Null Infinity

How can we talk about the "end" of the universe or the total energy radiated by a system over all time? In a relativistic universe where space and time are dynamic, simply going "infinitely far away" is not a well-defined concept. This ambiguity poses a fundamental problem in general relativity, hindering our ability to make global statements about physical systems like colliding black holes. The solution lies not in reaching infinity, but in bringing infinity to a finite, analyzable boundary.

This article introduces null infinity, the elegant geometric construction that serves as the ultimate boundary of spacetime. We will explore how this concept provides a rigorous stage for the cosmic drama to unfold. The following chapters will delve into its core principles and its far-reaching applications. First, "Principles and Mechanisms," will introduce the mathematical tools, like conformal compactification and Penrose diagrams, used to define null infinity and examine its structure as a celestial bulletin board for cosmic "news." Following that, "Applications and Interdisciplinary Connections," will demonstrate how null infinity offers a powerful lens to understand causality, horizons, radiated energy, and a fascinating set of infinite symmetries that connect to the deepest puzzles in modern physics.

Where does our universe end? If you point your telescope in some direction, you can see galaxies billions of light-years away. But where does the spacetime itself stop? This is a much trickier question than asking where the "stuff" in the universe runs out. It's a question about the very fabric of reality. In physics, especially when we want to ask questions about the whole system, like "How much total energy was radiated by that black hole merger?", we need a way to talk about "infinity" precisely. We can't just send our detector "infinitely far away" and wait "infinitely long".

The brilliant insight, championed by Roger Penrose, was to not try to go to infinity, but to bring infinity to us. The technique is called ​​conformal compactification​​, and a wonderful analogy is a photographer's fisheye lens. A fisheye lens can capture an entire 180-degree vista—from your feet to the horizon in every direction—and map it onto a single, finite, circular photograph. The things far away look squished and distorted near the edge of the photo, but their relative positions are preserved. Most importantly, if two people in the distance were walking towards each other, you'd see their images in the photo moving towards each other. The causal relationships are maintained.

Conformal compactification does something similar for spacetime. We invent a new set of mathematical "coordinates" that take the infinite expanse of spacetime and map it onto a finite region. Let's see how this works for the simplest case: the flat, empty spacetime of special relativity, called Minkowski space. We can describe any event by its time $t$ and its distance $r$ from an origin. The paths of light rays moving radially outwards or inwards are described by constant values of $u = t-r$ and $v = t+r$, respectively. These are called ​​null coordinates​​, and they stretch from $-\infty$ to $+\infty$.

To tame these infinities, we pass them through a function that maps an infinite range to a finite one, the arctangent function. We define new coordinates, say $\tilde{u} = \arctan(u)$ and $\tilde{v} = \arctan(v)$. Now, as $u$ and $v$ go from $-\infty$ to $+\infty$, our new coordinates $\tilde{u}$ and $\tilde{v}$ only wander between $-\frac{\pi}{2}$ and $\frac{\pi}{2}$. We have successfully "brought infinity in to a finite distance".

The mathematical "lens" we used is called a ​​conformal factor​​, often denoted by the Greek letter Omega, $\Omega$. For Minkowski space, this factor turns out to be a tidy expression, $\Omega = \cos(T) + \cos(\chi)$, where $T$ and $\chi$ are coordinates related to our $\tilde{u}$ and $\tilde{v}$. This factor rescales the spacetime metric, squishing the infinite volume into a finite one but—and this is the genius of it—it does so in a way that preserves the paths of light rays. The angle between two intersecting light rays in the real spacetime is the same as the angle between their images in our new, compact picture. Causal structure, the bedrock of physics, is perfectly preserved. The resulting map is what we call a ​​Penrose diagram​​.

Now that we have our "photograph" of the entire spacetime, what does the edge look like? This edge is the boundary of spacetime. It's not all the same; different parts of the edge correspond to different kinds of infinity.

• ​​Timelike Infinity​​: Imagine you are an observer just floating in space, not moving very fast. Your worldline is "timelike". As you live on forever, your path on the Penrose diagram ends at a single point called ​​future timelike infinity​​, or $i^+$. It’s the ultimate destination for all slow-moving observers. Similarly, the common origin point in the infinite past is ​​past timelike infinity​​, $i^-$.

• ​​Spacelike Infinity ($i^0$)​​: This is the point representing being infinitely far away in space, at a finite time. It's the destination for anything that could hypothetically travel faster than light (which, of course, nothing can).

• ​​Null Infinity​​: This is the most interesting part. Where do light rays and gravitational waves go? They travel along null paths. On our diagram, all light rays that travel outwards forever end up on a boundary called ​​future null infinity​​, or $\mathscr{I}^+$ (pronounced "scri-plus"). All incoming light rays originate from ​​past null infinity​​, or $\mathscr{I}^-$ ("scri-minus"). These are not points, but entire surfaces. For our 4D spacetime, $\mathscr{I}^+$ has the structure of a cylinder with the topology of a sphere ($S^2$) times the real line ($\mathbb{R}$). You can think of it as a celestial sphere at the end of time, with "time" on this boundary marked by the retarded time coordinate $u$.

A key feature of this boundary is that it is itself a ​​null surface​​. This means that the boundary itself behaves like a path of light. Light rays don't crash into $\mathscr{I}^+$; they gracefully become part of it, their worldlines blending into the null generators of the surface. This is a subtle but profound geometric point: the "end" of spacetime for light is a place that shares the character of light itself.

These abstract ideas become wonderfully concrete when we look at a black hole. In the Penrose diagram for a Schwarzschild black hole (derived from the Kruskal extension), we see these boundaries clearly. The region outside the black hole is connected to $\mathscr{I}^-$, where light from the distant past comes from, and $\mathscr{I}^+$, where escaping light goes to. The event horizon is precisely the line that separates the region from which light can escape to $\mathscr{I}^+$ from the region where it is trapped and must hit the singularity.

So, we have this beautiful boundary, this celestial sphere at the end of time. What is it good for? It turns out to be the ultimate bulletin board for the universe. Any event that radiates energy—a supernova, a binary star system, a black hole collision—sends out gravitational waves and light that travel across the cosmos and eventually arrive at $\mathscr{I}^+$.

This is the key to solving a very old problem: how do you define the total energy of a gravitating system? For a static, lonely star, you could just go far away and measure its gravitational pull. This gives the ​​ADM mass​​. But what if the system is dynamic, like two black holes spiraling into each other? They are furiously shedding energy in the form of gravitational waves. The total mass must be decreasing.

The ADM mass, defined at spacelike infinity ($i^0$), is constant and can't capture this mass loss. To do that, we must measure the energy where the escaping radiation actually goes. And that place is future null infinity. Hermann Bondi and his collaborators developed a framework to do just this. The ​​Bondi mass​​, $M_B(u)$, is the mass-energy of the system measured on a spherical "cut" of $\mathscr{I}^+$ at a particular moment of retarded time $u$. As $u$ increases, we see the mass change. The rate of change, $-\frac{dM_B}{du}$, precisely equals the power being radiated away in gravitational waves.

But what is this radiation when it arrives at $\mathscr{I}^+$? It's "news" from the source. The ​​Bondi news function​​, $N(u, \theta, \phi)$, is the variable that quantifies this. Imagine the celestial sphere at $\mathscr{I}^+$ as a perfectly smooth sphere in quiet times. When a gravitational wave passes through, it distorts the sphere, creating a "shear". The news function is nothing more than the time rate of change of this shear. If the shear isn't changing, the news is zero, and no energy is being radiated. If the news is non-zero, it means ripples are actively passing through, and the Bondi mass will decrease. $\mathscr{I}^+$ acts as the ultimate detector screen, and the news function is the signal written upon it.

The picture gets even more elegant when we look closer at the structure of the gravitational field as it arrives at $\mathscr{I}^+$. The field doesn't just arrive as a single monolithic block. It "peels" away in layers of decreasing strength, a phenomenon known as the ​​Peeling Theorem​​. If we track the different components of the gravitational field (specifically, the five complex ​​Weyl scalars​​, $\Psi_k$, in the Newman-Penrose formalism) as we move away from a source along a light ray, we find a beautiful hierarchy:

• $\Psi_0$ falls off fastest, like $1/r^5$. It's associated with incoming radiation.
• $\Psi_1$ falls off like $1/r^4$. It's related to the source's momentum.
• $\Psi_2$ falls off like $1/r^3$. This is the "Coulomb-like" part, the piece related to the total mass of the source. It's the gravitational analogue of the static electric field from a charge.
• $\Psi_3$ falls off like $1/r^2$.
• $\Psi_4$ falls off slowest of all, like $1/r$. This is the prize. This is the purely transverse, outgoing gravitational radiation—the "news". It must fall off as $1/r$ because the energy flux is proportional to the amplitude squared ($\sim 1/r^2$), and when integrated over the area of a sphere ($\sim r^2$), this gives a total power that is constant, independent of distance.

This peeling property is a profound statement about the structure of general relativity. It allows us to unambiguously separate the long-range, static "charge" part of gravity from the dynamic, information-carrying "radiation" part. The radiation is the outermost, most weakly interacting layer of the gravitational field peeling off the source.

This brings us to one of the deepest questions in general relativity. If $\mathscr{I}^+$ is the ultimate screen upon which the universe's story is projected, what can appear on it? We know black holes have singularities at their centers, points of infinite density and curvature where physics breaks down. But they are hidden behind an event horizon. In the language of null infinity, an event horizon is the boundary of the region from which no causal signal can ever reach $\mathscr{I}^+$. The singularity is censored from our view.

But could a singularity exist without a horizon? Could a point of infinite curvature be "naked", visible to the outside world? A ​​naked singularity​​ would be an object from which a light ray could escape and travel all the way to an observer at $\mathscr{I}^+$. On a Penrose diagram, this would correspond to a null line connecting the jagged line of the singularity directly to the boundary of $\mathscr{I}^+$. The ​​Weak Cosmic Censorship Conjecture​​ is the profound, and as yet unproven, idea that Nature forbids this. It conjectures that any singularity formed from the gravitational collapse of realistic matter will always be clothed by an event horizon. In essence, it says that the universe politely shields the observers at $\mathscr{I}^+$ from witnessing the breakdown of its own laws.

The story of null infinity does not end there. It has a final, stunning chapter that connects to the deepest puzzles of quantum gravity. In physics, symmetries are paramount. Symmetries of spacetime lead to conservation laws. At null infinity, we have the expected symmetries of special relativity (the Lorentz group), but in the 1960s, Bondi, van der Burg, Metzner, and Sachs (BMS) discovered an infinite number of additional symmetries called ​​supertranslations​​. You can think of these as an independent time shift at every single point on the celestial sphere at $\mathscr{I}^+$.

Every one of these infinite symmetries must have an associated conserved charge. For decades, this was a mathematical curiosity. But recently, it has been placed at the center of a potential resolution to the ​​black hole information paradox​​. The paradox is this: if you make a black hole out of a book, and the black hole then evaporates away into a featureless bath of Hawking radiation, where did the information in the book go? Is it destroyed, violating a core tenet of quantum mechanics?

The "soft hair" proposal suggests that the information is never lost. When the matter forming the book collapses into a black hole, it imprints its pattern onto the event horizon in the form of these infinite BMS charges. This imprint is the "soft hair" of the black hole. Then, as the black hole evaporates, the outgoing Hawking radiation is not perfectly random; it is subtly correlated in a way that carries this information back out to future null infinity, where it can be read from the final state's BMS charges.

In this picture, null infinity is the ultimate accountant's ledger. The total BMS charge of the matter that falls in at $\mathscr{I}^-$ must exactly equal the total BMS charge of all the radiation that comes out at $\mathscr{I}^+$. Information is conserved. The boundary of spacetime, once a mere mathematical tool for taming infinities, has become a candidate for the physical repository of all information in the universe. It is a testament to the power of a good idea that this elegant geometric construction now sits at the frontier of our quest to unite gravity and the quantum world.

After our journey through the intricate machinery of general relativity, you might be left with a feeling of... isolation. Spacetime is vast, curved, and dynamic. How can we ever get a complete, objective picture of anything? If we are at different points in space, we see different things. If we move differently, our clocks tick at different rates. Is there any absolute stage on which the grand cosmic drama unfolds? Remarkably, the answer is yes, and its name is null infinity.

Null infinity, or $\mathcal{I}$ ("scri," for script 'I'), is not a place you can travel to. It is a boundary to spacetime, the ultimate destination for all rays of light. Think of it as a gigantic, all-encompassing celestial sphere at the end of time. It is the final screen onto which all information that can ever escape a physical system is projected. By studying the structure of this boundary and the patterns written upon it, we gain a powerful, unified perspective on the physics of the universe. The applications of this concept are not just theoretical curiosities; they provide profound answers to fundamental questions about causality, energy, and even the nature of symmetries themselves.

One of the first and most profound uses of null infinity is to draw a definitive map of causality. It allows us to state with mathematical precision what is, and is not, observable.

Let's begin with a surprising example that doesn't even involve gravity. Imagine an astronaut, let's call her Rindy, in flat, empty Minkowski space. She fires her rockets in such a way that she maintains a constant proper acceleration forever. From her perspective, a strange thing happens: a part of the universe becomes forever inaccessible. There is a boundary, a horizon, behind which events can occur that she will never see, no matter how long she waits. Light signals from that region can never catch up to her accelerating ship. The amazing thing is that the concept of null infinity allows us to precisely identify this boundary. The light ray that just barely fails to reach her defines her horizon, and this critical ray has a specific, identifiable endpoint on future null infinity, $\mathcal{I}^+$. This "Rindler horizon" shows that the notion of a 'point of no return' is fundamentally tied to the observer's trajectory, and null infinity provides the framework to analyze it.

Now, let's turn gravity back on. When a massive star runs out of fuel, it collapses under its own weight, and if it's massive enough, it creates a much more permanent and absolute kind of horizon: an event horizon. Null infinity gives us the most elegant and powerful definition of a black hole. A black hole is, quite simply, the region of spacetime that cannot send signals to future null infinity. It is the set of all events that are not in the "causal past" of $\mathcal{I}^+$, which we denote as $J^-(\mathcal{I}^+)$. In plain English, a black hole is the ultimate cosmic prison: it's the place from which no message can ever be posted on the great cosmic bulletin board at the end of time.

This definition is beautifully illustrated by Penrose diagrams, which are themselves constructed using the idea of null infinity. On such a diagram, the fates of different objects are clear. A photon emitted from just outside the black hole escapes its pull and travels along a straight 45-degree line until it reaches $\mathcal{I}^+$. A courageous (or unlucky) astronaut who falls into the black hole has their worldline end inexorably at the singularity. But what about an astronaut who successfully escapes the black hole's gravity and travels away forever? Her worldline does not end on null infinity. Instead, it terminates at a different point called future timelike infinity, $i^+$, the final destination for all massive observers who live forever. Null infinity is a special boundary reserved for massless particles, like the photons that carry information across the cosmos.

So, we have this giant screen at the edge of the universe. What's playing? The answer is: everything. Every flash of light, every gravitational tremor, every dynamic event in the history of the universe leaves its signature on null infinity.

Let's start with a simple firefly blinking in empty space. The light from that flash expands outwards as a sphere. When this sphere of light arrives at $\mathcal{I}^+$, it paints a kind of picture. From the pattern of arrival times—the retarded time $u$ as a function of direction on the celestial sphere—we can, in principle, reconstruct exactly when and where the firefly blinked. Null infinity acts as a perfect, information-preserving celestial recorder.

The real showstoppers, however, are not flashes of light but ripples in the fabric of spacetime itself—gravitational waves. When two black holes spiral into each other and merge, they send out powerful waves of gravitational radiation. This isn't matter or light traveling through spacetime; it is spacetime itself oscillating. The "news" of this cataclysmic event is encoded in a quantity at null infinity called the ​​Bondi news tensor​​, often represented by a complex function $N(u, \theta, \phi)$. The astonishing connection is that the tidal forces of the wave, the physical stretching and squeezing that an instrument like LIGO actually measures, are directly related to how fast this news function is changing with time, $\partial N / \partial u$. The broadcast is the derivative of the news!

This news carries energy. Just as a broadcasting radio antenna radiates electromagnetic energy, an accelerating, asymmetric system radiates gravitational energy. This radiated energy is lost from the system forever. General relativity provides a precise way to measure the total mass-energy of an isolated system as seen from infinitely far away: the ​​Bondi mass​​. As the system radiates gravitational waves, its Bondi mass decreases. The rate of mass loss is directly proportional to the squared magnitude of the Bondi news.

The physical meaning of this is beautifully illustrated by considering a black hole that is accreting matter. Imagine a spherically symmetric rain of dust falling into a black hole. The mass of the central object is certainly increasing. But is any news being broadcast to the distant universe? No. The flow of energy is purely inward. Consequently, an observer at future null infinity sees the Bondi mass remain perfectly constant. No energy has been radiated away, so the total energy of the isolated system has not changed. This provides a crucial distinction between the locally measured mass and the total, globally defined energy.

This framework allows us to analyze the energetics of any dynamic process, from a charged black hole neutralizing itself by emitting electromagnetic radiation to the complete evaporation of a black hole via Hawking radiation. The entire, aeons-long life and death of a black hole, from its formation to its final puff of radiation, is mapped onto a finite interval of retarded time $u$ at future null infinity, a complete story written on the final page of spacetime.

We arrive now at the most modern, and perhaps most profound, application of null infinity. It turns out that this boundary is not just a passive screen for viewing the universe. It has a rich life of its own. It possesses a deep and beautiful set of symmetries, far more extensive than we ever expected, and these symmetries have physical consequences.

In physics, symmetries are paramount. As Emmy Noether taught us, every continuous symmetry of a physical system corresponds to a conserved quantity. For decades, it was believed that the only symmetries of an isolated system in general relativity were the ten familiar Poincaré symmetries: translations in time (conservation of energy), translations in space (conservation of momentum), rotations (conservation of angular momentum), and boosts. However, by carefully studying the structure of spacetime near null infinity, physicists in the 1960s, led by Bondi, van der Burg, Metzner, and Sachs, discovered an infinite number of new asymptotic symmetries. This infinite-dimensional group is now known as the BMS group.

What do these new symmetries mean? For a long time, the implications were obscure. But in recent years, a stunning synthesis has emerged connecting three seemingly disparate topics: the ​​asymptotic symmetries​​ of the BMS group, ​​soft theorems​​ in quantum field theory (which govern the behavior of very low-energy particles), and a physical phenomenon called the ​​gravitational memory effect​​ (a permanent distortion of spacetime left behind by a burst of gravitational waves).

We can see a tangible example of this connection in a simpler theory: quantum electrodynamics (QED). Here too, there exists an infinite set of asymptotic symmetries at null infinity, known as large gauge transformations. Each of these symmetries gives rise to a new conserved quantity, a "soft charge." When particles scatter off one another—for instance, an electron and a positron annihilating to produce a muon and an anti-muon—the process must obey this infinite family of new conservation laws. The total soft charge carried in by the initial particles must be precisely balanced by the charge carried out by the final particles, plus any created in the form of low-energy "soft" photons. These laws impose powerful, non-trivial constraints on the possible outcomes of any interaction in the universe, all dictated by the symmetry structure of null infinity.

The study of null infinity is not a closed chapter of physics. It is a vibrant, active area of research that is forcing us to rethink our understanding of foundational principles. It is a testament to the remarkable power of a good physical idea that this mathematical boundary, conceived to bring precision to the study of radiation in general relativity, has become a central stage where the deepest connections between gravity, spacetime, and quantum mechanics are being discovered.