The epistemological status of multiverse hypotheses: the limits of scientific falsifiability and the mhesi multi-axis evaluative framework

Multiverse hypotheses — a family of cosmological and physical proposals according to which the universe we observe is one of many physically realised universes, the others being either causally disconnected from ours, or distinguished by different fundamental constants, or distinguished by different fundamental laws — have moved since the 1990s from speculative ideas at the margins of theoretical physics into widely discussed components of mainstream cosmological discourse. Four principal variants crystallised in the 2016–2022 literature: eternal inflation plus the string landscape, Tegmark's Level IV mathematical universe, the Everettian multiplicity of quantum-mechanical branches extended by parameter variation, and bubblenucleation cosmology. The empirical status of these hypotheses is in every case the same: no direct observation of another universe is possible in principle, because the defining property of another universe is precisely that it is not causally accessible from ours. In this article I propose, as the original contribution, the Multiverse Hypothesis Epistemic Status Index (MHESI), a normalised composite metric — bounded on [0,1] — that integrates five epistemological dimensions (the strict Popperian falsifiability principle, Lakatosian research-programme progressivity, Dawid's non-empirical confirmation, internal mathematical-physical coherence, and predictive distinctiveness from non-multiverse alternatives) and returns a quantitative ranking of competing multiverse proposals along the axis of epistemic-scientific status. Applied to the four canonical multiverse variants, MHESI returns values in the range 0.20–0.50, indicating that no current multiverse hypothesis fully satisfies any operational demarcation criterion, but that the variants differ considerably in which criterion they best satisfy.