Time flows forward. Objects fall, entropy increases, causes precede effects. The arrow of time feels fundamental — written into the structure of the universe rather than emerging from it. But fundamental physics at the microscopic level is time-reversal symmetric: the laws of motion work equally well run backward. The arrow is emergent, and where it emerges from remains one of the deepest questions in physics.
Reader note (2026-11-17): Checked against the full text of arXiv:2603.12434 (Susskind; v1), which I had earlier held as matching on a lighter read. The central claim matches: the paper argues that time reversal (CRT) in de Sitter space is a gauge symmetry hidden by spontaneous symmetry breaking, with a closed curve and a holonomy that exchanges forward-going and backward-going clocks as the smoking gun. Three points differ. First, I searched the full text for arrow, entropy, thermodynamic and second law and found no hits: the paper is about whether time reversal is a gauge symmetry, and it speaks of the direction of a clock, so the essay's framing of the result as an account of where the arrow of time (entropy increase, causes before effects) comes from is the essay's own. Second, the paper's analogy is the Higgs phenomenon in QED with a charged scalar field, plus the holonomy of an electron carried around an SU(2) times U(1) string; I searched for W and Z, W boson, Z boson and electroweak and found no hits, so the essay's symmetry between the W and Z bosons is not from the paper. Third, the paper says the author cannot tell whether Harlow or Witten agree with him, and it ends by saying that without the holonomy there would be no track of the symmetry, which the author thinks is what Coleman and Witten had in mind, so reconciliation of the two positions is the essay's word; the diagrams are worked in 1 plus 1 dimensional de Sitter space, with higher dimensions left as an exercise for the reader.
Susskind (arXiv:2603.12434, March 2026) proposes that time-reversal symmetry in de Sitter space — the geometry that describes an expanding universe with a positive cosmological constant, which our universe approximately is — is a gauge symmetry. Gauge symmetries are redundancies in description, not physical symmetries: changing the gauge doesn't change the physics. If time reversal is a gauge symmetry, forward and backward are different descriptions of the same physics.
But we observe time asymmetry. Susskind's resolution: the gauge symmetry is spontaneously broken. Like the Higgs mechanism, which breaks the electroweak symmetry to give particles mass, something in the cosmological state breaks time-reversal symmetry to give the universe a direction.
The evidence is a closed curve in de Sitter space — a specific geometric path that, when traversed, flips forward-going clocks to backward-going clocks. This is the holonomy: a rotation acquired by parallel transport around a closed loop. The holonomy is the gauge transformation. Its existence means time reversal is available as a symmetry operation. Its spontaneous breaking means the universe has chosen a direction from among equivalent options.
The structural lesson is a reconciliation of two opposing positions. Harlow and Numasawa argued time reversal is a gauge symmetry. Witten argued it is not. Susskind's answer: it is, but you can't see it, because it's hidden by the same mechanism that hides the symmetry between the W and Z bosons. The arrow of time is not fundamental and not absent — it is fundamental symmetry rendered invisible by the state of the universe.