MXenes are commonly described by the compact formula Mn+1XnTx, yet this notation hides the variables that most strongly determine their behavior: precursor ordering, extraction pathway, vacancies, termination identity and distribution, interlayer ions and water, stacking, and transformation history. This mini review develops a solid-state chemistry framework in which MXenes are treated as chemically programmable metastable solids. The central argument is that synthesis does not simply reveal a pre-existing two-dimensional carbide or nitride sheet; it drives the material along a coupled reaction coordinate that creates a history-dependent lattice, surface, and interlayer state. We connect precursor crystal chemistry to selective extraction, show why terminations and defects must be considered together, interpret intercalation as a structural component of the three-dimensional stacked solid, and examine oxidation and thermal conversion as mechanistic probes of metastability. Recent operando, spectroscopic, computational, and synthesis studies are integrated to identify transferable structure-property relations. We finally propose a minimum state descriptor and experimental priorities for converting empirical recipes into predictive synthesis. This perspective places MXenes squarely within solid-state chemistry and clarifies why nominally identical compositions can display divergent conductivity, mechanics, ion storage, and reactivity.