Fine, but my point is that you could make myriad clocks with all sorts of T formulas, and not have a quantitative explanation of what causes the required change in the quantities T depends on by its formula, so that their values in the other frame work out to match the T in the other frame...