Excipients are conventionally described through intrinsic material properties, pharmacopeial specifications, and functional labels such as binder, disintegrant, solubiliser, stabiliser, or release modifier. This vocabulary has supported pharmaceutical development for decades because it simplifies excipient selection and links material identity to expected product performance. Yet the same vocabulary becomes unstable when dosage forms are compositionally dense, structurally heterogeneous, and highly dependent on manufacturing history. The central problem is that excipient performance in complex dosage forms often deviates from what would be predicted by isolated material tests. A polymer that stabilises supersaturation in one amorphous solid dispersion may fail in another, while a lipid excipient that improves solubilisation under one digestion condition may promote precipitation under another. Such behaviour suggests that excipient functionality is not merely carried by the excipient molecule, but is produced within the dosage form system. This article proposes a theoretical reframing of excipient functionality as a system property. In this view, functionality emerges from the combined effects of formulation composition, spatial architecture, and processing history. The purpose is not to replace molecular or compendial characterisation, but to relocate those measurements within a broader systems framework. The proposed theory defines excipient functionality as an emergent outcome of interactions among drugs, excipients, process energy, phase behaviour, and microstructural organisation. It explains why apparently similar formulations can display different dissolution, supersaturation, release, or stability behaviours when their processing route or internal architecture differs. Three tables are used to contrast the reductionist and system-property paradigms, map overlooked interactions, and identify design implications. Adopting a system-property view would shift pharmaceutical development from selecting excipients as isolated ingredients toward designing excipient functions as relational outcomes. It would encourage formulation scientists to evaluate not only what an excipient is, but what it becomes within a particular dosage form. This perspective offers a conceptual basis for more predictive, adaptive, and robust pharmaceutical product design.