

...

"At the time, this maneuverability was believed to be absolutely essential to fighter superiority. If your airplane is going to stall before mine, I can shoot you out of the sky in a heartbeat."

Greater subsonic maneuverability, same motivation for the Su- 47