de Sitter extremal surfaces, time contours, complexifications and pseudo-entropies

This paper investigates no-boundary de Sitter extremal surfaces and their associated pseudo-entropies by utilizing complex time contours to analyze both large and small subregions, revealing geometric equivalences and suggesting that de Sitter entropy inequalities can be encoded via analytic continuation of AdS counterparts.

The Gist

The Big Picture: A Universe with a "Ghostly" Mirror

Imagine our universe is like a giant, expanding balloon (this is de Sitter space, or dS). In the famous theory of holography, the idea is that everything happening inside this 3D balloon is actually a "shadow" or a projection of information living on the 2D surface of the balloon.

Usually, physicists study this using Anti-de Sitter (AdS) space, which is like a bowl. In a bowl, if you want to know how much two regions on the edge are "entangled" (connected by quantum secrets), you draw a smooth, real string (a surface) from the edge down into the bowl and back up. The length of that string tells you the amount of entanglement.

But our universe is a balloon, not a bowl. The rules are different. The "surface" of our universe is at the very end of time (the future). When physicists try to draw those same "entanglement strings" in our balloon universe, they run into a problem: the strings don't want to be real.

The Problem: The Strings Turn into Ghosts

In this paper, the author (K. Narayan) investigates what happens when we try to measure the "entanglement" between different patches of the universe's future edge.

1. The "Real" Strings Break: For big patches of the universe, you can draw a path that goes partly through real time and partly through "imaginary" time (a mathematical trick). It looks like a real shape in a diagram.
2. The "Ghost" Strings Appear: But for small patches, the math forces the string to become entirely "imaginary." It doesn't exist in our normal spacetime; it exists in a parallel, mathematical "ghost world" (an auxiliary AdS space).

The author calls the result of these measurements "Pseudo-Entropy." Think of it like a "ghostly shadow" of entanglement. It's not a normal number (it can be complex, like $3 + 4i$), but it still holds deep meaning about how the universe is connected.

The Key Discovery: Different Paths, Same Destination

The most surprising part of the paper is about Time Contours.

Imagine you are trying to get from City A to City B.

• Route 1 (The Scenic Route): You drive through real mountains and valleys (Real time + Euclidean time).
• Route 2 (The Warp Drive): You take a shortcut through a wormhole in a parallel dimension (Complex time).

Usually, these routes are totally different. But the author discovers that in this specific universe, Route 1 and Route 2 are actually the same trip.

If you look at the "map" of time (which is a complex plane, like a map with North, South, East, and West), you can wiggle Route 1 around until it looks exactly like Route 2 without hitting any roadblocks. Because you can morph one into the other, they give the exact same answer for the "ghostly entropy."

The Analogy: It's like having two different recipes for a cake. One uses real eggs, the other uses "imaginary" eggs. But if you can mathematically prove that the "imaginary" eggs are just a different way of describing the real ones, then both recipes make the exact same cake.

The "Inflation" Trick: Small Becomes Big

The paper also explores a cool geometric trick involving light rays.

Imagine you are standing at the North Pole of a static, frozen version of the universe. You shine a flashlight forward.

• If you shine it on a tiny spot right next to you, the beam travels across the universe and hits the future edge (the horizon).
• Because the universe is expanding so fast, that tiny spot you shone on gets stretched out massively by the time it hits the future.

The author shows that a tiny, regulated patch near the North Pole is mathematically equivalent to a giant, massive patch at the very end of the universe. This allows them to calculate the total entropy of the universe (the "size" of the whole balloon) by just looking at the tiny patch near the pole.

Why Does This Matter?

1. It Unifies Two Worlds: It suggests that our weird, expanding universe (dS) is secretly just a "rotated" version of the nice, stable bowl universe (AdS). If you turn the universe sideways in the math, the weird "ghost" entanglement becomes the normal "real" entanglement we understand.
2. It Solves the "Small Patch" Problem: It explains how to calculate quantum connections for small parts of the universe, which was previously a mystery. The answer lies in these "ghost" surfaces that live in complex time.
3. It Redefines Entropy: It tells us that in our universe, the "entropy" (disorder) isn't just a number; it's a complex, multi-dimensional shape that connects the beginning, middle, and end of time in ways we are just starting to understand.

Summary in One Sentence

The paper discovers that the quantum "glue" holding our universe together can be measured using invisible, ghostly paths in a parallel mathematical world, and that these ghostly paths are actually just different views of the same real-world geometry, revealing that our expanding universe is deeply connected to the stable universes we already know how to study.

Technical Summary

1. Problem Statement

The paper addresses the challenge of defining holographic entanglement entropy in de Sitter (dS) space, specifically within the context of the dS/CFT correspondence. Unlike Anti-de Sitter (AdS) space, where the Ryu-Takayanagi (RT) and Hubeny-Rangamani-Takayanagi (HRT) prescriptions yield real-valued areas for extremal surfaces anchored to the boundary, dS space presents unique difficulties:

• Non-unitary Duals: The dual CFT at future infinity ($I^+$) is expected to be non-unitary (ghost-like), implying that the reduced density matrix is actually a transition matrix ($\rho_A \sim |f\rangle\langle i|$). Consequently, the resulting entropy is a pseudo-entropy, which is generally complex-valued.
• Absence of Real Turning Points: In global no-boundary dS, extremal surfaces anchored at $I^+$ do not possess real "turning points" where they return to the boundary. Instead, they often require complexification or become timelike surfaces.
• Ambiguity in Surface Selection: For generic subregions, it is unclear which extremal surfaces (real, timelike, or complex) correctly encode the boundary pseudo-entropy, especially as subregion sizes vary from maximal to small.

2. Methodology

The author employs a combination of geometric analysis, analytic continuation, and complex contour integration to study extremal surfaces in $dS_3$ and higher dimensions ($dS_{d+1}$).

• Analytic Continuation: The study utilizes the mapping $L_{AdS} \to i l_{dS}$ to relate dS extremal surfaces to their AdS counterparts. This allows the derivation of dS surface profiles and areas from known AdS solutions.
• Complex Time Contours: Extremal surface area integrals are defined via contours in the complex time plane ($r$ or $\tau$). The paper analyzes how these contours deform around singularities (poles) and how different contours (e.g., real timelike paths vs. imaginary time paths) relate to one another.
• Replica Geometry: The paper examines bulk replica geometries with conical deficits at $I^+$ to verify that the $n \to 1$ limit of the replica partition function yields the extremal surface areas.
• Light Ray Mapping: In the static patch, the author uses light rays to map subregions on constant time slices to subregions at the future boundary ($I^+$), establishing a geometric correspondence between static patch observers and the global boundary.

3. Key Contributions and Results

A. Classification of Extremal Surfaces by Subregion Size

The paper distinguishes between two regimes based on the size of the boundary subregion at $I^+$:

1. Large Subregions:

   • Timelike + Euclidean Surfaces: These exist as connected surfaces. They consist of a timelike segment in the Lorentzian region (purely imaginary area) joined to a spatial segment wrapping a Euclidean hemisphere.
   • Complex Surfaces: Complex extremal surfaces also exist for large subregions.
   • Result: Both types yield identical areas. The total area contains a divergent imaginary term (cutoff dependent) and a finite real term equal to half the de Sitter entropy ($S_{dS}/2$).

2. Small Subregions:

   • Breakdown of Real Surfaces: As the subregion size decreases, the "tilt" of the timelike+Euclidean surface increases until the Euclidean hemisphere component ceases to exist as a real spatial surface.
   • Existence of Complex Surfaces: Only complex extremal surfaces (living in an auxiliary AdS space via analytic continuation) exist for sufficiently small subregions.
   • Result: The area formula remains structurally identical to the large subregion case, but the real de Sitter entropy term arises non-trivially from the complex time contour integration, with no direct geometric hemisphere interpretation.

B. Equivalence of Time Contours and Replica Geometries

A central finding is the equivalence of distinct extremal surfaces for the same boundary subregion:

• Contour Deformation: The time contours for the "timelike+Euclidean" surface (real path $r \in [0, l] \cup [l, R_c]$) and the "complex" surface (imaginary path $r = i\rho$) can be deformed into one another in the complex time plane without obstruction (skirting the pole at $r=l$).
• Implication: These surfaces are physically equivalent for the purpose of calculating pseudo-entropy. This implies an equivalence between the corresponding complex replica geometries (one Lorentzian/Euclidean hybrid, the other purely complex/hyperbolic), suggesting they are related by complex coordinate transformations.

C. Entropy Inequalities and Subregion Duality

• Analytic Continuation of AdS Inequalities: For multiple small subregions, the paper demonstrates that dS pseudo-entropy inequalities (e.g., mutual information, tripartite information) encode the familiar AdS entropy inequalities via analytic continuation ($il \to L$).
• Phase Transitions: The "disentangling transition" (switching from connected to disconnected surfaces) observed in AdS is encoded in dS through these complex surfaces.
• Subregion Duality: Geometric subregion duality (mapping a boundary region to a bulk wedge) is not manifest directly in dS geometry but is encoded via the analytic continuation to the auxiliary AdS space.

D. Static Patch and Left-Right Entanglement

• Light Ray Mapping: The author establishes a mapping between small subregions on the constant time slice of the static patch (near the North/South poles) and large subregions at $I^+$ via light rays. The relation is an inversion: $r_{static} \sim l^2 / r_{future}$.
• Left-Right Extremal Surfaces: By considering extremal surfaces connecting the regulated North and South static patches (analogous to the Hartman-Maldacena surface in black holes), the paper derives an area that equals the total de Sitter entropy in the limit where the subregions approach the poles.
• Interpretation: This suggests that the de Sitter entropy can be interpreted as the entanglement entropy between the left and right static patch observers, emerging from a spatial extremal surface rather than a complex one.

4. Significance

• Resolution of Ambiguity: The paper resolves the ambiguity of which extremal surface to choose in dS/CFT by demonstrating that multiple surfaces (real and complex) are equivalent via contour deformation, provided they share the same endpoints.
• Unified Picture of Pseudo-Entropy: It provides a coherent framework for understanding complex-valued pseudo-entropies, showing that the "real" de Sitter entropy term emerges naturally from complex time contours even when no real Euclidean hemisphere exists.
• Connection to Timelike Entanglement: The work bridges the gap between dS holography and the study of timelike entanglement, suggesting that the exotic features of dS/CFT (non-unitarity, complex areas) are consistent with known structures in timelike entanglement studies.
• Static Patch Entropy: The derivation of de Sitter entropy from a spatial left-right extremal surface in the static patch offers a new perspective on the origin of dS entropy, potentially linking it to entanglement between causally disconnected regions without relying solely on the complex $I^+$ boundary.

In summary, Narayan's work advances the understanding of holography in de Sitter space by rigorously defining extremal surfaces via complex time contours, establishing the equivalence of different geometric saddles, and demonstrating how standard AdS entanglement structures are encoded in dS through analytic continuation.