One of the things that many shoes are marketed as providing for the foot is "torsional stability." What that basically means is that they have something in to prevent the sole from twisting at the midfoot. Some of the shoes that offer this also claim to be designed to support "natural" ways of running.
Fully natural running isn't done in rigid supportive shoes. Our feet didn't evolve with that kind of shoe, and they've only been adopted recently. I think a lot of why they have been adopted is the "just-so" logic (anecdotally-based) that has been presented for years. Once we adopted the assumption that too much pronation was bad, we consumers were primed and ready for a long line of technologies designed to prevent or reduce it. From posting, to anti-torsion trusses in the midfoot area of the sole, more structured uppers, and a variety of other patented "features," and buy them we have. Since we already bought into the just-so logic, these anti-pronation features simply made sense to us, and thus the need to reduce pronation became an item of "common sense" in the shoe industry, among runners, and even among podiatrists, sports medicine specialists, orthopedists, physical therapists, chiropractors, and many others.
The term "common sense" usually refers to something there is little empirical evidence for, but which makes sense, based on a series of assumptions. The trouble with common sense, however, is that things that make sense aren't always correct. Sometimes, our common sense is based on erroneous assumptions, and we don't recognize it, because the assumptions are simply believed and are rarely ever tested. The assumptions are mere items of faith. They may be based on what we think of as trustworthy concepts and even observations, but think back on your life at how may times that you thought for perhaps years that you understood how something worked, only to learn that you were wrong. How may times have you thought you understood the relationship between two things, only to discover eventually that the actual relationship was quite different than you believed?
Well, I think that's what's going on in with regard to the human foot. For years, not much thought was put into how it did what it did, or even what exactly it was doing. Then, when we began to get interested, a lot of correct and erroneous ideas about the structure and function of our feet were throw against the great wall of human knowledge. Now, the great wall of knowledge is a funny thing. When you throw things against it that are correct, sometimes they don't stick the first time, but once they do, they hold fast. Conversely, when you throw things that are incorrect against the wall, sometimes they stick for awhile, as if they are actually correct, before they fall off or are knocked loose by new ideas that happen to be correct.
When we became interested in the structure and function of our feet, and began to look closely at them, the context of the entire body, the biomechanics of locomotion, and the evolution of bipedal locomotion was often not part of the equation, and a lot of our ideas about these things were still poorly developed as well (and still are in many ways). Unfortunately, that led to a lot of erroneous ideas being thrown against the wall, and some seemed to stick, they were accepted as truth for decades with little question. Well, some of those ideas have recently begun to come loose from the wall, as a handful of thoughtful folks are beginning to question the assumptions on which the original ideas were based. A lot of this questioning arose from more recent efforts to reconcile what we think we know about the structure and function of the foot with what we observe in the real world, as we seek a cohesive synthesis of the biomechanics of bipedal locomotion (two-legged walking and running, that is).
That's all very interesting, if esoteric, however. So, let's just cut to the chase. What it means to you and me, is that we probably adhere to some ideas about our feet which will be considered obsolete in a few years, the biggest of which is that heel-first striking is a natural or more efficient way to run, the definition of overpronation, and the idea that overpronation is the root of virtually all evil associated with our feet. The more one considers the weight of scientific evidence now becoming available, the more one has to accept these conclusions, I think. Or, alternatively, we can simply have faith in the old paradigm, and cling to it beyond reason.
I hesitate to suggest that we should go to one extreme or another with regard to these ideas. I think rather than assuming that there is a completely right and wrong way to think about them (pronation, high-tech shoes, foot strike, etc...), we should be carefully considering and studying them to determine if parts of these ideas or some ideas they have spawned have merit, and what the limits of that merit are.
Should we all be running around barefoot? Well, if it weren't a good idea to at least have something to prevent cuts, abrasions, and contusions on certain terrain, humans would probably not have bothered developing foot coverings of any kind for summer use. But, they did and footwear really caught on, in general. That's not to say footwear, particularly athletic footwear, hasn't been taken in some ridiculous directions, but it does suggest that a little protection can offer some advantages over being completely barefoot. In turn, it doesn't likewise suggest that more protection and support is better. Indeed, the evidence suggest that our feet serve us better if they are allowed their full range of motion.
Which brings me back to the real topic I started with... Torsional stability. Shoes seek to prevent the forefoot from rolling inward by securing the forefoot and heel and adding plates and trusses to prevent the sole from twisting. It seems obvious that the goal is to prevent or reduce pronation, but here's the thing. When I run barefoot, or in my cross country flats that offer virtually no torsional stability, and I run with a forefoot-first foot strike, my forefoot is able to settle at whatever angle is necessary to adapte to unevenness on the ground due to rocks or sticks or tree roots, etc... adapting to uneven ground surfaces is what the foot seems well adapted for. When you prevent it, and make the entire foot adapt to uneven surfaces, you change that midfoot torsional adaptation to instability at the ankle, adding lateral stresses to both the ankle and the knee.
Why would a runner ever want to do that, for Pete's sake? I think I'd much rather have a natural forefoot strike, and allow my forefoot to adapt to irregularities on the running surface without transmitting those lateral forces upward to the ankle and knee, if I can help it. Based on the set of assumptions that I accept today, torsional stability just seems like a recipe for ankle and knee injuries.
To read the marketing materials of some running shoe manufacturers, the only aspect of running form that is important is which part of the foot strikes the ground first on each stride. If a runner is forefoot striking, that is enough to qualify as "natural" running. I find this idea lacking, however. The foot is adapted for more than just that. The same company also claims their shoes have flexible soles, because they bent relatively easily. The problem is, in my view, they don't twist easily, so the foot can do what it has evolved to do and adapt to effectively to the ground.
I'll stick with shoes that have more torsional and longitudinal flex, thank you, rather than transmit instability and lateral forces to more injury prone ankles and knees that aren't as able to deal with them.