Reading The Design of Everyday Things, is a great trip; I recommend it because it leads down the path of why design works, and maybe the summary answer is: because it is human.
What this means, practically, is that we don’t have to think about interfacing with it, we just use it; the thinner the manual, the better the design.
A human has a training program that is about 10-15 years long in which they master their senses and their motor skills, from stereo vision to talking , to walking and working with 2 hands and two feet and a skin sensitive to a wide set of stimuli. Not to mention hearing, sonic location on 3D, and chemical analysis through nose and taste.
This training should not be discarded in the name of technology and its interface: if you learned a qwerty keyboard when you were 5, great, but don’t assume that it is as natural as doing what you do to communicate with another person through speech, gesture and nuance.
Technology is getting close to the point (so close!) that these natural interactions are possible. Keyboards are being assisted with gestures and voice everywhere and the fact that the manuals for these devices are thin or nonexistant says we are on the right track. I have an Acer touchscreen computer running windows 7, and although it has a keyboard and mouse, I hardly use them – they seem alien and disconnected when a few gestures and screen handwriting (or even onscreen keyboard) work just fine.
Another example is robotics. Replacing humans on the International Space Station for spacewalks and repairs has been a goal of the Robonaut program. NASA chose a human analogue for the robot, intentionally mapping arms, hands, finger joints, and stereo cameras to the dimensions and actions of a human body. Although technically not the best solution, nor the simplest, they feel the intuitive Avatar effect will more than compensate in terms of learning curve and efficiency. Once this design is solid, it can be reused in other missions, and the strange sight of a human torso, arms, and head centaur-ed onto space probes and submersible robots may become common.
So why didn’t technology mimc us in the past (or even now)? technical limitations. Good design overcame some, but not all, especially as functioonality increased beyond what could be presented intuitively. A telephone from 1955? Easy to figure out. A telephone from 1995 office? Needs training, manual, and reference card just to check voicemail. A phone from 2010? Again, no training anymore. The technology caught up to the new functionality with voice dialing, touch interface and good UI design.
This is a constant race, though and expect there to be lulls and leaps in usability as functionality outstrips interface technology, and interfaces struggle to catch up.