• No results found

Subtle Points About Saddle Points

N/A
N/A
Protected

Academic year: 2021

Share "Subtle Points About Saddle Points"

Copied!
34
0
0

Loading.... (view fulltext now)

Full text

(1)

Subtle Points

About Saddle Points

in the S-Matrix Theory

Sebastian Mizera (IAS)

(2)

We’ll look at aspects of saddle points in Lorentzian scattering processes without counterparts in other signatures

I) worldlines II) worldsheets

(3)

One of the oldest open problems in QFT: crossing symmetry

Are the two scattering processes related by analytic continuation?

[Stückelberg, Gell-Mann, Goldberger, Thirring, …]

(on-shell)

(mom. cons.)

(4)

Partial progress in axiomatic QFT with a mass gap:

Some extensions to string field theory .

[Bros, Epstein, Glaser ‘65] [Bros ‘86]

[De Lacroix, Erbin, Sen ‘18]

(5)

Technical challenges:

• Massless particles

• Different number of in/out states

(e.g., exchanging a particle for an antiparticle)

• More than five external states

(where constraints from the space-time dimension first start to matter)

(6)

Conceptual challenge:

• Little to no physics understanding

(previous proofs hinge on the use of analytic completion theorems)

(7)

Historically abandoned; replaced with a much stronger mathematical assumption

The S-matrix theory should not be based on assumptions whose physical content we don’t understand!

crossing symmetry

maximal analyticity

contrast with physical assumptions:

causality, locality, unitarity

(8)

What’s the new strategy?

(9)

Revisit the question of crossing symmetry in perturbation theory Crossing symmetry absence of a certain class of singularities

[Witten]

internal particles on-shell (any number of loops)

aligned along two beams

(10)

Simplification coming with perturbation theory:

singularities wordline saddle points

algebraic problem

(11)

Contribution from a single worldline Feynman diagram:

Schwinger parameters

determinant of the Laplacian

worldline action

polynomial numerator (from the OPE’s)

(12)

The worldline action takes the form

Before applying saddle-point analysis (really, stratified Morse theory) we need to make the Green’s functions holomorphic

internal masses

Green’s function

(13)

Saddle point equations:

(in this talk we ignore boundary saddles)

Wordline action is special because of homogeneity under dilations:

(14)

Three important consequences:

• The action vanishes on the saddle points, .

• Integrating out the overall scale gives .

• equations on independent Schwinger parameters, leaving at least one constraint on the external kinematics

degree of divergence

(15)

For instance at 4-pt, in terms of the Mandelstam invariants and

(while Euclidean scattering doesn’t have any saddles at all)

Re(s) Re(t)

0 saddle points

1 saddle point

2 saddle points

Lorentzian kinematics

(16)

Physical meaning of intermediate particles becoming classical on-shell states (also known as anomalous thresholds)

Saddle point conditions equivalent to Landau equations

internal propagators on-shell

(17)

Causality/convergence at infinity is imposed by rotating the worldlines by an infinitesimal amount in complex directions:

The action acquires a small positive imaginary part

except at saddle points

(18)

Resolves the branch cuts

Starting point for general dispersion relations beyond EFT?

analyticity

doesn’t imply anything about analyticity away from the physical kinematics

s s

(19)

Analytic continuation of external energies within the complexified lightcone (say at 4-pt):

z

+1 1

(preserves on-shell conditions and mom. cons.)

lightcone coordinates

(20)

Every internal momentum can be decomposed as

Putting them on-shell implies .

Unknown if such anomalous thresholds exist in an arbitrary theory

1 2

3

4

(21)

For planar scattering amplitudes such singularities never appear:

analyticity when rotating the energies

propagators along the perimeter can never go on-shell!

p

±1

p

±1

f

e

> 0

z

±1

p

±2

g

e

> 0

z

±1

p

±2

cyclic ordering

(1234)

(22)

A B C D

!I II!

IV V

III!

B A

C D

C¯ A

B¯ D

B C¯

A¯ D

A C¯

B¯ D

B C¯

¯ D A

[details in hep-th/2104.12776]

Sequence of rotating the energies between crossing channels:

(23)

Composition of such moves proves crossing symmetry for planar amplitudes at every order in perturbation theory

with any masses, spins, multiplicity, …

(for , processes with consecutive in/out states in CPT-invariant theories)

(24)

Similar arguments extend to many non-planar Feynman diagrams, e.g.,

But, more generally, crossing symmetry remains an open problem!

(25)

How does it generalize to strings?

(Spoiler: we don’t know)

(26)

All the QFT anomalous thresholds should be present

for every mass level

(27)

Therefore, Lorentzian string scattering at any genus

must involve an infinite number of saddle points!

(28)

Before applying the saddle-point analysis, one needs to make

the Green’s functions holomorphic (say, for open strings)

(29)

Allows for “winding” of strings around each other

Tricky to interpret physically

[cf. Grinberg, Maldacena ‘20 in black hole physics]

(30)

Already non-trivial at genus zero

On each of the sheets counted by , the number of saddle points is .

worldsheet action (finite)

integers depending on the crossing channel and cyclic ordering

(31)

For example, 4-pt with Chan–Paton ordering (1234) in the (2,2) signature gives a single saddle

In the Lorentzian signature we have an infinite lattice of saddles

[Gross, Mende ‘87]

[Fairlie, Roberts ‘72]

N 1, 1

w12 w23

N0, 1 N1, 1 N2, 1

N 1,0 N0,0 N1,0 N2,0 N 1,1 N0,1 N1,1 N2,1

N 1,2 N0,2 N1,2 N2,2

N 1, 2 N0, 2 N1, 2 N2, 2

N 2, 1

N 2,0 N 2,1

N 2,2

N 2, 2

(32)

[details in hep-th/1910.11852 with Pokraka]

Re(s) Re(t)

1

0

2

0 1

0 2

0

0

w12

w23

0 0 0 0

0 0 +1 +1 +1

0 0 0 1 1

0 0 0 0 0

0 0 0 0 0

· · ·

· · ·

0

w12 w23

0 0 0 0

0 0 0 0 0

0 0 +1 0 0

0 0 0 0 0

0 0 0 0 0

0

w12 w23

0 0 0 0

0 0 0 0 0

0 +1 0 0 0

0 +1 1 0 0

0 +1 1 0 0

... ...

s-channel

u-channel t-channel

All the saddles resum to

resonances

(33)

Summary

• Singularities as worldline saddle points

• Crossing symmetry for planar amplitudes

• Challenges in generalizations to string theory

(34)

Thank you!

References

Related documents