Towards a Manifestly Causal Approach to Particle Scattering

Robert Dickinson, Jeff Forshaw, Ross Jenkinson, Peter Millington · 2025

Read Towards a Manifestly Causal Approach to Particle Scattering by Robert Dickinson, Jeff Forshaw, Ross Jenkinson, Peter Millington (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2502-18551.txt.

Introduction

Robert Dickinson, Jeff Forshaw, Ross Jenkinson, and Peter Millington propose a new way to compute particle scattering that keeps cause and effect visible. Their paper appeared in 2025. Standard methods work with amplitudes and square them at the end. The new approach works directly at the level of probabilities, summing over unseen final states from the start. Delayed signals, which respect causality, appear on their own.

The authors draw diagrams for the probability terms, much as Feynman diagrams picture amplitude terms. They conjecture a set of rules that generates every such diagram for a broad class of processes. Fixed-order examples confirm the rules. They hope the built-in causality will clarify why certain infinite terms cancel in gauge theories, a long-standing puzzle in scattering calculations.

Catalog record

CREATORRobert Dickinson, Jeff Forshaw, Ross Jenkinson, Peter Millington
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE88 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2502-18551

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Towards a Manifestly Causal Approach to Particle Scattering Robert Dickinson, Jeff Forshaw, Ross Jenkinson, Peter Millington Abstract We introduce a new, probability-level approach to calculations in scalar field particle scattering. The approach involves the implicit summation over final states, which makes causality manifest since retarded propagators emerge naturally. Novel diagrams represent algebraic terms at the probability level, akin to Feynman diagrams at the amplitude level. We present a list of rules which generate all probability-level diagrams for particle scattering processes in which one is fully inclusive over final states that contain no initial-state particles. The inclusivity and causal structure of this formalism may offer insights into the cancellation of infrared divergences if applied to gauge theory calculations. I Introduction In relativistic quantum field theories (QFTs), causality is encoded in the vanishing of the commutator (or anti-commutator) of field operators. However, scattering calculations are usually performed in a manner in which causality is not manifest, due to the ubiquity of the Feynman propagator as opposed to the retarded propagator (which is manifestly causal). Furthermore, Bogoliubov’s Condition of Causality Bogolyubov and Shirkov 1959 shows that causality is manifest and meaningful only at the probability level. The S -matrix does not directly demonstrate causality because it represents only the transition amplitude, not the full probabilistic outcome of a process. This distinction underscores the importance of observable probabilities over amplitudes when discussing physical principles like causality. With this in mind, a manifestly causal, probability-level formalism for calculations in QFT was developed Dickinson et al. 2014 ; Dickinson et al. 2016 ; Dickinson et al. 2017a ; Dickinson et al. 2017b by applying a generalisation of the Baker-Campbell-Hausdroff lemma to the transition probability, naturally expressed in terms of commutators of fields. This method has been used to study the Fermi two-atom problem Dickinson et al. 2017a and the Unruh effect Dickinson et al. 2025 . In this paper, we apply the formalism to particle scattering processes for inclusive final states where we demand only that there are no initial-state particles in the final state. Since all of the expressions are at the probability level, they are algebraically more complicated than in the usual amplitude-level approach. In this initial study, we start with simpler scalar field theories, which offer an instructive analogy to gauge theories such as Quantum Electrodynamics (QED) and Quantum Chromodynamics (QCD). The result is a new probability-level, diagrammatic method for calculating scattering probabilities in which the retarded propagator plays a key role, making causality explicit. The appearance of retarded propagators may offer a novel window on infrared divergences, which arise in part due to low-energy (soft) massless particles having arbitrarily large wavelengths that have effects over infinite separations. Retarded propagators are zero for spacelike separations and carry a distinct analytic structure compared to the Feynman propagator, and this approach may help quell these infinite-distance contributions. The Bloch-Nordsieck Bloch and Nordsieck 1937 and the Kinoshita-Lee-Nauenberg (KLN) Kinoshita 1962 ; Lee and Nauenberg 1964 theorems ensure that, while individual amplitudes may diverge, physical observables (like cross sections and decay rates) remain finite when real and virtual soft emissions are properly summed. Our formalism allows one to sum implicitly over all relevant final states from the beginning of the calculation and thereby explore the infrared behaviour from a new angle. An example of a typical gauge theory process of interest is electron-positron ( e^{+}e^{-} ) pair annihilation to a quark-antiquark ( q\overline{q} ) pair, mediated by a photon ( \gamma ), with gluon ( g ) corrections to the final state. The Feynman diagram for this process is pictured on the right of Fig. 1 . A scalar analogue of this process is \psi\psi\rightarrow\chi\rightarrow\phi\phi with h corrections to the final state, where \psi,\chi,\phi, and h are all scalar fields, as pictured on the left of Fig. 1 . In this paper, we first consider the decay process \chi\rightarrow X (Sections IV and V ), where X represents a general final state, before turning to the annihilation process \psi\psi\rightarrow X (Section VI ). For both processes, we are fully inclusive over final states, except for demanding that they contain no initial-state particles. This first study allows us to demonstrate how our new formalism plays out in a simple example, without the technical complications of spinor and gauge structure. In Section II , we introduce the probability-level formalism used throughout the paper. Section III defines a new diagrammatic method and outlines the general rules for generating the set of diagrams relevant to any scalar field scattering process for which the final state contains anything except initial-state particles. For those interested in the calculational details, Sections IV - VI contain algebraic derivations of the complete sets of relevant diagrams without using the rules. We observe that distinct final states are intrinsically summed over and are not separable in our results, and the retarded self-energy emerges naturally. The calculations look complicated in the intermediate steps, but they reduce considerably. This suggests the existence of a more fundamental set of rules. Section VII concludes. Throughout this paper, we adopt natural units c=\hbar=1 and the ‘mostly-minus’ metric signature (+---) . Figure 1 : Traditional, amplitude-level diagrams for the processes of interest. Right: The gauge theory process e^{-}e^{+}\rightarrow\gamma\rightarrow q\overline{q} , with gluon ( g ) corrections. Left: An analogous toy-model in which \psi,\chi,\phi and h are all real scalar fields. II Probability-Level Formulation Suppose that a system is initially ( t\rightarrow-\infty ) described by a density operator \rho_{0}=\ket{i}\bra{i} and that a measurement outcome is described by an effect operator E . In general, E is an element of a Positive Operator-Valued Measure, but in the case of an exclusive final state, \ket{f} , it becomes a projection operator, \displaystyle E=\ket{f}\bra{f}. (1) The probability of the measurement outcome, \mathbb{P} , is then given by \mathbb{P}\ =\ \tr(E \rho_t)\,, (2) where \rho_{t}\ \equiv\ U\,\rho_{0}\,U^{\dagger} (3) is the density operator at time t and U=\mathrm{T}\left\{\exp\left(\frac{1}{i}\int^{\infty}_{-\infty}\mathrm{d}t^{\prime}\>\mathcal{H}_{\text{int}}(t^{\prime})\right)\right\} (4) is the unitary evolution operator ( \mathrm{T} indicates time ordering) for a given interaction Hamiltonian, \mathcal{H}_{\text{int}}(t^{\prime}) . From Eq. ( 2 ), the probability of the measurement outcome can be written as \mathbb{P}\ =\ \Braket{i|U^{\dagger}\>E\>U|i}. (5) Using Eq. ( 1 ), \mathbb{P}\ =\ \bra{i}U^{\dagger}\ket{f}\bra{f}U\ket{i}\ =\ \big\lvert\bra{f}U\ket{i}\big\rvert^{2}\,. (6) This is the usual expression for the probability of an initial state, \ket{i} , evolving into a final state, \ket{f} . Rather than calculating the amplitude and squaring, we can instead work directly with the expression in Eq. ( 5 ). We do so with a generalization of the Baker-Campbell-Hausdorff lemma as in Franson and Donegan 2002 ; Dickinson et al. 2014 ; Cliche and Kempf 2010 . One commutes the effect operator, E , through the evolution operator, U , resulting in \displaystyle\mathbb{P} \displaystyle=\sum_{j=0}^{\infty}\int^{\infty}_{-\infty}\mathrm{d}t^{\prime}_{1}\mathrm{d}t^{\prime}_{2}\dots\mathrm{d}t^{\prime}_{j}\Theta_{12\dots j}\Braket{i|\mathcal{F}_{j}|i}, (7) where \displaystyle\Theta_{ijk\dots} \displaystyle\equiv\begin{cases}1,&\text{if }t^{\prime}_{i}>t^{\prime}_{j}>t^{\prime}_{k}\dots\\ 0,&\text{otherwise}\end{cases}, (8) \displaystyle\mathcal{F}_{0} \displaystyle=E, (9) and \displaystyle\mathcal{F}_{j} \displaystyle=\frac{1}{i}\left[\mathcal{F}_{j-1},\mathcal{H}_{\text{int}}(t^{\prime}_{j})\right]. (10) In general, E may be written as the sum of the effect operators of each final state, which themselves are products of Hermitian operators in different Hilbert spaces, i.e., E=\sum_{\kappa}\prod_{i}E^{\mathscr{H}_{i}}_{(\kappa)}, (11) where the superscripts indicate the Hilbert spaces of each operator, and \{\kappa\} is the set of final states. If all final states of a particular Hilbert space are summed over, then the effect operator in that Hilbert space is the identity operator,

Read the complete text