Thesis. The SI second is a convention pinned to a hyperfine transition of cesium-133 because that transition is stable, reproducible, and politically uncontested — not because cesium is metaphysically privileged. ODTOE replaces this with a stricter claim: the natural unit of time for any observer O is the period of O's own coherence cycle. Cesium clocks work for us because human bodies and human civilizations happen to share a coherence band that brackets cesium frequencies.
What the SI second actually is
Since 1967, the second is defined as the duration of 9,192,631,770 periods of the radiation corresponding to the transition between two hyperfine levels of the ground state of cesium-133 at 0 K. Beautiful, precise, and entirely operational. It does not claim that this duration is "what a second really is" — it claims that if you want a stable, reproducible second, here is one.
That last clause is where ODTOE pushes.
The hidden observer in metrology
Every measurement of "how long" presupposes an observer who can register the start and the end. The cesium clock works because we, as biological-and-cognitive observers, can register events on the time scale of seconds to fractions of seconds, and the cesium transition runs fast enough that we can count many of its cycles within our own perceptual window.
For an observer whose coherence cycles run a million times faster, the cesium tick would be unbearably slow — like trying to keep time by watching glaciers. For an observer a million times slower, the cesium tick would be invisible noise. The unit only "works" within a band, and the band is set by the observer's B(O, C), not by cesium.
The time as a strange loop paper makes the formal version of this argument: time is the coherence cycle of self-referential observers, and any other "time" is a derived quantity, indexed to a specific observer's clock.
Why this matters for relativity
In standard general relativity, time dilation is a property of frames: two clocks at different gravitational potentials tick at different rates. In ODTOE's reading, this is the geometric shadow of a deeper fact: the two clocks are observers whose coherence rates are differently modulated by the local curvature of the configuration field H. The chirality of time / Kozyrev paper shows that this also predicts a small directional asymmetry that pure GR does not — Kozyrev's experimental observations are unparsimonious in GR but natural in ODTOE.
Three things this changes
- Atomic clocks are convenient, not fundamental. Picking cesium was an engineering choice; an extraterrestrial civilization with different coherence biology would naturally pick a different oscillator. There is no "absolute" frequency.
- Biological time is not a "felt" approximation of "real" time. It is real time, indexed to a real observer. Cesium time is real time, indexed to a different (effective) observer — the metrological collective. Both are configurations.
- Time dilation in extreme regimes is observer-coherence dilation. Near a black hole, near light speed, in superfluid phases — the deviations from clock universality are signatures of the coherence field's nontrivial geometry, not of "time itself" slowing down. See Temporal asymmetry for the configuration-field formulation.
What about practical metrology
Nothing in ODTOE breaks practical metrology. Cesium clocks remain the best we have for human-scale civilization, and they remain wildly useful for GPS, financial markets, particle physics. ODTOE just declines to treat them as the absolute reference. They are an excellent convention. Conventions are not foundations.
The time basic introduction lays out which parts of time-talk survive ODTOE and which do not, in a form accessible without the formal apparatus.
Cite this post
Pankratov, A. (2026). Why Cesium-133 Defines a Second — and Why ODTOE Says That's Wrong. ODTOE Blog. https://odtoe.org/blog/why-cesium-133-defines-second-and-odtoe-disagrees