The Kuramoto order parameter summarizes collective synchronization in a single complex number: R tells you how synchronized the oscillators are, and Ψ tells you the direction they point. When oscillators lock phases, R approaches 1 and Ψ is the shared phase angle. The framework is standard in every synchronization textbook.
Veronica Sanz asks: when does Ψ actually mean anything?
The question sounds trivial. Ψ is defined — take the argument of the complex order parameter. Compute it. There's a number. But in incoherent states, where oscillators point in every direction, that number is a ghost. It's computable but not estimable. Add a small amount of noise or change your finite sample, and Ψ jumps wildly. The quantity is formally defined and operationally meaningless.
Sanz proposes an operational emergence criterion: Ψ is a real macroscopic coordinate only when it can be robustly estimated from noisy, finite measurements. Not just computed from the exact state vector in a clean simulation. Estimated from the kind of data an experiment actually produces.
The distinction separates two things that are usually conflated. Mathematical definability means a formula exists. Operational emergence means the formula's output is stable under perturbation — you get approximately the same answer from different samples of the same system. In coherent states, Ψ passes both tests: the formula exists and its output is robust. In incoherent states, only the first test passes. The formula exists. Its output is noise.
This matters because the onset of synchronization is conventionally described as R increasing from 0. But Sanz's criterion reveals a second transition hidden inside: the global phase Ψ goes from operationally meaningless to operationally meaningful. These two transitions — amplitude and phase becoming emergent — can occur at different parameter values. The system can be partially synchronized (R > 0) while the phase Ψ remains a ghost (not robustly estimable).
The time-dependent coupling makes this sharper. When coupling ramps up quickly, the system freezes into a partially ordered state where R is nonzero but the phase never stabilizes. The ramp protocol determines whether the phase emerges. Slow ramps: the system tracks the synchronization transition and Ψ becomes real. Fast ramps: the system gets stuck and Ψ remains computable but meaningless. The procedure creates or prevents the emergence.
The structural point extends beyond oscillator networks. Any system described by an order parameter has both a mathematical definition and an operational one. The mathematical definition exists everywhere — it's a formula, always computable. The operational definition exists only where the formula's output is stable to perturbation. The gap between these two definitions is where ghosts live: quantities that exist in the equations but not in the world.