M theory is a name for a more unified theory that has the different string theories, as we know them, as limits, and which also can reduce, under appropriate conditions, to eleven-dimensional supergravity. There's this picture that we all have to draw where different string theories are limits of this M theory, where M stands for Magic, Mystery or Matrix, but it also sometimes is seen as standing for Murky, because the truth about M theory is Murky. And the different limits, where the main parameter simplifies, give the different string theories -- Type IIA, Type IIB, Type I, and there's eleven-dimensional supergravity, which turns out to be an important limit even though it isn't part of the systematic perturbation expansion, then there's the E8XE8 heterotic string, and there's SO(32) heterotic string.
So M-theory is a name for this picture, this more general picture that will generate the different limits through the different string theories. The parameters in this picture we can think of being roughly hbar, which is Planck's constant, and that determines how important the quantum effects are, and the other parameter is alpha prime, which is the tension, related to the tension of the string, that determines how important stringy effects are. So traditionally, a physicist looking at Type IIA, for example, by traditional weak coupling methods, explores this little region, and if asked how his theory is related to Type I theory, the answer would have to be, "Well I don't know, that's something else."
And likewise, if you ask this observer what happens for strong coupling, the traditional answer was, "Well I don't know." In graduate courses, you learn that you can do more or less anything for weak coupling, but you can't do anything for strong coupling. What happened in the 90s was that we learned how to do a little bit for strong coupling, and it turned out that the answer is Type IIA at strong coupling turns out to be Type I in a slightly different limit, SO(32) heterotic, and so on. So we built up this more unified picture, but we still don't understand what it means
Dr. Ed Witten, Princeton University
My favorite passage from Dr. Witten's statement:
And the different limits, where the main parameter simplifies, give the different string theories -- Type IIA, Type IIB, Type I, and there's eleven-dimensional supergravity, which turns out to be an important limit even though it isn't part of the systematic perturbation expansion
My second favorite passage:
...but we still don't understand what it means