Does TsT preserve physics because it's a true duality, or just an effective mathematical reparametrization?

Your question cuts to something philosophers and physicists have debated intensely: when two mathematical formulations describe the same physical system, are they *genuinely equivalent* or merely two languages for the same underlying truth?

TsT (T-duality S-duality T-duality, a chain of string theory dualities) sits in a subtle territory. The short answer: it's likely more than mere relabeling, but the question of what counts as "true" equivalence is still contested.

Here's why the distinction matters. A "true duality" means two theories are physically indistinguishable—they make identical predictions, describe the same degrees of freedom, and differ only in which variables you choose to highlight. A "reparametrization," by contrast, is just swapping notation: like writing $E = mc^2$ versus $m = E/c^2$. The physics doesn't change; you've just renamed the players [AG-2025.08-036].

The philosophical crux is theoretical equivalence: do two formulations "say the same thing in different words"? [AG-2025.08-036]. This is harder to define than it sounds. A cautionary example from the archive: teleparallel gravity—which describes gravity via torsion in flat spacetime instead of curvature in curved spacetime—is *often claimed* to be equivalent to general relativity. Yet [AG-2024.06-418] argues the theories are not categorically equivalent, because teleparallel gravity posits additional geometric structure (the flat connection) that general relativity does not require. In GR-equivalent teleparallel models, that extra structure is "pure gauge"—unobservable and arbitrary [AG-2024.11-413]. That suggests a reparametrization, not a duality.

String theory dualities like TsT differ from this case: they typically *do* map between regimes where coupling strengths or energy scales flip in a way that makes previously intractable calculations tractable. If you can't compute in one description but can in its dual, the duality has real explanatory teeth. However, [AG-2025.08-036] notes that dualities raise questions about scientific realism: if string theory can be written in multiple inequivalent ways (M-theory's many formulations), which one reflects "reality"?

The honest answer: TsT likely preserves the physics in a deep sense—it's a symmetry of the string worldsheet equations—but whether that makes it a "true duality" or an "effective reparametrization" depends on philosophical commitments about what counts as the same physics [AG-2025.08-036]. In practice, physicists treat it as a true duality because it connects weakly-coupled and strongly-coupled regimes that would otherwise be opaque to calculation.

The provided archive doesn't directly address TsT's status, so I can't ground a more specific verdict in these papers.