Experiments describe physical regularities of remarkable stability. That stability does not, on its own, establish that the laws are eternal. Physics therefore distinguishes an observation — the absence of detected variation within a given domain — from a more general hypothesis about the origin or the immutability of the laws.

What can be measured

A physical variation must be expressed through a dimensionless quantity. The fine-structure constant α and the proton-to-electron mass ratio are two examples. An isolated variation of a dimensioned quantity would depend on the chosen system of units and would not, on its own, provide an experimental statement.

Atomic-clock comparisons look for a relative drift between different transitions. Astronomical spectra test far greater epochs and distances, with other instrumental and astrophysical uncertainties. Geophysical and cosmological data add further distinct scales. No single method therefore covers every possible form of variation on its own.

Published results impose very strict limits without bringing to light any confirmed drift. Their precision depends on the constant, the period and the model used. It would be misleading to summarise the whole by a single bound. Jean-Philippe Uzan's review presents the principles and the constraints of these various tests. Modern clock comparisons illustrate the sensitivity attained at laboratory scale.

Phase transitions and effective laws

The primordial Universe went through changes of temperature and density. In physical models, some of these changes may correspond to phase transitions. If a first-order transition took place, the nucleation and collision of bubbles could have produced a gravitational-wave background. Work in this area evaluates the frequencies and amplitudes that might be accessible to observatories such as LISA.

A phase transition does not demonstrate that the fundamental laws have changed. It describes a change of state, of symmetry or of vacuum within a theoretical framework that keeps its equations. The effective properties of particles and interactions may then differ from one regime to another. This distinction avoids confusing the evolution of a physical state with that of the rules used to describe it.

The question of the electroweak vacuum

Starting from the measured parameters of the Standard Model, several calculations place the electroweak vacuum near a boundary between stability and metastability. This conclusion depends notably on the mass of the Higgs boson, on that of the top quark and on the precision of the calculations. Metastability means that a state is long-lasting without necessarily being the state of minimum energy.

These calculations do not constitute an observation of vacuum decay. They evaluate a possibility internal to the model and show that the estimated lifetime may be far greater than the present age of the Universe. They therefore do not allow one to announce a future change of the laws, nor to infer from it an instability observable at our scale.

The limits of interpretation

A regularity may be compatible with several explanations. It may reflect a fundamental constant, the stabilisation of a field, or a dynamic whose variation lies below the detection thresholds. To decide between these possibilities, a hypothesis must specify the quantity concerned, the form of the expected variation and the protocol capable of distinguishing it from a systematic error.

The weak anthropic principle reminds us that our observations are conditioned by the existence of observers. It proves neither the variation of the constants nor their local character. Used with caution, it signals a selection bias. It does not replace a quantitative prediction.

SUMMARY

Measurements support a strong stability of the constants without demonstrating their absolute immutability. Phase transitions concern physical regimes and do not prove an evolution of the fundamental laws. Any more ambitious proposal must produce a measurable variation and a test capable of isolating it.

References