Well, I'm hardly an expert in mirowave design. I can't prove that what you're saying is incorrect, but I know that a micrwave oven sets up orthagonal electric and magnetic fields. The magnetic field doesn't cause any acceleration (it can change the direction of a moving charged particle, but the tangential speed remains constant), but the electric field does cause oscillation of the dipole.
What actually happens is that each end of the dipole is pulled in opposite directions. The positive end is "pushed" and the negative end is "pulled" (assuming the convention of drawing electric field lines originating at the positive source of the field). This causes a torque, which is the cross product of the electric dipole moment and the electric field. The rotational motion caused by the torque will stop however once the dipole is perpendicular to the field, which is why a constant electric field is not used. The field is varied regularly with a frequency which puts it in the microwave range of EM radiation, which allows optimal rotation back and forth. In other words, it takes ~ 1/(2.48 x 10^9) seconds for one complete motion, which is why the field polarity (or direction of the field lines) is reversed ~ 2.48 x 10^9 times per second. Intensity of the field could be easily varied to increase the magnitude of the torque produced, and thus the amount of power generated by the motion (which would determine the rate at which the water heated up).
However, this could just be another side of the coin which you were arguing. By modelling the orthogonal fields as photons one could argue that the photons would be absorbed, causing a specific excitation of an electron which would then relax, releasing the energy as a bond vibration. Like I said, I'm no expert at this. However the field explanation would allow for any molecule with an overall electric dipole to be heated, while the photon explanation would only allow for molecules with an electronic configuration permitting an excited state (which is not a nonbonding orbital) with the precise amount of energy supplied by a photon in the microwave range.
Either way, chances are that D_DOOD's molecule of interest would not be damaged (assuming that it doesn't decompose when heated).