S-Matrix Unitarity, Analyticity, and Resonances

Unitarity relates a physical-cut discontinuity of a scattering amplitude to on-shell intermediate states; analyticity organizes such boundary values into poles, cuts, sheets, and crossed channels. Together they turn probability conservation into the optical theorem and partial-wave bounds, distinguish stable particles from resonances, and constrain—but do not by themselves determine—high-energy behavior.

Enter through $S$-matrix unitarity for normalization, through partial waves for angular-momentum bounds, through analyticity for sheets and crossing, or through the resonance page if your immediate question concerns unstable states. The causal-growth and Regge pages are depth routes: their conclusions require the earlier unitarity and analytic-domain assumptions.

Check your entry point

Before using a cut or an Argand plot, verify that you can:

• separate $S=1+iT$ from the stripped $\mathcal M$ and its overall delta function;
• write Lorentz-invariant phase space and the two-body flux; and
• state the difference between a pole and a threshold cut.

Repair these in S-Matrix and T-Matrix Normalization, Cross Sections and Decay Rates, and Poles, Cuts, Thresholds, and Stable Particles. For the analytic route, also be able to name a branch, sheet, and continuation path; Branches, Sheets, Analytic Continuation, and Monodromy reviews this material.

Choose a route

Goal	Suggested route	Result
Probability conservation and inclusive rates	S-matrix unitarity → optical theorem	Derive $2\operatorname{Im}\mathcal M$ as an on-shell state sum and match the forward limit to $\sigma_{\mathrm{tot}}$.
Angular-momentum bounds	S-matrix unitarity → partial waves	Put each elastic partial amplitude on its Argand circle and diagnose inelasticity.
Crossed amplitudes and sheets	Analyticity and crossing → causal domains	Identify boundary values and state the hypotheses behind fixed-$t$ dispersion and growth bounds.
Unstable states	Partial waves → analyticity → resonance poles	Distinguish bound, virtual, resonance, peak, cusp, and Breit–Wigner data.
High-energy orientation	Partial waves → causal domains → Regge limits	Keep fixed-angle and fixed-transfer limits distinct and qualify Regge claims.

How the constraints fit together

With $S=1+iT$, unitarity gives

$$T-T^\dagger=iT^\dagger T.$$

The right side inserts a complete set of physical states. In a forward matrix element it becomes an inclusive total rate; after angular projection it becomes a separate nonlinear condition for each partial wave. Analyticity then interprets the same on-shell thresholds as branch cuts and connects different channel boundary values when a justified continuation path exists.

These implications are one-way unless extra hypotheses are supplied. Microcausality and the spectrum condition support analytic domains, but a fixed-$t$ dispersion relation also needs growth control and enough subtractions. A resonance pole affects a line shape, but background, thresholds, and coupled channels prevent a generic peak from identifying the pole. Complex angular momentum can organize fixed-transfer asymptotics, but it does not determine fixed-angle behavior.

The operator, optical, partial-wave, and pole relations are derived in Schwartz 2014, ch. 24, printed pp. 452–477 and Weinberg 1995, §§ 3.6–3.8, printed pp. 147–165. Modern boundary-value and sheet conventions, including their multichannel limits, are reviewed in Mizera 2023, open lecture-note PDF, §§ 3.3 and 5.2, pp. 81–88 and 131–140 and Particle Data Group 2025, review 50, §§ 50.1.1–50.1.2, printed pp. 3–8, PDF.

Chapter guide

1. S-Matrix Unitarity derives the complete-state relation with the chapter normalization and perturbative order counting.
2. The Optical Theorem and Cut Interpretation takes the forward limit, recovers the exact flux factor, and explains what a diagrammatic cut represents without proving the Cutkosky rules.
3. Partial-Wave Unitarity fixes a four-dimensional scalar normalization, derives the Argand circle and its inelastic interior, and checks threshold scaling.
4. Analyticity and Crossing of Amplitudes distinguishes physical sheets, upper and lower boundary values, crossed cuts, and qualified crossing.
5. Resonance Poles, Riemann Sheets, and Unstable States separates pole data from Breit–Wigner parameters, peaks, cusps, and asymptotic states.
6. Causality, Growth, and Analytic Domains states which locality, spectrum, mass-gap, and growth assumptions underwrite analytic domains, subtractions, and Froissart-type conclusions.
7. High-Energy and Regge Limits distinguishes fixed angle from fixed transfer and introduces complex angular momentum, signature, Regge poles, and cuts with their limitations.

Shared normalization and sheet language

The first three pages use

$$\begin{aligned} \mathcal M(s,z) &=16\pi\sum_{\ell=0}^{\infty}(2\ell+1)\\ &\quad{}\times a_\ell(s)P_\ell(z),\\ S_\ell&=1+2i\rho(s)a_\ell(s). \end{aligned}$$

where $\rho(s)=\sqrt{1-4m^2/s}$ for equal-mass elastic scattering. The dimensionless variable $b_\ell=\rho a_\ell$ lies on the elastic Argand circle. This displayed $16\pi$ convention is for distinguishable scalar channels; normalized identical-boson states commonly use a $32\pi$ expansion with even $\ell$ only. Other normalizations are translated by preserving $S_\ell$.

The physical sheet is the one reached from real Euclidean kinematics without crossing a singularity; above an $s$-channel threshold, the physical amplitude is the upper-rim boundary value $s+i0$. “Second sheet” is not a global label in a multichannel problem: a sheet must be specified by the sign choice of each channel momentum.

Review the chapter

Starting only from $S^\dagger S=1$, derive the forward optical theorem, project it onto one elastic partial wave, and draw where the corresponding amplitude lies in the Argand plane. Then explain which analytic continuation moves a resonance pole off the physical sheet and why neither a bump nor a Breit–Wigner fit uniquely fixes that pole near a threshold.

A successful response preserves the flux and $\rho$ factors, names the sheet, and distinguishes an assumption from a theorem.

One compact coefficient check runs through the first three pages. In $\lambda\phi^4/4!$ theory, the identical two-particle state gives

$$2\operatorname{Im}\mathcal M^{(1)}(s,0) =\frac1{2!}\frac{\rho(s)}{8\pi}\lambda^2,$$

so $\operatorname{Im}\mathcal M^{(1)}=\lambda^2\rho/(32\pi)$. The same coefficient follows from the logarithmic discontinuity of the one-loop scalar bubble and from the $\ell=0$ projection in the normalized identical-boson convention $\mathcal M=32\pi\sum_{\ell\,\mathrm{even}}(2\ell+1)a_\ell P_\ell$. If these routes disagree, the likely fault is an identical-state factor, use of the distinguishable-particle $16\pi$ expansion, or a discontinuity-versus-imaginary-part factor of $2i$.

References

• Mizera, Sebastian. “Physics of the Analytic S-Matrix.” Physics Reports 1047 (2024): 1–92. DOI. Open PDF.
• Particle Data Group. “Resonances.” In Review of Particle Physics, 2025 Update, review 50, 2025. Official PDF.
• Schwartz, Matthew D. Quantum Field Theory and the Standard Model. Cambridge University Press, 2014, ch. 24, printed pp. 452–477. doi:10.1017/9781139540940.
• Weinberg, Steven. The Quantum Theory of Fields, Volume I: Foundations. Cambridge University Press, 1995, §§ 3.6–3.8, printed pp. 147–165. doi:10.1017/CBO9781139644167.