Real-Time QFT, Kinetic Theory, and Hydrodynamics

Schwinger–Keldysh, kinetic theory, and hydrodynamics describe the same real-time problem at different resolutions. The closed-time-path functional retains quantum correlations and the initial density matrix; Kadanoff–Baym or $n$PI equations close a hierarchy; kinetic theory assumes quasiparticles or controlled scattering; hydrodynamics keeps only conserved and parametrically slow fields. Moving down this chain gains reach and loses microscopic information, so every handoff needs a scale-separation and error argument.

Evidence cutoff. 11 August 2026.

Required background. Closed-time-path generating functionals supplies contour and initial-state data, distribution functions and transport equations supplies kinetic closure, and conservation laws and hydrodynamic fields supplies the infrared degrees of freedom. Helpful background. Kadanoff–Baym evolution supplies two-point closure, linearized relaxation modes supplies rates, and frame-invariant dissipative data supplies physical transport variables.

A hierarchy of real-time descriptions

Targets include response and spectral functions, equilibration, transport coefficients, expanding plasmas, critical dynamics, open-system noise, and late-time fluctuations. Required inputs are the Hamiltonian or effective action, initial density matrix, conserved charges, sources, hierarchy of time and length scales, and the observable’s resolution.

Description	Applicability and inputs	Main approximation or error
Exact Schwinger–Keldysh functional	Arbitrary initial density matrix and real-time correlators on a doubled contour	Formal identity until evaluated; sign/phase problem and renormalization
2PI/$n$PI or Kadanoff–Baym	Correlators with a specified loop, coupling, or $1/N$ closure	Memory-kernel and vertex truncation; conservation and symmetry must be preserved
Kinetic theory	Dilute excitations, quasiparticles, or another controlled on-shell/gradient limit	Collision-kernel order, off-shell width, molecular chaos, gradient expansion
Deterministic hydrodynamics	Wavelengths long compared with nonhydrodynamic relaxation; conserved/slow fields	Constitutive gradient truncation, frame choice, causal completion
Fluctuating/SK hydrodynamic EFT	Near-local equilibrium with noise and KMS constraints	Loop-sensitive transport renormalization, stochastic discretization, finite cell size

Schwinger’s source-contour formulation and Keldysh’s nonequilibrium Green functions make normalization, causal response, and statistical correlators simultaneous (Schwinger 1961; Keldysh 1965). The contour does not choose a closure. A 2PI effective action can preserve conservation laws at a fixed truncation but may violate crossing or miss vertex physics; direct Kadanoff–Baym evolution is limited by memory cost and initialization.

Kinetic theory follows after a Wigner transform and a separation among microscopic, collision, and macroscopic scales. At weak coupling, systematically matched collision operators yield leading-order transport coefficients; Arnold, Moore, and Yaffe demonstrate how screening, collinear splitting, and the Landau–Pomeranchuk–Migdal effect all enter the leading result (Arnold, Moore, and Yaffe 2003). A relaxation-time ansatz may be useful phenomenology, but it does not inherit that accuracy.

Hydrodynamics requires slow conserved modes, not necessarily a dilute gas. Schwinger–Keldysh hydrodynamic effective actions encode unitarity, fluctuation–dissipation, and dynamical KMS constraints (Crossley, Glorioso, and Liu 2017). Recent work extends this logic to approximate symmetries and derives the parametric form of relaxation terms (Hongo et al. 2024). These symmetry constraints fix allowed structures; their coefficients still require matching or data.

Error propagation and benchmarks

A complete error chain separates initial-state uncertainty; contour/source discretization; UV regulator and renormalization; hierarchy closure; quasiparticle and gradient errors; collision-kernel perturbative order; constitutive truncation; stochastic sampling; and matching covariance. Transport coefficients extracted by fitting hydrodynamics to a microscopic evolution inherit both microscopic and hydrodynamic errors. Varying the hydrodynamic starting time is not a substitute for adding the missing nonhydrodynamic modes.

Useful benchmarks are free Gaussian quenches, exactly solvable large-$N$ or kinetic models, linear-response Kubo relations, weak-coupling transport matched between diagrams and Boltzmann equations, Bjorken or Gubser expansion, and critical diffusion with known dynamic universality. A kinetic and hydrodynamic computation are not independent if the latter’s coefficients and initial state were fitted to the former. Stronger validation withholds an observable or compares to lattice spectral constraints, experiment, or a distinct microscopic theory.

Failure modes and claim ceilings

Pinch singularities signal the need for resummation; secular growth can invalidate finite-order perturbation theory. A quasiparticle description fails when widths are comparable to energies. A gradient series can be asymptotic even when hydrodynamics predicts observables well; causal stability depends on the chosen completion and coefficients. Near a critical point, additional slow fields must be promoted. Hydrodynamic agreement does not prove local thermal equilibrium or identify microscopic carriers.

Claim ceiling. Exact Schwinger–Keldysh identities organize observables but do not solve an interacting theory. A controlled kinetic calculation establishes transport only in its scale and coupling regime. Hydrodynamics establishes universal long-wavelength form and matched coefficients, not microscopic thermalization or a unique UV theory.

Choosing the resolution scale

Question	Appropriate starting point	Do not use it to claim
Short-time quantum coherence or arbitrary initial state	Schwinger–Keldysh / Kadanoff–Baym	A closed numerical result without a tested truncation
Weakly coupled scattering and transport	Kinetic theory with matched collision kernel	Strong-coupling accuracy by analogy
Late-time, long-wavelength response	Hydrodynamics with frame-invariant data	Quasiparticles or local equilibrium without separate evidence
Noise, full counting statistics, long-time tails	Fluctuating or SK hydrodynamic EFT	Regulator-independent coefficients without renormalization
Overlapping regime	Match adjacent descriptions on the same observables	Independence if all inputs came from one calculation

Related routes include thermal and nonequilibrium field theory, real-time continuum dynamics from regulators, the hydrodynamic-attractor dossier, and the hydrodynamic-attractor evidence brief.

Evidence cutoff and change criteria

The finite selection covers the contour foundations, controlled weak-coupling transport, nonequilibrium closure, hydrodynamic effective actions, and approximate-symmetry extensions. Targeted arXiv, journal, transport, and citation-chain searches covered public evidence through 11 August 2026. Reassessment is triggered by a closure counterexample, a new matched perturbative order, a transport benchmark failure, a causal-stability revision, or a new slow mode changing the EFT.

References

• Arnold, Peter, Guy D. Moore, and Laurence G. Yaffe. “Transport Coefficients in High Temperature Gauge Theories: (II) Beyond Leading Log.” Journal of High Energy Physics 2003, no. 5 (2003): 051. DOI.
• Crossley, Michael, Paolo Glorioso, and Hong Liu. “Effective Field Theory of Dissipative Fluids.” Journal of High Energy Physics 2017, no. 9 (2017): 095. DOI.
• Hongo, Masaru, Noriyuki Sogabe, Mikhail A. Stephanov, and Ho-Ung Yee. “Schwinger–Keldysh Effective Action for Hydrodynamics with Approximate Symmetries.” arXiv:2411.08016 (2024). arXiv.
• Keldysh, L. V. “Diagram Technique for Nonequilibrium Processes.” Soviet Physics JETP 20 (1965): 1018–1026. DOI.
• Schwinger, Julian. “Brownian Motion of a Quantum Oscillator.” Journal of Mathematical Physics 2 (1961): 407–432. DOI.