Does TsT preserve the same number of physical degrees of freedom across both formulations?

# TsT and Degrees of Freedom: A Subtle Problem

Your question touches on a deep tension in modified teleparallel theories that the archive reveals clearly.

The short answer: no, TsT (Teleparallel Symmetric Theory) formulations do not unambiguously preserve the same degree of freedom count across different mathematical frameworks — and this is a live problem, not a solved one.

Here's why:

The Core Issue: Hidden Structure

Teleparallel theories add a geometric layer on top of ordinary spacetime: a flat parallel transport (or connection structure) that isn't present in general relativity's formulation. In the GR-equivalent case, this structure is pure gauge — it's unphysical window-dressing. But when you modify the theory (like in f(T) gravity), this structure becomes physical [AG-2024.05-450]. The trouble is that counting degrees of freedom in these theories is technically treacherous.

Two Counting Methods, Conflicting Results

There are (at least) two rigorous ways to count degrees of freedom:

1. Hamiltonian/constraint analysis — the Dirac-Bergmann method applied to the Lagrangian. When done carefully in modified teleparallel gravity, it predicts "much more dynamics than what has ever been seen" [AG-2024.05-450].

2. Linear perturbation analysis — what you see by examining small fluctuations around a background spacetime. In f(T) gravity, this often gives fewer propagating modes than the Hamiltonian method predicts [AG-2024.05-450].

The fact that these disagree suggests either (a) hidden instabilities ("strong coupling"), (b) mistakes in the simpler treatments, or (c) both [AG-2024.05-450].

Modern Progress: Equation-Based Counting

Recently, a new method has emerged that bypasses the Lagrangian/Hamiltonian dichotomy and counts degrees of freedom directly from field equations [AG-2025.01-981, AG-2025.09-408]. When applied to f(Q) gravity (the nonmetricity analogue of f(T)), this yields seven propagating degrees of freedom [AG-2025.09-408] — a result traditional methods couldn't reliably reach.

Real Example: The Mismatch

In quadratic scalar-nonmetricity theory, Hamiltonian analysis gives 10 DOFs in one case, but linear perturbations around cosmological backgrounds show only 6 or 5 modes [AG-2025.12-739]. This discrepancy hints that formulation matters—what you count as "physical" depends on how you ask the question.

Bottom Line

TsT theories sit at the frontier where the standard toolkit (Lagrangian → Hamiltonian → perturbations) breaks down. A reformulation from one framework to another does not guarantee the same degree of freedom count emerges transparently. The newer equation-based methods [AG-2025.09-408] offer hope for systematic, formulation-independent answers, but they're still being developed.

The archive doesn't yet have a paper directly comparing TsT across multiple formulations side-by-side.