I have a very basic problem in understanding the relationship between electric field and displacement. If a field is applied to a dielectric, it is clear that this will cause the material to polarize, and one can define the permittivity of the material to quantify the size of this effect. Since we can write [itex]D=\epsilon E[/itex], we can also define the inverse permittivity [itex]E=\kappa D[/itex]. But conceptually, I can't make sense of the displacement as the independent variable. How does a polarization give rise to a macroscopic field in the same direction? If anything, a polarization seems associated with a field in the opposite direction (by imagining the situation in the center of a dipole), although I know that macroscopic E&M says nothing about these microscopic fields. If I think about the case of stress and strain, I can easily imagine how either one gives rise to the other. Where is my problem?