HM08 — Grassmannian Quantisation

Field Value
Domain Positive Geometry / Scattering Amplitudes
System Massless scattering in N=4 SYM
Group GL(k) × GL(n−k) acting on Gr⁺(k, n)
H^k tier
ISA Hum (β = it/ℏ) — Grassmannian basis
Status Conjectural (conceptual framework)
Opcodes ORBIT · EMIT
Papers Paper 620

Physical system

There are two standard quantisation paradigms:

  1. Canonical quantisation (first quantisation): fixed particle number, Hilbert space ψ, operators acting on ℋ. Replace Poisson brackets with commutators. The ISA grading: H¹ (quantum phases, interference, no EMIT).

  2. Path-integral quantisation (second quantisation): variable particle number, Fock space ⊕_N ℋ^{⊗N}, field operators ψ̂(x) creating/destroying quanta. The ISA grading: H³ Feynman basis (EMIT + PROPAGATE + RENORM).

The amplituhedron proposes a third:

  1. Grassmannian quantisation: classical scattering data (n particles, helicities k, external momenta Z) maps to a positive region 𝒜_{n,k,L}(Z) ⊂ Gr⁺(k, k+4; L). The amplitude is the volume of this region. No Fock space, no Lagrangian, no virtual particles.

ISA interpretation

In ISA language, Grassmannian quantisation replaces the Fock-space tower ⊕_N ℋ^{⊗N} with the positive Grassmannian — a real manifold with boundary, not a complex vector space.

First and second quantisation emerge as limits:

  • First quantisation = L=0 (external data Z only; no loop variables)
  • Second quantisation = L > 0 (loop positivity constraints = virtual effects)

The Hum ISA opcode programme is:

  1. EMIT: specify the theory (n, k, coupling g, particle content)
  2. ORBIT(Gr⁺(k, k+4; L)): compute the volume

That is all. PROPAGATE and RENORM do not appear — they are Feynman-basis artefacts of the second-quantisation expansion.

The three quantisation paradigms as ISA programmes

Paradigm ISA basis Key opcodes Fock space? RENORM?
Canonical (1st) Origami/Motive ORBIT, TWIST No No
Feynman (2nd) Hum (Feynman) EMIT, PROPAGATE, RENORM Yes Yes
Grassmannian (3rd) Hum (Grassmannian) EMIT, ORBIT No No

The third paradigm has no standard name in the QFT literature. “Positive geometry quantisation” (Arkani-Hamed) and “amplituhedron quantisation” have been used informally.

The distinction between real and virtual particles

In the Feynman basis: external (real) particles are ORBIT entries; internal (virtual) particles are PROPAGATE steps. The distinction is fundamental — real particles are on-shell (p² = m²), virtual particles are off-shell.

In the Grassmannian basis: both real and virtual particles are columns in the (n+2L) × (k+4) matrix (Z_1, …, Z_n, D^{(1)}, …, D^{(L)}). Real particles = visible columns Z_i; virtual particles = hidden columns D^{(i)} subject to positivity. The real/virtual distinction is not fundamental — it is a labelling of which columns are “visible” (given as input data) vs “hidden” (integrated over).

This dissolves the on-shell/off-shell distinction that underlies the need for PROPAGATE and RENORM.

Connections

  • HM07 (Amplituhedron): HM07 is the concrete realisation; HM08 is the conceptual framework. HM07 gives the opcode count (1 vs ≥ 10); HM08 explains why (Fock space vs positive Grassmannian).
  • P01 (OPU): the Orbit Processing Unit (Paper 598) runs ORBIT on Gr(k,n) for molecular orbitals. Grassmannian quantisation = QFT running on the OPU. The two applications are: chemistry (orbital geometry) and scattering amplitudes.
  • RS01 (Rising Sea ISA, future): the Rising Sea ISA (Paper 621) shows that the β-plane fibration contains Grassmannian quantisation as the β=it/ℏ fibre. The “third paradigm” is not a new foundation but a specific fibre in the categorical fibration.

Open questions

  1. Gravity at H³: the m=6 generalised amplituhedron (Arkani-Hamed, post-2013) suggests graviton amplitudes live at H³ in the Grassmannian basis. If so, non-renormalisability of perturbative gravity is a Feynman-basis artefact at any loop order.

  2. Grassmannian Lamb shift: the Lamb shift in the Grassmannian basis would be a boundary-constrained ORBIT. What is the positive geometry? (x620f, open)

  3. ℤ/4ℤ closure: IMAGINE⁴ = MARK (Paper 621, x621d). Does Grassmannian quantisation extend to ℤ/8ℤ (Bott periodicity)? Is there an 8-periodic positive geometry?


Part of the ISA Zoo. Hum ISA reference: Paper 620.