199
The Rest is Silence: Empirically Equivalent Hypotheses about the Universe
Because we so often talk about finding an Earth 2.0, I’m reminded that the discipline of astrobiology all too easily falls prey to an earthly assumption: Intelligent beings elsewhere must take forms compatible with our planet. Thus the recent post on SETI and fireflies, one I enjoyed writing because it explores how communications work amongst non-human species here on Earth. Learning about such methods may lessen whatever anthropomorphic bias SETI retains. But these thoughts also emphasize that we continue to search in the dark. It’s a natural question to ask just where SETI goes from here. What happens if in all our work, we continue to confront silence? I’ve been asked before what a null result in SETI means – how long do we have to keep doing this before we simply acknowledge that there is no one out there? But a better question is, how would we ever discover a definitive answer given the scale of the cosmos? If not in this galaxy, maybe in Andromeda? If not there, M87?
In today’s essay, Nick Nielsen returns to dig into how these questions relate to the way we do science, and ponders what we can learn by continuing to push out into a universe that remains stubbornly unyielding in its secrets. Nick is an independent scholar in Portland OR whose work has long graced these pages. Of late he has been producing videos on the philosophy of history. His most recent paper is “Human Presence in Extreme Environments as a Condition of Knowledge: An Epistemological Inquiry.” As Breakthrough Listen continues and we enter the era of the Extremely Large Telescopes, questions like these will continue to resonate.
by J. N. Nielsen
What would it mean for humanity to be truly alone in the universe? In an earlier Centauri Dreams post, SETI’s Charismatic Megafauna, I discussed the tendency to focus on the extraterrestrial equivalent of what ecologists sometimes call “charismatic megafauna”—which in the case of SETI consists of little green men, space aliens, bug-eyed monsters, Martians, and their kin—whereas life and intelligence might take very different forms from those with which we’re familiar. [1] We might not feel much of a connection to the discovery of an exoplanet covered in a microbial mats, which couldn’t respond to us, much less communicate with us, but it would be evidence that there is other life in the universe, which suggests there may be other life yet to be found, which also would mean that, as life, we aren’t utterly alone in the universe. This in turn suggests the alternative view that we might be utterly alone, without a trace of life beyond Earth, and this gets to some fundamental questions. One way to cast some light on these questions is through a thought experiment that would bring the method of isolation to bear on the problem. I will focus on a single, narrow, unlikely scenario as a way to think about what it would mean to be truly alone in the universe.
Suppose, then, we find ourselves utterly alone in the universe—not only alone in the sense of there being no other intelligent species with whom we could communicate, and no evidence of any having existed in the universe’s past (from which we could experience unidirectional communication), but utterly alone in the sense that there’s not any sign of life in the universe, not even microbes. This scenario begins where we are today, inhabiting Earth, looking out into the cosmos to see what we can see, listening for SETI transmissions, trying to detect life elsewhere, and planning missions and designing spacecraft to extend this search further outward into the universe. This thought experiment, then, is consistent with what we know of the universe today; it is empirically equivalent to a universe positively brimming with other life and other civilizations that we just haven’t yet found; at our current level of technology and cosmological standing, we can’t distinguish between the two scenarios.
There is a cluster of related problems in the philosophy of science, including the underdetermination of theories, the possibility of empirically equivalent theories, theory choice, and holism in confirmation. I’m going to focus on the possibility of empirically equivalent theories, but what follows could be reformulated in terms of the others. What is it for a theory to be underdetermined? “To say that an outcome is underdetermined is to say that some information about initial conditions and rules or principles does not guarantee a unique solution.” (Lipton 1991: 6) If there’s no unique solution, there may be many possible solutions. Empirically equivalent theories are these many possible solutions. [2]
The discussion of empirically equivalent theories today has focused on the expansion of the consequence class of a theory, i.e., adopting auxiliary hypotheses so as to derive further testable consequences. We’re going to look at this through the other end of the telescope, however. Two theories can have radically different consequence classes while our ability conduct observations that would confirm or disconfirm these consequence classes is so limited that the available empirical evidence cannot distinguish between the two theories. That our ability to observe changes, and therefore the scope of the empirical consequence class changes, due to technologies and techniques of observation has been called “variability of the range of observation” (VRO) and the “inconstancy of the boundary of the observable.” (discussed in Laudan and Leplin 1991). Given VRO, there may be a time in the history of science when the observable consequence classes of two theories coincide, even while their unobservable consequence class ultimately diverges; at this time, the two theories are empirically equivalent in the sense that no current observation can confirm one while disconfirming the other. This is why we build larger telescopes and more powerful particle accelerators: to gain access to observations that can decide between theories that are empirically equivalent at present, but which have divergent consequence classes.
Returning to our thought experiment, where we began as we are today (unable to distinguish between a populous universe and terrestrial exceptionalism)—what do we do next? In our naïveté we make progress with our ongoing search. We build better telescopes, and we orbit larger and more sophisticated telescopes, with the intention of performing exoplanet atmospheric spectroscopy. We build spacecraft that allow us to explore our solar system. We go to Mars, but we don’t find anything there; no microbes in the permafrost or deep in subterranean bodies of water, and no sign of any life in the past. But we aren’t discouraged by this, because it’s always been possible that there was never life on Mars. There are many other places to explore in our solar system. Eventually we travel to interesting places like Titan, with its own thick atmosphere. We find this moon to be scientifically fascinating, but, again, no life of any kind is found. We send probes into subsurface liquid water oceans, first on Enceladus, then Europa, and we find nothing more complex in these waters than what we see in the astrochemistry of deep space: some simple organic molecules, but no macromolecules. Again, these worlds are scientifically fascinating, but we don’t find life and, again, we aren’t greatly bothered because we’ve only recently accustomed ourselves to the idea that there might be life in these oceans, and we can readily un-accustom ourselves as quickly. But it does raise questions, and so we seek out all the subsurface oceans in our solar system, even the brine pockets under the surface of Ceres, this time with a little more urgency. Again, we find many things of scientific interest, but no life, and no other unexpected forms of emergent complexity.
Suppose we exhaust every potential niche in our solar system, from the ice deep in craters on Mercury, to moons and comets in the outer solar system, and we find no life at all, and nothing like life either—no weird life (Toomey 2013), no life-as-we-do-not-know-it (Ward 2007), and no alternative forms of emergent complexity that are peers of life (Nielsen 2024). All the while as we’ve been exploring our solar system, our cosmological “backyard” as it were, we’ve continued to listen for SETI signals, and we’ve heard nothing. And we’ve continued to pursue exoplanet atmospheric spectroscopy, and we have a few false positives and a few mysteries—as always, scientifically interesting—but no life and no intelligence betrays itself. Now we’re several hundred years in the future, with better technology, better scientific understanding, and presumably a better chance of finding life, but still nothing.
If we had had some kind of a hint of possible life on another world, we could have had some definite target for the next stage of our exploration, but so far we’ve drawn a blank. We could choose our first interstellar objective by flipping a coin, but instead we choose to investigate the strangest planetary system we can find, with some mysterious and ambiguous observations that might be signs of biotic processes we don’t understand. And so we begin our interstellar exploration. Despite choosing a planetary system with ambiguous observations that might betray something more complex going on, once we arrive at the other planetary system and investigate it, we once again come up empty-handed. The investigation is scientifically interesting, as always, but it yields no life. Suppose we investigate this other planetary system as thoroughly as we’ve investigated our own solar system, and the whole thing, with all its potential niches for life, yields nothing but sterile, abiological processes, and nothing that on close inspection can’t be explained by chemistry, mineralogy, and geology.
Again we’re hundreds of years into the future, with interstellar exploration under our belt, and we still find ourselves alone in the cosmos. Not only are we alone in the cosmos, but the rest of the cosmos so far as we have studied it, is sterile. Nothing moves except that life that we brought with us from Earth. Still hundreds of years into the future and with all this additional exploration, and the scenario remains consistent with the scenario we know today: no life known beyond Earth. We can continue this process, exploring other scientifically interesting planetary systems, and trying our best to exhaustively explore our galaxy, but still finding nothing. At what threshold does this unlikelihood rise to the level of paradoxicality? Certainly at this point the strangeness of the situation in which we found ourselves would seem to require an explanation. So instead of merely searching for life, wherever we go we also seek to confirm that the laws of nature we’ve formulated to date remain consistent. That is to say, we test science for symmetry, because if we are able to find asymmetry, we will have found a limit to scientific knowledge.
We don’t have any non-arbitrary way to limit the scope of our scientific findings. If any given scientific findings could be shown to fail under translation in space or translation in time, then we would have reason to restrict their scope. Indeed, if we were to discover that our scientific findings fail beyond a given range in space and time, there would be an intense interest in exploring that boundary, mapping it, and understanding it. Eventually, we would want to explain this boundary. But without having discovered this boundary, we find ourselves in a quandary. Our science ought to apply to the universe entire. At least, this is the idealization of scientific knowledge that informs our practice. “On the one hand, there are truths founded on experiment, and verified approximately as far as almost isolated systems are concerned; on the other hand, there are postulates applicable to the whole of the universe and regarded as rigorously true.” (Poincaré 1952: 135-136) Earth and its biosphere are effectively an isolated system in Poincaré’s sense. We’ve constructed a science of biology based on experimentation within that isolated system (“verified approximately as far as almost isolated systems are concerned”), and the truths we’ve derived we project onto the universe (“applicable to the whole of the universe”). But our extrapolation of what we observe locally is an idealization, and our projecting a postulate onto the universe entire is equally an idealization. We can no more realize these idealizations in fact than we can construct a simple pendulum in fact. [3]
We need to distinguish between, on the one hand, that idealization used in science and without which science is impossible (e.g., the simple pendulum mentioned above), and, on the other hand, that idealization that is impossible for science to capture in any finite formalization, but which can be approximated (like the ideal isolation of experiment discussed by Poincaré). Holism in confirmation, to which I referred above (and which is especially associated with Duhem-Quine thesis), is an instance of this latter kind of idealization. Both forms of idealization force compromises upon science through approximation; we accept a result that is “good enough,” even if not perfect. Each form of idealization implies the other, as, for example, the impossibility of accounting for all factors in an experiment (idealized isolation) implies the use of a simplified (ideal) model employed in place of actual complexity. Thus one ideal, realizable in theory, is substituted for another ideal, unrealizable in theory.
Our science of life in the universe, i.e., astrobiology, involves these two forms of idealization. Our schematic view of life, embodied in contemporary biology (for example, the taxonomic hierarchy of kingdom, phylum, class, order, family, genus, and species, or the idealized individuation of species), is the idealization realizable in theory, while the actual complexity of life, the countless interactions of actual biological individuals within a population both of others of its own species and individuals of other species, not to mention the complexity of the environment, is the idealization unrealizable in theory. The compromises we have accepted up to now, which have been good enough for the description of life on Earth, may not be adequate in an astrobiological context. Thus the testing of science for symmetries in space and time ought to include the testing of biology for symmetries, but, since in this thought experiment there are no other instances of biology beyond Earth, we cannot test for symmetry in biology as we would like to.
Suppose that our research confirms that as much of our science as can be tested is tested, and this science is as correct as it can be, and so it should be predictive, even if it doesn’t seem to be doing a good job at predicting what we find on other worlds. We don’t have to stop there, however. If we don’t find other living worlds in the cosmos, we might be able to create them. Exploring the universe on a cosmological scale would involve cosmological scales of time. If we were to travel to the Andromeda galaxy and back, about four million years would elapse back in the Milky Way. If we were to travel to other galaxy clusters, tens of millions of years or hundreds of millions of years would elapse. These are biologically significant periods of time, by which I mean these are scales of time over which macroevolutionary processes could take place. Our cosmological exploration would give us an opportunity to test that. In the sterile universe that we’ve discovered in this thought experiment, we still have the life from Earth that we’ve brought to the universe, and over biological scales of time life from Earth could go on to its own cosmological destiny. In our exploration of a sterile universe, we could plant the seeds of life from Earth and seek to create the biological universe we expected to find. The adaptive radiation of Earth life, facilitated by technology, could supply to other worlds the origins of life, and if origins of life were the bottleneck that produced a sterile universe, then once we supply that life to other worlds, these other worlds should develop biospheres in a predictable way (within expected parameters).
It probably wouldn’t be as easy as leaving some microbes on another planet or moon; we would have to prepare the ground for them so they weren’t immediately killed by the sterile environment. In other words, we would have to practice terraforming, at least to the extent of facilitating the survival, growth, and evolution of rudimentary Earth life on other worlds. If every attempt at terraforming immediately failed, that would be as strange as finding the universe to be sterile, and perhaps more inexplicable. But that’s a rather artificial scenario. It’s much more realistic to imagine that we attempt the terraforming of many worlds, and, despite some initial hopeful signs, all of our attempts at terraforming eventually die off, all for apparently different reasons, but none of them “take.” This would be strange, but we could still seek some kind of scientific explanation for this that demonstrated truly unique forces to be at work on Earth that allowed the biosphere not only to originate but to survive over cosmological scales of time (the “rare Earth” hypothesis with a vengeance).
If the seeding of Earth life on other worlds didn’t end in this strange way (as strange as the strangeness of exploring a sterile universe, so it’s a continued strangeness), but rather some of these terraforming experiments were successful, what comes next could entail a number of possible outcomes of ongoing strangeness. Leaving our galaxy for a few billion years of exploration in other galaxies, upon our return we could study these Earth life transplantations. Transplanted Earth life on other worlds could very nearly reproduce the biosphere on Earth, which would suggest very tight constraints of convergent evolution. If origins of life are very rare, and conditions for the further evolution of life are tightly constrained by convergent evolution, that would partially explain why we found a sterile universe, but the conditions would be far stronger than we would expect, and that would be scientifically unaccountable.
Another strange outcome would be if our terraformed worlds with transplanted Earth life all branched out in radically different directions over our multi-billion year absence exploring other galaxies. We would expect some branching out, but there would be a threshold of branching out, with none of the biospheric outcomes even vaguely resembling any of the others, that would defy expectations, and, in defying expectations, we would once again find ourselves faced with conditions much stronger than we would expect. In all these cases of strangeness—the strangeness of all our engineered biospheres failing, the strangeness of our engineered biospheres reproducing Earth’s biosphere to an unexpected degree of fidelity, and the strangeness of our engineered biospheres all branching off in radically different directions—we would confront something scientifically unaccountable. Even though we have no experience of other biospheres, we still have expectations for them based on the kind of norms we’ve come to expect from hundreds of years of practicing science, and departure from the norms of naturalism is strange. All of these scenarios would be strange in the sense of defying scientific expectations, and that would make them all scientifically interesting.
These scenarios are entirely consistent with our current observations, so that a sterile universe with Earth as the sole exception where life is to be found is, at the present time, empirically equivalent with a living universe in which life is commonplace. However, the exploration of our own solar system could offer further confirmation of a sterile universe, or disconfirm it, or modify it. If, as in the preceding scenario, we find nothing at all beyond Earth in our solar system, this will increase the degree of confirmation for the sterile universe hypothesis (which we could also call terrestrial exceptionalism). If we were to find life elsewhere in our solar system, but molecular phylogeny shows that all life in our solar system derives from a single origins of life event, then we will have demonstrated that life as we know it can be exchanged among worlds, but the likelihood of independent origins of life events would be rendered somewhat less probable, especially if we were to determine that any of the over life-bearing niches in our solar system were not only habitable, but unambiguously urable. [4]
If we were to find life elsewhere in our solar system and molecular phylogeny shows that these other instances of life derive from independent origins of life events, then this would increase the degree of confirmation of the predictability of origins of life events on the basis of our present understanding of biology. The number of distinct origins of life events could serve as a metric to quantify this. [5] If we were to find life elsewhere in the solar system and this life consists of an eclectic admixture of life with the same origins event as life on Earth, and life derived from distinct origins events, then we would know both that the distribution of life among worlds and origins of life were common, and on this basis we would expect to find the same in the cosmos at large. An exacting analysis of this maximal life scenario would probably yield interesting details, such as particular forms of life that appear the most readily once boundary conditions have been met, and particular forms of life that are more finicky and don’t as readily appear. Similarly, among life distributed across many worlds we would likely find that some varieties are more readily distributed than others.
If the solar system is brimming with life, we could still maintain that the rest of the cosmos is sterile, reproducing the same scenario as above, but the scenario would be less persuasive, or perhaps I should say less frightening, knowing that life had originated elsewhere and was not absolutely unique to Earth. Nevertheless, we could yet be faced with a scenario that is even more inexplicable than the above (call it the “augmented Fermi paradox” if you like). If we found our solar system to be brimming with life, with life easily originating and easily transferable among astronomical bodies, increasing our confidence that life is common in the universe and widely distributed, and then we went out to explore the wider universe and found it to be sterile, we would be faced with an even greater mystery than the mystery we face today. The dilemma imposed upon us by the Fermi paradox can yet take more severe forms than the form in which we know it today. The possibilities are all the more tantalizing given that at least some of these questions will be answered by evidence within our own solar system.
It seems likely that the Fermi paradox is an artifact of the contemporary state of science, and will persist as long as science and scientific knowledge retains its current state of conceptual development. Anglo-American philosophy of science has tended to focus on confirmation and disconfirmation of theories, while continental philosophy of science has developed the concept of idealization [6]; I have drawn on both of these traditions in the above thought experiment, and it will probably require resources from both of these traditions to resolve the impasse we find ourselves at present. Because science and scientific knowledge itself would be called into question in this scenario, there would be a need for human beings themselves to travel to the remotest parts of the universe to ensure the integrity of the scientific process and the data collected (Nielsen 2025b), and this will in turn demand heroic virtues (Nielsen 2025) on the part of those who undertake this scientific research program.
Thanks are due to Alex Tolley for suggesting this.
Notes
1. I have discussed different definitions of life in (Nielsen 2023), and I have formulated a common theoretical framework for discussing forms of life and intelligence not familiar to us in (Nielsen 2024b) and (Nielsen 2025a).
2. The discussion of empirically equivalent theories probably originates in (Van Fraassen 1980).
3. I am using “simple pendulum” here in the sense of an idealized mathematical model of a pendulum that assumes a frictionless fulcrum, a weightless string, a point mass weight bob, absence of air drag, short amplitude (small-angle approximation where sinθ≈θ), inelasticity of pendulum length, rigidity of the pendulum support, and a uniform field of gravity during operation of the pendulum. Actual pendulums can be made precise to an arbitrary degree, but they can never exhaustively converge on the properties of an ideal pendulum.
4. “Urable” planetary bodies are those that are, “conducive to the chemical reactions and molecular assembly processes required for the origin of life.” (Deamer, et al. 2022)
5. The degree of distribution of life from a single origins of life event, presumably a function of the particular form of life involved, the conditions of carriage (i.e., the mechanism of distribution), and the structure of the planetary system in question, would provide another metric relevant to assessing the ability of life to survive and reproduce on cosmological scales.
6. Brill has published fourteen volumes on idealization in the series Poznań Studies in the Philosophy of the Sciences and the Humanities.
References
Deamer, D., Cary, F., & Damer, B. (2022). Urability: A property of planetary bodies that can support an origin of life. Astrobiology, 22(7), 889-900.
Laudan, L. and Leplin, J. (1991). “Empirical Equivalence and Underdetermination.” Journal of Philosophy. 88: 449–472.
Lipton, Peter. (1991). Inference to the Best Explanation. Routledge.
Nielsen, J. N. (2023). “The Life and Death of Habitable Worlds.” Chapter in: Death And Anti-Death, Volume 21: One Year After James Lovelock (1919-2022). Edited by Charles Tandy. 2023. Ria University Press.
Nielsen, J. N. (2024a). Heroic virtues in space exploration: everydayness and supererogation on Earth and beyond,” Heroism Sci. doi:10.26736/hs.2024.01.12
Nielsen, J. N. (2024b). Peer Complexity in Big History. Journal of Big History, VIII(1); 83-98.
DOI | https://doi.org/10.22339/jbh.v8i1.8111 (An expanded version of this paper is to appear as “Humanity’s Place in the Universe: Peer Complexity, SETI, and the Fermi Paradox” in Complexity in Universal Evolution—A Big History Perspective.)
Nielsen, J.N. (2025a). An Approach to Constructing a Big History Complexity Ladder. In: LePoire, D.J., Grinin, L., Korotayev, A. (eds) Navigating Complexity in Big History. World-Systems Evolution and Global Futures. Springer, Cham. https://doi.org/10.1007/978-3-031-85410-1_12
Nielsen, J.N. (2025b). Human presence in extreme environments as a condition of knowledge: an Epistemological inquiry. Front. Virtual Real. 6:1653648. doi: 10.3389/frvir.2025.1653648
Poincaré, Henri. (1952). Science and Hypothesis. Dover.
Toomey, D. (2013). Weird life: The search for life that is very, very different from our own. WW Norton & Company.
Van Fraassen, B. C. (1980). The scientific image. Oxford University Press.
Ward, P. (2007). Life as we do not know it: the NASA search for (and synthesis of) alien life. Penguin.
In today’s essay, Nick Nielsen returns to dig into how these questions relate to the way we do science, and ponders what we can learn by continuing to push out into a universe that remains stubbornly unyielding in its secrets. Nick is an independent scholar in Portland OR whose work has long graced these pages. Of late he has been producing videos on the philosophy of history. His most recent paper is “Human Presence in Extreme Environments as a Condition of Knowledge: An Epistemological Inquiry.” As Breakthrough Listen continues and we enter the era of the Extremely Large Telescopes, questions like these will continue to resonate.
by J. N. Nielsen
What would it mean for humanity to be truly alone in the universe? In an earlier Centauri Dreams post, SETI’s Charismatic Megafauna, I discussed the tendency to focus on the extraterrestrial equivalent of what ecologists sometimes call “charismatic megafauna”—which in the case of SETI consists of little green men, space aliens, bug-eyed monsters, Martians, and their kin—whereas life and intelligence might take very different forms from those with which we’re familiar. [1] We might not feel much of a connection to the discovery of an exoplanet covered in a microbial mats, which couldn’t respond to us, much less communicate with us, but it would be evidence that there is other life in the universe, which suggests there may be other life yet to be found, which also would mean that, as life, we aren’t utterly alone in the universe. This in turn suggests the alternative view that we might be utterly alone, without a trace of life beyond Earth, and this gets to some fundamental questions. One way to cast some light on these questions is through a thought experiment that would bring the method of isolation to bear on the problem. I will focus on a single, narrow, unlikely scenario as a way to think about what it would mean to be truly alone in the universe.
Suppose, then, we find ourselves utterly alone in the universe—not only alone in the sense of there being no other intelligent species with whom we could communicate, and no evidence of any having existed in the universe’s past (from which we could experience unidirectional communication), but utterly alone in the sense that there’s not any sign of life in the universe, not even microbes. This scenario begins where we are today, inhabiting Earth, looking out into the cosmos to see what we can see, listening for SETI transmissions, trying to detect life elsewhere, and planning missions and designing spacecraft to extend this search further outward into the universe. This thought experiment, then, is consistent with what we know of the universe today; it is empirically equivalent to a universe positively brimming with other life and other civilizations that we just haven’t yet found; at our current level of technology and cosmological standing, we can’t distinguish between the two scenarios.
There is a cluster of related problems in the philosophy of science, including the underdetermination of theories, the possibility of empirically equivalent theories, theory choice, and holism in confirmation. I’m going to focus on the possibility of empirically equivalent theories, but what follows could be reformulated in terms of the others. What is it for a theory to be underdetermined? “To say that an outcome is underdetermined is to say that some information about initial conditions and rules or principles does not guarantee a unique solution.” (Lipton 1991: 6) If there’s no unique solution, there may be many possible solutions. Empirically equivalent theories are these many possible solutions. [2]
The discussion of empirically equivalent theories today has focused on the expansion of the consequence class of a theory, i.e., adopting auxiliary hypotheses so as to derive further testable consequences. We’re going to look at this through the other end of the telescope, however. Two theories can have radically different consequence classes while our ability conduct observations that would confirm or disconfirm these consequence classes is so limited that the available empirical evidence cannot distinguish between the two theories. That our ability to observe changes, and therefore the scope of the empirical consequence class changes, due to technologies and techniques of observation has been called “variability of the range of observation” (VRO) and the “inconstancy of the boundary of the observable.” (discussed in Laudan and Leplin 1991). Given VRO, there may be a time in the history of science when the observable consequence classes of two theories coincide, even while their unobservable consequence class ultimately diverges; at this time, the two theories are empirically equivalent in the sense that no current observation can confirm one while disconfirming the other. This is why we build larger telescopes and more powerful particle accelerators: to gain access to observations that can decide between theories that are empirically equivalent at present, but which have divergent consequence classes.
Returning to our thought experiment, where we began as we are today (unable to distinguish between a populous universe and terrestrial exceptionalism)—what do we do next? In our naïveté we make progress with our ongoing search. We build better telescopes, and we orbit larger and more sophisticated telescopes, with the intention of performing exoplanet atmospheric spectroscopy. We build spacecraft that allow us to explore our solar system. We go to Mars, but we don’t find anything there; no microbes in the permafrost or deep in subterranean bodies of water, and no sign of any life in the past. But we aren’t discouraged by this, because it’s always been possible that there was never life on Mars. There are many other places to explore in our solar system. Eventually we travel to interesting places like Titan, with its own thick atmosphere. We find this moon to be scientifically fascinating, but, again, no life of any kind is found. We send probes into subsurface liquid water oceans, first on Enceladus, then Europa, and we find nothing more complex in these waters than what we see in the astrochemistry of deep space: some simple organic molecules, but no macromolecules. Again, these worlds are scientifically fascinating, but we don’t find life and, again, we aren’t greatly bothered because we’ve only recently accustomed ourselves to the idea that there might be life in these oceans, and we can readily un-accustom ourselves as quickly. But it does raise questions, and so we seek out all the subsurface oceans in our solar system, even the brine pockets under the surface of Ceres, this time with a little more urgency. Again, we find many things of scientific interest, but no life, and no other unexpected forms of emergent complexity.
Suppose we exhaust every potential niche in our solar system, from the ice deep in craters on Mercury, to moons and comets in the outer solar system, and we find no life at all, and nothing like life either—no weird life (Toomey 2013), no life-as-we-do-not-know-it (Ward 2007), and no alternative forms of emergent complexity that are peers of life (Nielsen 2024). All the while as we’ve been exploring our solar system, our cosmological “backyard” as it were, we’ve continued to listen for SETI signals, and we’ve heard nothing. And we’ve continued to pursue exoplanet atmospheric spectroscopy, and we have a few false positives and a few mysteries—as always, scientifically interesting—but no life and no intelligence betrays itself. Now we’re several hundred years in the future, with better technology, better scientific understanding, and presumably a better chance of finding life, but still nothing.
If we had had some kind of a hint of possible life on another world, we could have had some definite target for the next stage of our exploration, but so far we’ve drawn a blank. We could choose our first interstellar objective by flipping a coin, but instead we choose to investigate the strangest planetary system we can find, with some mysterious and ambiguous observations that might be signs of biotic processes we don’t understand. And so we begin our interstellar exploration. Despite choosing a planetary system with ambiguous observations that might betray something more complex going on, once we arrive at the other planetary system and investigate it, we once again come up empty-handed. The investigation is scientifically interesting, as always, but it yields no life. Suppose we investigate this other planetary system as thoroughly as we’ve investigated our own solar system, and the whole thing, with all its potential niches for life, yields nothing but sterile, abiological processes, and nothing that on close inspection can’t be explained by chemistry, mineralogy, and geology.
Again we’re hundreds of years into the future, with interstellar exploration under our belt, and we still find ourselves alone in the cosmos. Not only are we alone in the cosmos, but the rest of the cosmos so far as we have studied it, is sterile. Nothing moves except that life that we brought with us from Earth. Still hundreds of years into the future and with all this additional exploration, and the scenario remains consistent with the scenario we know today: no life known beyond Earth. We can continue this process, exploring other scientifically interesting planetary systems, and trying our best to exhaustively explore our galaxy, but still finding nothing. At what threshold does this unlikelihood rise to the level of paradoxicality? Certainly at this point the strangeness of the situation in which we found ourselves would seem to require an explanation. So instead of merely searching for life, wherever we go we also seek to confirm that the laws of nature we’ve formulated to date remain consistent. That is to say, we test science for symmetry, because if we are able to find asymmetry, we will have found a limit to scientific knowledge.
We don’t have any non-arbitrary way to limit the scope of our scientific findings. If any given scientific findings could be shown to fail under translation in space or translation in time, then we would have reason to restrict their scope. Indeed, if we were to discover that our scientific findings fail beyond a given range in space and time, there would be an intense interest in exploring that boundary, mapping it, and understanding it. Eventually, we would want to explain this boundary. But without having discovered this boundary, we find ourselves in a quandary. Our science ought to apply to the universe entire. At least, this is the idealization of scientific knowledge that informs our practice. “On the one hand, there are truths founded on experiment, and verified approximately as far as almost isolated systems are concerned; on the other hand, there are postulates applicable to the whole of the universe and regarded as rigorously true.” (Poincaré 1952: 135-136) Earth and its biosphere are effectively an isolated system in Poincaré’s sense. We’ve constructed a science of biology based on experimentation within that isolated system (“verified approximately as far as almost isolated systems are concerned”), and the truths we’ve derived we project onto the universe (“applicable to the whole of the universe”). But our extrapolation of what we observe locally is an idealization, and our projecting a postulate onto the universe entire is equally an idealization. We can no more realize these idealizations in fact than we can construct a simple pendulum in fact. [3]
We need to distinguish between, on the one hand, that idealization used in science and without which science is impossible (e.g., the simple pendulum mentioned above), and, on the other hand, that idealization that is impossible for science to capture in any finite formalization, but which can be approximated (like the ideal isolation of experiment discussed by Poincaré). Holism in confirmation, to which I referred above (and which is especially associated with Duhem-Quine thesis), is an instance of this latter kind of idealization. Both forms of idealization force compromises upon science through approximation; we accept a result that is “good enough,” even if not perfect. Each form of idealization implies the other, as, for example, the impossibility of accounting for all factors in an experiment (idealized isolation) implies the use of a simplified (ideal) model employed in place of actual complexity. Thus one ideal, realizable in theory, is substituted for another ideal, unrealizable in theory.
Our science of life in the universe, i.e., astrobiology, involves these two forms of idealization. Our schematic view of life, embodied in contemporary biology (for example, the taxonomic hierarchy of kingdom, phylum, class, order, family, genus, and species, or the idealized individuation of species), is the idealization realizable in theory, while the actual complexity of life, the countless interactions of actual biological individuals within a population both of others of its own species and individuals of other species, not to mention the complexity of the environment, is the idealization unrealizable in theory. The compromises we have accepted up to now, which have been good enough for the description of life on Earth, may not be adequate in an astrobiological context. Thus the testing of science for symmetries in space and time ought to include the testing of biology for symmetries, but, since in this thought experiment there are no other instances of biology beyond Earth, we cannot test for symmetry in biology as we would like to.
Suppose that our research confirms that as much of our science as can be tested is tested, and this science is as correct as it can be, and so it should be predictive, even if it doesn’t seem to be doing a good job at predicting what we find on other worlds. We don’t have to stop there, however. If we don’t find other living worlds in the cosmos, we might be able to create them. Exploring the universe on a cosmological scale would involve cosmological scales of time. If we were to travel to the Andromeda galaxy and back, about four million years would elapse back in the Milky Way. If we were to travel to other galaxy clusters, tens of millions of years or hundreds of millions of years would elapse. These are biologically significant periods of time, by which I mean these are scales of time over which macroevolutionary processes could take place. Our cosmological exploration would give us an opportunity to test that. In the sterile universe that we’ve discovered in this thought experiment, we still have the life from Earth that we’ve brought to the universe, and over biological scales of time life from Earth could go on to its own cosmological destiny. In our exploration of a sterile universe, we could plant the seeds of life from Earth and seek to create the biological universe we expected to find. The adaptive radiation of Earth life, facilitated by technology, could supply to other worlds the origins of life, and if origins of life were the bottleneck that produced a sterile universe, then once we supply that life to other worlds, these other worlds should develop biospheres in a predictable way (within expected parameters).
It probably wouldn’t be as easy as leaving some microbes on another planet or moon; we would have to prepare the ground for them so they weren’t immediately killed by the sterile environment. In other words, we would have to practice terraforming, at least to the extent of facilitating the survival, growth, and evolution of rudimentary Earth life on other worlds. If every attempt at terraforming immediately failed, that would be as strange as finding the universe to be sterile, and perhaps more inexplicable. But that’s a rather artificial scenario. It’s much more realistic to imagine that we attempt the terraforming of many worlds, and, despite some initial hopeful signs, all of our attempts at terraforming eventually die off, all for apparently different reasons, but none of them “take.” This would be strange, but we could still seek some kind of scientific explanation for this that demonstrated truly unique forces to be at work on Earth that allowed the biosphere not only to originate but to survive over cosmological scales of time (the “rare Earth” hypothesis with a vengeance).
If the seeding of Earth life on other worlds didn’t end in this strange way (as strange as the strangeness of exploring a sterile universe, so it’s a continued strangeness), but rather some of these terraforming experiments were successful, what comes next could entail a number of possible outcomes of ongoing strangeness. Leaving our galaxy for a few billion years of exploration in other galaxies, upon our return we could study these Earth life transplantations. Transplanted Earth life on other worlds could very nearly reproduce the biosphere on Earth, which would suggest very tight constraints of convergent evolution. If origins of life are very rare, and conditions for the further evolution of life are tightly constrained by convergent evolution, that would partially explain why we found a sterile universe, but the conditions would be far stronger than we would expect, and that would be scientifically unaccountable.
Another strange outcome would be if our terraformed worlds with transplanted Earth life all branched out in radically different directions over our multi-billion year absence exploring other galaxies. We would expect some branching out, but there would be a threshold of branching out, with none of the biospheric outcomes even vaguely resembling any of the others, that would defy expectations, and, in defying expectations, we would once again find ourselves faced with conditions much stronger than we would expect. In all these cases of strangeness—the strangeness of all our engineered biospheres failing, the strangeness of our engineered biospheres reproducing Earth’s biosphere to an unexpected degree of fidelity, and the strangeness of our engineered biospheres all branching off in radically different directions—we would confront something scientifically unaccountable. Even though we have no experience of other biospheres, we still have expectations for them based on the kind of norms we’ve come to expect from hundreds of years of practicing science, and departure from the norms of naturalism is strange. All of these scenarios would be strange in the sense of defying scientific expectations, and that would make them all scientifically interesting.
These scenarios are entirely consistent with our current observations, so that a sterile universe with Earth as the sole exception where life is to be found is, at the present time, empirically equivalent with a living universe in which life is commonplace. However, the exploration of our own solar system could offer further confirmation of a sterile universe, or disconfirm it, or modify it. If, as in the preceding scenario, we find nothing at all beyond Earth in our solar system, this will increase the degree of confirmation for the sterile universe hypothesis (which we could also call terrestrial exceptionalism). If we were to find life elsewhere in our solar system, but molecular phylogeny shows that all life in our solar system derives from a single origins of life event, then we will have demonstrated that life as we know it can be exchanged among worlds, but the likelihood of independent origins of life events would be rendered somewhat less probable, especially if we were to determine that any of the over life-bearing niches in our solar system were not only habitable, but unambiguously urable. [4]
If we were to find life elsewhere in our solar system and molecular phylogeny shows that these other instances of life derive from independent origins of life events, then this would increase the degree of confirmation of the predictability of origins of life events on the basis of our present understanding of biology. The number of distinct origins of life events could serve as a metric to quantify this. [5] If we were to find life elsewhere in the solar system and this life consists of an eclectic admixture of life with the same origins event as life on Earth, and life derived from distinct origins events, then we would know both that the distribution of life among worlds and origins of life were common, and on this basis we would expect to find the same in the cosmos at large. An exacting analysis of this maximal life scenario would probably yield interesting details, such as particular forms of life that appear the most readily once boundary conditions have been met, and particular forms of life that are more finicky and don’t as readily appear. Similarly, among life distributed across many worlds we would likely find that some varieties are more readily distributed than others.
If the solar system is brimming with life, we could still maintain that the rest of the cosmos is sterile, reproducing the same scenario as above, but the scenario would be less persuasive, or perhaps I should say less frightening, knowing that life had originated elsewhere and was not absolutely unique to Earth. Nevertheless, we could yet be faced with a scenario that is even more inexplicable than the above (call it the “augmented Fermi paradox” if you like). If we found our solar system to be brimming with life, with life easily originating and easily transferable among astronomical bodies, increasing our confidence that life is common in the universe and widely distributed, and then we went out to explore the wider universe and found it to be sterile, we would be faced with an even greater mystery than the mystery we face today. The dilemma imposed upon us by the Fermi paradox can yet take more severe forms than the form in which we know it today. The possibilities are all the more tantalizing given that at least some of these questions will be answered by evidence within our own solar system.
It seems likely that the Fermi paradox is an artifact of the contemporary state of science, and will persist as long as science and scientific knowledge retains its current state of conceptual development. Anglo-American philosophy of science has tended to focus on confirmation and disconfirmation of theories, while continental philosophy of science has developed the concept of idealization [6]; I have drawn on both of these traditions in the above thought experiment, and it will probably require resources from both of these traditions to resolve the impasse we find ourselves at present. Because science and scientific knowledge itself would be called into question in this scenario, there would be a need for human beings themselves to travel to the remotest parts of the universe to ensure the integrity of the scientific process and the data collected (Nielsen 2025b), and this will in turn demand heroic virtues (Nielsen 2025) on the part of those who undertake this scientific research program.
Thanks are due to Alex Tolley for suggesting this.
Notes
1. I have discussed different definitions of life in (Nielsen 2023), and I have formulated a common theoretical framework for discussing forms of life and intelligence not familiar to us in (Nielsen 2024b) and (Nielsen 2025a).
2. The discussion of empirically equivalent theories probably originates in (Van Fraassen 1980).
3. I am using “simple pendulum” here in the sense of an idealized mathematical model of a pendulum that assumes a frictionless fulcrum, a weightless string, a point mass weight bob, absence of air drag, short amplitude (small-angle approximation where sinθ≈θ), inelasticity of pendulum length, rigidity of the pendulum support, and a uniform field of gravity during operation of the pendulum. Actual pendulums can be made precise to an arbitrary degree, but they can never exhaustively converge on the properties of an ideal pendulum.
4. “Urable” planetary bodies are those that are, “conducive to the chemical reactions and molecular assembly processes required for the origin of life.” (Deamer, et al. 2022)
5. The degree of distribution of life from a single origins of life event, presumably a function of the particular form of life involved, the conditions of carriage (i.e., the mechanism of distribution), and the structure of the planetary system in question, would provide another metric relevant to assessing the ability of life to survive and reproduce on cosmological scales.
6. Brill has published fourteen volumes on idealization in the series Poznań Studies in the Philosophy of the Sciences and the Humanities.
References
Deamer, D., Cary, F., & Damer, B. (2022). Urability: A property of planetary bodies that can support an origin of life. Astrobiology, 22(7), 889-900.
Laudan, L. and Leplin, J. (1991). “Empirical Equivalence and Underdetermination.” Journal of Philosophy. 88: 449–472.
Lipton, Peter. (1991). Inference to the Best Explanation. Routledge.
Nielsen, J. N. (2023). “The Life and Death of Habitable Worlds.” Chapter in: Death And Anti-Death, Volume 21: One Year After James Lovelock (1919-2022). Edited by Charles Tandy. 2023. Ria University Press.
Nielsen, J. N. (2024a). Heroic virtues in space exploration: everydayness and supererogation on Earth and beyond,” Heroism Sci. doi:10.26736/hs.2024.01.12
Nielsen, J. N. (2024b). Peer Complexity in Big History. Journal of Big History, VIII(1); 83-98.
DOI | https://doi.org/10.22339/jbh.v8i1.8111 (An expanded version of this paper is to appear as “Humanity’s Place in the Universe: Peer Complexity, SETI, and the Fermi Paradox” in Complexity in Universal Evolution—A Big History Perspective.)
Nielsen, J.N. (2025a). An Approach to Constructing a Big History Complexity Ladder. In: LePoire, D.J., Grinin, L., Korotayev, A. (eds) Navigating Complexity in Big History. World-Systems Evolution and Global Futures. Springer, Cham. https://doi.org/10.1007/978-3-031-85410-1_12
Nielsen, J.N. (2025b). Human presence in extreme environments as a condition of knowledge: an Epistemological inquiry. Front. Virtual Real. 6:1653648. doi: 10.3389/frvir.2025.1653648
Poincaré, Henri. (1952). Science and Hypothesis. Dover.
Toomey, D. (2013). Weird life: The search for life that is very, very different from our own. WW Norton & Company.
Van Fraassen, B. C. (1980). The scientific image. Oxford University Press.
Ward, P. (2007). Life as we do not know it: the NASA search for (and synthesis of) alien life. Penguin.
Anonymous
No comments yet.