1
New Uses for the Eschaton
One way to examine problems with huge unknowns – SETI is a classic example – is through the construction of a so-called ‘toy model.’ I linger a moment on the term because I want to purge the notion that it infers a lightweight conclusion. A toy model simplifies details to look for the big picture. It can be a useful analytical tool, a way of screening out some of the complexities in order to focus on core issues. And yes, it’s theoretical and idealized, not predictive.
But sometimes a toy model offers approaches we might otherwise miss. Consider how many variables we have to work with in SETI. What kind of signaling strategy would an extraterrestrial civilization choose? What sort of timeframe would it operate under? What cultural values determine its behavior? What is its intent? You can see how long this list can become. I’ll stop here.
The toy model I want to focus on today is one David Kipping uses in a new paper called “The Eschatian Hypothesis.” The term refers to what we might call ‘final things.’ Eschaton is a word that turns up in both cosmology and theology, in the former case talking about issues like the ultimate fate of the cosmos. So when Kipping (Columbia University) uses it in a SETI context, he’s going for the broadest possible approach, the ‘big picture’ of what a detection would look like.
I have to pause here for a moment to quote science fiction writer Charles Stross, who finds uses for ‘eschaton’ in his Singularity Sky (Ace, 2004), to wit:
I am the Eschaton. I am not your God.
I am descended from you, and exist in your future.
Thou shalt not violate causality within my historic light cone. Or else.
Love the ‘or else.’
Let’s now dig into the new paper. Published in Research Notes of the AAS, the paper homes in on a kind of bias that haunts our observations. Consider that the first exoplanets ever found were at the pulsar PSR 1257+12. Or the fact that the first main sequence star with a planet was found to host a ‘hot Jupiter,’ which back in 1995, when 51 Pegasi b was discovered, wasn’t even a category anyone ever thought existed. The point is that we see the atypical first precisely because such worlds are so extreme. While our early population of detections is packed with hot Jupiters, we have learned that these worlds are in fact rarities. We begin to get a feel for the distribution of discoveries.
Hot Jupiters, in other words, are ‘loud.’ They’re the easiest of all radial velocity planet signatures to find. And yet they make up less than one percent of the exoplanets we’ve thus far found. The issue is broad. From the paper:
…over-representation of unusual astronomical phenomena in our surveys is not limited to exoplanetary science. One merely needs to look up at the night sky to note that approximately a third of the naked-eye stars are evolved giants, despite the fact less than one percent of stars are in such a state—a classic observational effect known as Malmquist bias (K. G. Malmquist 1922). Or consider that a supernova is expected roughly twice per century in Milky Way-sized galaxies (G. A. Tammann et al. 1994)—an astoundingly rare event. And yet, despite being an inherently rare type of transient, astronomers routinely detect thousands of supernovae every year (M. Nicholl 2021), as a product of their enormous luminosities.
That’s quite a thought. Go for a walk on a clear winter evening and look up. So many of the stars you’re seeing are giants in the terminal stages of their lifetimes. Those we can see at great range, but our nearest star, Proxima Centauri, demands a serious telescope for us to be able to see it. So we can’t help the bias that sets in until we realize how much of what we are seeing is rare. Sometimes we have to step back and ask ourselves why we are seeing it.
In SETI terms, Kipping steps back from the question to ask whether the first signatures of ETI, assuming one day they appear, will not be equally ‘loud,’ in the same way that supernovae are loud but actually quite rare. We might imagine a galaxy populated by stable, quiescent populations that we are not likely to see, cultures whose signatures are perhaps already in our data and accepted as natural. These are not the civilizations we would expect to see. What we might detect are the outliers, unstable cultures breaking into violent disequilibrium at the end of their lifetimes. These, supernova style, would be the ones that light up our sky.
Kipping’s toy model works on variables of average lifetime and luminosity, examining the consequences on detectability. A loud civilization is one that becomes highly visible for a fraction of its lifetime before going quiet for the rest. The model’s math demonstrates that a civilization that is 100 times louder than its peers – through any kind of disequilibrium with its normal state, as for example nuclear war or drastic climate change – becomes 1000 times more detectable. A supernova is incredibly rare, but also incredibly detectable.
Image: The toy model at work. This is from Kipping’s Cool Worlds video on the Eschatian Hypothesis.
The Eschatian search strategy involves wide-field, high cadence surveys. In other words, observe at short intervals and keep observing with rapid revisit times to the same source. A search like this is optimized for transients, and the author points out that a number of observatories and observing programs are “moving toward a regime where the sky is effectively monitored as a time-domain data set.” The Vera Rubin Observatory moves in this direction, as does PANOPTES (Panoptic Astronomical Networked Observatories for a Public Transiting Exoplanet Survey). The latter is not a SETI program, but its emphasis on short-duration, repeatable events falls under the Eschatian umbrella.
Rather than targeting narrowly defined technosignatures, Eschatian search strategies would instead prioritize broad, anomalous transients—in flux, spectrum, or apparent motion—whose luminosities and timescales are difficult to reconcile with known astrophysical phenomena. Thus, agnostic anomaly detection efforts (e.g., D. Giles & L. Walkowicz 2019; A. Wheeler & D. Kipping 2019) would offer a suggested pathway forward.
I’ve often imagined the first SETI detection as marking a funeral beacon, though likely not an intentional one. The Eschatian Hypothesis fits that thought nicely, but it also leaves open the prospect of what we may not detect until we actually go into the galaxy, the existence of civilizations whose lifetimes are reckoned in millions of years if not more. The astronomer Charles Lineweaver has pointed out that most of our galaxy’s terrestrial-class worlds are two billion years older than Earth. Kipping quotes the brilliant science fiction writer Karl Schroeder when he tunes up an old Arthur Clarke notion: Any sufficiently advanced civilization will be indistinguishable from nature. Stability infers coming to terms with societal disintegration and mastering it.
Cultures like that are going to be hard to distinguish from background noise. We’re much more likely to see a hard-charging, shorter-lived civilization meeting its fate.
The paper is Kipping, “The Eschatian Hypothesis,” Research Notes of the AAS Vol. 9, No. 12 (December, 2025), 334. Full text.
But sometimes a toy model offers approaches we might otherwise miss. Consider how many variables we have to work with in SETI. What kind of signaling strategy would an extraterrestrial civilization choose? What sort of timeframe would it operate under? What cultural values determine its behavior? What is its intent? You can see how long this list can become. I’ll stop here.
The toy model I want to focus on today is one David Kipping uses in a new paper called “The Eschatian Hypothesis.” The term refers to what we might call ‘final things.’ Eschaton is a word that turns up in both cosmology and theology, in the former case talking about issues like the ultimate fate of the cosmos. So when Kipping (Columbia University) uses it in a SETI context, he’s going for the broadest possible approach, the ‘big picture’ of what a detection would look like.
I have to pause here for a moment to quote science fiction writer Charles Stross, who finds uses for ‘eschaton’ in his Singularity Sky (Ace, 2004), to wit:
I am the Eschaton. I am not your God.
I am descended from you, and exist in your future.
Thou shalt not violate causality within my historic light cone. Or else.
Love the ‘or else.’
Let’s now dig into the new paper. Published in Research Notes of the AAS, the paper homes in on a kind of bias that haunts our observations. Consider that the first exoplanets ever found were at the pulsar PSR 1257+12. Or the fact that the first main sequence star with a planet was found to host a ‘hot Jupiter,’ which back in 1995, when 51 Pegasi b was discovered, wasn’t even a category anyone ever thought existed. The point is that we see the atypical first precisely because such worlds are so extreme. While our early population of detections is packed with hot Jupiters, we have learned that these worlds are in fact rarities. We begin to get a feel for the distribution of discoveries.
Hot Jupiters, in other words, are ‘loud.’ They’re the easiest of all radial velocity planet signatures to find. And yet they make up less than one percent of the exoplanets we’ve thus far found. The issue is broad. From the paper:
…over-representation of unusual astronomical phenomena in our surveys is not limited to exoplanetary science. One merely needs to look up at the night sky to note that approximately a third of the naked-eye stars are evolved giants, despite the fact less than one percent of stars are in such a state—a classic observational effect known as Malmquist bias (K. G. Malmquist 1922). Or consider that a supernova is expected roughly twice per century in Milky Way-sized galaxies (G. A. Tammann et al. 1994)—an astoundingly rare event. And yet, despite being an inherently rare type of transient, astronomers routinely detect thousands of supernovae every year (M. Nicholl 2021), as a product of their enormous luminosities.
That’s quite a thought. Go for a walk on a clear winter evening and look up. So many of the stars you’re seeing are giants in the terminal stages of their lifetimes. Those we can see at great range, but our nearest star, Proxima Centauri, demands a serious telescope for us to be able to see it. So we can’t help the bias that sets in until we realize how much of what we are seeing is rare. Sometimes we have to step back and ask ourselves why we are seeing it.
In SETI terms, Kipping steps back from the question to ask whether the first signatures of ETI, assuming one day they appear, will not be equally ‘loud,’ in the same way that supernovae are loud but actually quite rare. We might imagine a galaxy populated by stable, quiescent populations that we are not likely to see, cultures whose signatures are perhaps already in our data and accepted as natural. These are not the civilizations we would expect to see. What we might detect are the outliers, unstable cultures breaking into violent disequilibrium at the end of their lifetimes. These, supernova style, would be the ones that light up our sky.
Kipping’s toy model works on variables of average lifetime and luminosity, examining the consequences on detectability. A loud civilization is one that becomes highly visible for a fraction of its lifetime before going quiet for the rest. The model’s math demonstrates that a civilization that is 100 times louder than its peers – through any kind of disequilibrium with its normal state, as for example nuclear war or drastic climate change – becomes 1000 times more detectable. A supernova is incredibly rare, but also incredibly detectable.
Image: The toy model at work. This is from Kipping’s Cool Worlds video on the Eschatian Hypothesis.
The Eschatian search strategy involves wide-field, high cadence surveys. In other words, observe at short intervals and keep observing with rapid revisit times to the same source. A search like this is optimized for transients, and the author points out that a number of observatories and observing programs are “moving toward a regime where the sky is effectively monitored as a time-domain data set.” The Vera Rubin Observatory moves in this direction, as does PANOPTES (Panoptic Astronomical Networked Observatories for a Public Transiting Exoplanet Survey). The latter is not a SETI program, but its emphasis on short-duration, repeatable events falls under the Eschatian umbrella.
Rather than targeting narrowly defined technosignatures, Eschatian search strategies would instead prioritize broad, anomalous transients—in flux, spectrum, or apparent motion—whose luminosities and timescales are difficult to reconcile with known astrophysical phenomena. Thus, agnostic anomaly detection efforts (e.g., D. Giles & L. Walkowicz 2019; A. Wheeler & D. Kipping 2019) would offer a suggested pathway forward.
I’ve often imagined the first SETI detection as marking a funeral beacon, though likely not an intentional one. The Eschatian Hypothesis fits that thought nicely, but it also leaves open the prospect of what we may not detect until we actually go into the galaxy, the existence of civilizations whose lifetimes are reckoned in millions of years if not more. The astronomer Charles Lineweaver has pointed out that most of our galaxy’s terrestrial-class worlds are two billion years older than Earth. Kipping quotes the brilliant science fiction writer Karl Schroeder when he tunes up an old Arthur Clarke notion: Any sufficiently advanced civilization will be indistinguishable from nature. Stability infers coming to terms with societal disintegration and mastering it.
Cultures like that are going to be hard to distinguish from background noise. We’re much more likely to see a hard-charging, shorter-lived civilization meeting its fate.
The paper is Kipping, “The Eschatian Hypothesis,” Research Notes of the AAS Vol. 9, No. 12 (December, 2025), 334. Full text.
Anonymous