Since equations of state relate only pressure, volume, and temperature, they exclude stress fields other than pure hydrostatic pressure/tension. As such—they are not of themselves useful in describing the general stress‐deformation‐temperature response. Constitutive equations are required. The latter are, in turn, subsets of still more general expressions, the free energy functions. Rheological constitutive equations incorporate time, but introduce the need to carefully distinguish the independent variable as the stress or the deformation. Moreover, for glassy solids they must incorporate at least two distinct, yet interacting, memory functions, while providing for the physical aging process. The paper comments on thermodynamics and free energy functions and on some conceptual difficulties, including the definition of reference states for strain, especially volumetric strain. For glasses, the volumetric reference state is a particular problem because the unloaded state is metastable, and physical aging can occur. Hence, equation‐of‐state information in the otherwise undeformed state can provide a starting point for the development of constitutive equations. The underlying free energy function might be derived via statistical mechanics or taken from continuum mechanics. The Simha‐Somcynsky equation of state, based on statistical mechanics, is used to discussV‐Teffects explicitly. A possible means of modifying the underlying cell model is indicated, showing how the model can in principle be modified to produce a constitutive equation. Continuum mechanics approaches are then discussed. Examples are given of developments based on linear viscoelasticity theory, but directly incorporating stress‐induced volume changes (Knauss‐Emri and Shay‐Caruthers), and on large‐strain elastic theory (Peng‐Landel). Problems with all approaches to trying to describe glassy response are pointed out, especially in that they attack different facets of the overall problem.