Exact S-Matrix Bootstrap, CDD Freedom, and Completeness

Factorization reduces the relativistic scattering problem to two-body amplitudes, but unitarity, crossing, analyticity, symmetry, and a proposed spectrum generally do not determine those amplitudes uniquely. Even the one-species scalar problem admits multiplicative CDD factors. “Exact S matrix” therefore means an exact solution for a declared model and analytic class, with its particle content and CDD choice stated; algebraic consistency alone is not a construction of a local QFT.

Required background. Elasticity, factorization, and their hypotheses supplies the two-body reduction and its massive-asymptotic assumptions. Analyticity and crossing of amplitudes supplies physical sheets, crossed channels, and boundary values. Helpful background. The Yang–Baxter equation supplies matrix-index consistency when the amplitude is not scalar.

Scalar bootstrap conventions

Consider one stable, neutral particle of mass $m$ in a massive $1+1$-dimensional relativistic QFT. We normalize the two-body amplitude $S(\theta)$ as the factor acquired when two ordered asymptotic particles exchange, where $\theta=\theta_1-\theta_2$. On the real axis, physical unitarity and elastic completeness give

$$S(\theta)S(-\theta)=1.$$

For a self-conjugate scalar in this convention, crossing is

$$S(\theta)=S(i\pi-\theta).$$

We impose real analyticity as

$$S(\theta)^*=S(-\theta^*).$$

The physical sheet is the analytic continuation reached from real center-of-mass energy with the Feynman prescription. In the rapidity uniformization, its direct-channel strip is

$$0<\operatorname{Im}\theta<\pi.$$

Simple poles in that strip require interpretation; branch cuts associated with multiparticle production are absent in a strictly elastic factorized model after uniformization, but poles, zeros, and crossed images remain. These conventions agree with the scalar equations used by Zamolodchikov and Zamolodchikov 1979, §§ 2–3, pp. 257–269.

The constant amplitudes

$$S_0(\theta)=+1 \qquad\text{and}\qquad S_0(\theta)=-1$$

solve the three displayed equations. Which constant corresponds to the intended free theory depends on the exchange and statistics convention. This already shows that the functional equations require physical input.

CDD freedom

For real parameter $\alpha$, define

$$C_\alpha(\theta) = \frac{\sinh\theta+i\sin\alpha} {\sinh\theta-i\sin\alpha}.$$

Because $\sinh(-\theta)=-\sinh\theta$,

$$C_\alpha(\theta)C_\alpha(-\theta)=1.$$

Because $\sinh(i\pi-\theta)=\sinh\theta$,

$$C_\alpha(i\pi-\theta)=C_\alpha(\theta).$$

For real $\theta$, the numerator and denominator are complex conjugates, so $\lvert C_\alpha(\theta)\rvert=1$. Thus, if $S(\theta)$ satisfies scalar unitarity, crossing, and real analyticity, so does

$$\widetilde S(\theta)=S(\theta)C_\alpha(\theta).$$

For $-\pi<\alpha<0$, the denominator has no zero in the physical strip, so the factor adds no bound-state pole there; it does add physical-strip zeros. For $0<\alpha<\pi$, poles occur where $\sin(\operatorname{Im}\theta)=\sin\alpha$ and can change the proposed bound spectrum. More general products and ratios of such blocks produce the CDD ambiguity first isolated in dispersion theory by Castillejo, Dalitz, and Dyson 1956, pp. 453–458.

“Minimality” is therefore an additional selection principle, not an equation: one chooses the solution with no unnecessary physical-strip singularities and the smallest high-energy growth compatible with the declared spectrum and asymptotics. Different ultraviolet completions or integrable deformations can select different CDD factors while satisfying the same elementary functional equations.

Worked pole-free solution

The scalar sinh-Gordon scattering function provides a concrete single-species solution:

$$S_B(\theta) = \frac{\sinh\theta-i\sin(\pi B)} {\sinh\theta+i\sin(\pi B)}, \qquad 0<B<1.$$

This is $C_{-\pi B}(\theta)$. Direct substitution gives

$$S_B(\theta)S_B(-\theta)=1, \qquad S_B(i\pi-\theta)=S_B(\theta).$$

For a possible physical-strip pole set $\theta=iu$ with $0<u<\pi$. The denominator becomes

$$i\bigl[\sin u+\sin(\pi B)\bigr],$$

which cannot vanish because both terms are positive. Hence this convention has no physical-strip bound-state pole. Its numerator can vanish, so the absence of poles does not mean the amplitude is constant.

Matrix amplitudes and bootstrap inputs

With internal labels, the two-body amplitude is an intertwiner $S(\theta):V_a\otimes V_b\to V_c\otimes V_d$. A complete bootstrap must declare:

1. stable particle species, masses, charges, and conjugation;
2. representation projectors or invariant tensor structures;
3. Yang–Baxter consistency of alternative pairwise orders;
4. braiding unitarity and Hermitian analyticity;
5. crossing with an explicit charge-conjugation convention;
6. the physical sheet, allowed poles, residue normalization, and asymptotic bounds;
7. fusion equations for every accepted bound state; and
8. the remaining scalar factors, CDD choices, and completeness assumptions.

Symmetry and Yang–Baxter often determine ratios among tensor channels. Scalar factors remain to be fixed by analytic equations and spectrum data. A pole list alone is insufficient because anomalous thresholds can imitate bound-state singularities; the pole and fusion analysis supplies the residue and Coleman–Thun tests.

The integrability exact-data chain shows where spectrum and CDD assumptions enter. The exact and rigorous status comparison states the corresponding claim boundary: an exact meromorphic amplitude can be conditional on completeness and locality even when every displayed bootstrap equation is satisfied.

Algebraic solution versus local QFT

There are several increasingly strong statements:

• a function satisfies unitarity and crossing;
• a matrix amplitude also satisfies Yang–Baxter and symmetry;
• its poles and fusion rules close on a declared stable spectrum;
• finite-volume, form-factor, and ultraviolet checks match the proposed model; and
• local observables and the Hilbert-space theory are actually constructed with that scattering matrix.

The last statement is not automatic. A precise positive result is known for a bounded class with one neutral massive scalar species and no physical-strip bound-state poles. Lechner starts from regular bounded analytic factorizing scattering functions satisfying specified unitarity and crossing conditions, constructs wedge-local fields, and proves modular nuclearity. For general regular functions this yields nontrivial double-cone algebras above a finite splitting distance; the additional condition $S(0)=-1$ gives the result for arbitrarily small double cones. Within that class, the models are asymptotically complete and recover the prescribed scattering function Lechner 2008, Definitions 3.1 and 3.3, Theorem 5.6, p. 848; Theorem 5.8, p. 851; Proposition 6.2, p. 854; and Theorem 6.3, p. 855. The examples include scattering functions associated with sinh-Gordon and scaling Ising theory.

This is a rigorous existence result under its hypotheses, not a theorem that every algebraic solution, physical-strip pole pattern, matrix S matrix, or CDD deformation defines a local QFT. Bound-state poles and more general internal spaces require additional constructive machinery.

Completeness tests

No finite checklist proves completeness in every model, but independent tests can reveal an incomplete ansatz.

• Pole closure: every physical-strip pole is classified as a bound state, crossed image, or on-shell anomalous threshold; accepted bound states have consistent fusion amplitudes.
• Finite-volume check: Bethe–Yang levels agree at large circumference, with exponentially small wrapping corrections treated separately.
• Observable check: form factors satisfy exchange, crossing, and residue axioms, and spectral sums have a controlled tail in the regime quoted.
• Ultraviolet check: the thermodynamic Bethe ansatz approaches the expected effective central charge after bulk subtraction and convention matching.
• Existence check: a constructive theorem’s entire hypothesis set is verified when a rigorous local realization is claimed.

Agreement in the infrared and ultraviolet is strong model evidence but does not remove CDD freedom by logic alone, nor does it prove uniqueness among all local QFTs.

Common pitfalls

Calling minimality a consequence of unitarity. Minimality removes additional analytic structure by choice. CDD factors show that the functional equations alone do not do so.

Leaving the sheet implicit. A pole coordinate has no physical interpretation until the rapidity strip, sheet, and continuation convention are declared.

Equating an exact amplitude with a universal result. Exactness is internal to the specified integrable model. A nearby nonintegrable deformation can have production, widths, and branch cuts absent from the exact ansatz.

Exercises

1. Verify unitarity and crossing for $C_\alpha(\theta)$ and locate its physical-strip poles for $0<\alpha<\pi$.

Solution

Replacing $\theta$ by $-\theta$ exchanges numerator and denominator, so the product is one. Since $\sinh(i\pi-\theta)=\sinh\theta$, crossing is immediate. At $\theta=iu$, the denominator is $i(\sin u-\sin\alpha)$. It vanishes at $u=\alpha$ and $u=\pi-\alpha$, with coincidence at $\alpha=\pi/2$. Whether both apparent singularities are distinct physical-channel poles depends on cancellations and the complete amplitude.

2. Show that the sinh-Gordon fixture has no physical-strip pole for $0<B<1$, and explain why this check does not construct the theory.

Solution

At $\theta=iu$ with $0<u<\pi$, the denominator is $i[\sin u+\sin(\pi B)]$, which is nonzero because both sines are positive. This verifies one analytic property of the scattering function. It neither constructs local observable algebras nor proves asymptotic completeness; those conclusions require additional hypotheses such as those in the bounded Lechner construction.

References

• Castillejo, L., R. H. Dalitz, and F. J. Dyson. “Low’s Scattering Equation for the Charged and Neutral Scalar Theories.” Physical Review 101 (1956): 453–458. DOI.
• Lechner, Gandalf. “Construction of Quantum Field Theories with Factorizing S-Matrices.” Communications in Mathematical Physics 277 (2008): 821–860. DOI. Open PDF.
• Zamolodchikov, Alexander B., and Alexey B. Zamolodchikov. “Factorized S-Matrices in Two Dimensions as the Exact Solutions of Certain Relativistic Quantum Field Models.” Annals of Physics 120 (1979): 253–291. DOI.