A common source of discussion among teachers is on hip placement in Warrior 2, and the confusion over whether the hips should be “square” to the side or not.
My simple answer to this is a pretty flat: “NO”.
What follows is a little more examination that came out of a discussion I had some time ago with some other teachers. The original discussion was regarding whether or not the hips should be “square” — fully open and perpendicular to the front of the mat — or not. Included in that discussion was some discussion on “goddess” pose and “frog” pose (not bhekasana, the legs bent backbend from Ashtanga 2nd series, but rather mandukasana: a prone pose where the legs are abducted, bent at the knees, and the pelvis is to sink toward the floor). This is an extremely problematic pose for a variety of reasons: one obvious one is the knees, the second is the hips, and the third is the SI joints and the lower back (more on that in a moment).
Part of the concern about allowing the hips *not* to be square in Warrior 2 was about how the hip position affects the lumbar spine. Here’s my view:
The issue with the lumbar spine in Virabhadrasana II is pulling the back (straight leg) hip backward to the physiologically impossible “hips square” position. Pulling that hip back not only compromises the bent leg knee, but it also compresses the SI joint, which is the root of certain forms of lumbar pain — know that external rotation automatically compresses the SI joints, and is the primary reason why most back bends are done with internal rotation. What some describe as “destabilization” in the lumbar spine by bringing the back (left) hip forward is merely a mild twist: hips 120-150º from the bent leg femur, torso and arms 180º to the bent leg femur, for an over all twist of 30º-60º.
The issue with “frog” is not only bilateral abduction and external rotation (which will also compress the SI joints if counter action is not taken to draw the ASIS toward the floor) but gravity is also pulling the femur heads toward the inner thighs, pressurizing the iliofemoral ligaments, femoral arteries, and lower iliacus and psoas muscles, which, since those muscles pass behind the inguinal ligament, can in turn overstretch the inguinal ligament and set up the conditions for inguinal hernia. The pose itself may not cause certain injuries, but it does make one more prone to some of them.
The same pose, done lying on the back, with the feet on the wall and legs bent at the knee 90º and the legs abducted to the student’s capacity — which will be anywhere from 40º from the midline to about 75º degrees from the midline — does not cause as many problems, simply due to the difference in how gravity affects the hip joints, and by extension, the knees. In this position, not only can you ensure that there is no loaded torque in the knee joints due to foot/ankle position, but it also reverses the gravitational load so that the femur heads are drawn AWAY from the inner thighs, thus opening the femoral arteries, and taking pressure off the iliofemoral ligaments, iiliacus & psoas attachments, and by extension, the inguinal ligaments. Having the heels at or higher from the floor than the knees and then isometrically pushing them upward also encourages the external rotation necessary with abduction; whereas facing the floor really only encourages abduction.
It would be extremely rare, even if the student’s acetabulums are set more laterally, for the legs to open beyond 80º from the midline, or a total angle of 160º. Even if a student could open beyond this range, it could be indicative of the hip joint ligaments already being hyper lax, at which point it may be more beneficial from a functional perspective how to restabilize the joints at a far more conservative range of motion. The only time the hips should be square in Vira II is if the opening were a full 180º, but even if that were possible, it is unlikely to be healthy.
So, bringing the back hip forward in Virabhadrasana II (and Trikonasana, and Parsvakonasana) not only saves the front knee, but also saves the SI joints and lower back! There are additional actions that accompany the physical alignment that can further stabilize the lumbar, such as finding and engaging the lower transverse abdominals and lower multifidi (to be looked at soon!).
So, while the back leg hip probably shouldn’t come fully around toward the bent-leg, in a pose that would be closer to Warrior 1 than Warrior 2, but with an added near-90º twist in the spine, there must be some allowance made for the virtual impossibility of having 180º hip abduction. As with most standing poses, working the legs and femurs accurately will take care of the hips, and by extension, the spine.
Consider also how you move into Virabhadrasana II, as some entries are more likely to compress the hip joints, making connection to the back leg and extension through it more difficult; while other entries make it more possible to decompress the hip joints (or at least prevent further compression). We’ll look at those soon, too!
For now, let your back leg’s hip come forward, fully extend the back leg, and keep the front (bent) leg’s femur from collapsing inward.