Anthropocentric thinking is a persistent problem when we’re talking about extraterrestrial civilizations. Some of our most cherished notions can be invoked so effortlessly as to defy the imagination. The Copernican idea that life must exist elsewhere because elsewhere is bound to be more or less like here has had a long lifetime in SETI studies. It seems like the most basic common sense. And yet, as we’ll see once again today, scientific results continue to make it apparent that the fact that we are here does not mean that they are there.
Finding stellar systems more or less like our own continues to be difficult. I’m not going full ‘rare Earth’ here but acknowledging that our circumstances may be unusual enough to tamp down estimates of the number of life-supporting planets. For that matter, our lack of knowledge about abiogenesis itself makes the case that we cannot necessarily expect it around other suns. On this matter, at least, we should be able to gather data soon, perhaps with a Mars lander.
Which gets me to today’s interesting take on binary stars. The binaries – HD 129171 and HD 129209 – are G-class, both having formed from the same molecular cloud. This is a common enough scenario, because binary systems are extremely common. In fact, about half of the stars in the Milky Way have a companion star, and thus are likely to share a common chemistry.
About 180 light years from the Sun (based on Gaia data) in Boötes, the system is proving unusually helpful. Using the UVES spectrograph on ESO’s Very Large Telescope (VLT) in Chile, an international effort led by Anne Rathsam at the University of São Paulo in Brazil is using this binary as a laboratory to study chemical differences between the stars that may be telling us something about planet formation and orbital evolution. What is driving the differences?
All this has a bearing on life, because for life to occur, conditions must exist for long enough to let the necessary processes go to work. And that may be a problem. HD 129171 turns out to be enriched in refractory elements, while its companion HD 129209 is not. Refractory elements condense at high temperatures, as opposed to volatiles, and the proportion of one to the other can be an interesting diagnostic tool, telling us something about the system’s history. A star depleted in refractories is likely one that has spawned planets, while one heavy in these elements has likely swallowed planets whole.
Rathsam’s work, available in a paper just published in Astronomy & Astrophysics, focuses on beryllium, whose abundance here is interesting because it is not produced in stellar interiors. Measured properly, its presence points to a star that consumed planetary material long after formation. Lithium, beryllium, and boron, in fact, share characteristics that make them exceptional in terms of chemical emergence in the galaxy. Jorge Luis Melendez Moreno (USP), who served as study advisor to Rathsam on this paper, explains:
“All other chemical elements originate from primordial nucleosynthesis [the formation of the first atomic nuclei in the minutes following the Big Bang] or stellar nucleosynthesis [the nuclear fusion process that occurs inside stars throughout their lifetimes]. But not beryllium and boron. They primarily arise through a process called ‘cosmic spallation,’ in which high-energy particles fragment heavier nuclei, such as carbon, nitrogen, and oxygen, producing lighter elements.”
Image: The binary system of HD 129171 and HD 129209. Credit: Digital Sky Survey/Aladin/Anne Rathsam.
Lithium has its own uses as a marker, but the USP researchers found that beryllium is the more reliable tool, with a longer lasting chemical signature. HD 129171’s surplus of refractory elements (including magnesium, silicon, calcium and titanium) compared to its companion HD 129209 is accompanied by lithium and beryllium excess as well. The authors see evidence for rocky material equivalent to more than eleven times the mass of the Earth. This would have been ingested in the star, raising interesting questions about planet dynamics.
The infall of entire planets into a host star can be explained in various ways, from gravitational perturbations from other planets on eccentric orbits or skewed inclinations, to interactions within the early circumstellar disk driving young planets onto migratory trajectories. Although I hadn’t seen this paper before, the authors cite a 2025 study by Soares and team (citation below) that draws on simulations to show that about half of the stars simulated should ingest planets, leaving a chemical signature in perhaps 20 percent of them.
Evidently we can imagine system disruptions as a common occurrence in the Milky Way. We might add to this the fact that planet configurations with giant planets in circular outer orbits and rocky planets on inner ones are not common. Our observational data so far tends to confirm this, although tracking outer Jupiter-class planets with our current detection methods is always problematic given their sparse transit signatures. Overall, though, the picture that is emerging is that systems similar to the Solar System seem rare. If many systems are going through upheavals through planet migration and ingestion into the host, stable orbits may be trickier than we’ve thought. Rathsam comments:
“In our planetary system, the planets have relatively stable, low-eccentricity orbits. However, if planetary engulfment is common, it suggests that many systems undergo violent dynamic phases.”
Implications?
“Life wouldn’t just need billions of years to emerge and evolve. The planet would also have to remain in a sufficiently stable orbit to survive significant gravitational perturbations.”
If you’re wondering about our own star, it’s light in refractory materials relative to volatiles, which at least one recent study sees as a sign of planet formation at an early era. What became rocky planets and planetesimals, in other words, was never accreted into our star. I give that citation below. This and subsequent work seems to peg the Sun as the center of a dynamically quiet system compared to many, but be aware that the debate on this matter continues.
How useful is beryllium as a marker in such an analysis? The authors argue that despite the fact that beryllium is eventually depleted in a stellar interior, it persists long enough to provide reliable information. From the paper (my italics):
Refractory elements offer an excellent way to distinguish between the planet engulfment or the proto-cloud inhomogeneity scenarios. Since they have high condensation temperatures (≳1000 K), they are the primary constituents of the rocky material in planetary systems – terrestrial planets and cores of gaseous planets. In case of engulfment by a Sun-like star, this material is accreted by the star and then dissolved and mixed in the convective envelope, increasing the stellar surface abundances after the event (Sandquist et al. 2002). This process, however, produces a metal-rich outer layer with an unstable mean molecular weight gradient, which triggers thermohaline mixing (Théado & Vauclair 2012; Sevilla et al. 2022). For fragile elements such as Li and Be, this thermohaline mixing induces depletion, as it can carry these elements below the convective zone, into their burning regions. Thus, the chemical enrichment caused by the engulfment disappears over time. Nevertheless, Sevilla et al. (2022) demonstrated through simulations that the Li engulfment signature in stars with masses close to solar can be detected for ≥1 Gyr.
As far as I can tell, it’s the treatment of beryllium as a marker of engulfment that gives this paper its significance, as heretofore it has been used primarily as a diagnostic for the mixing of elements in stellar interiors. More broadly speaking, though, any reminders of the factors that make our Solar System adaptable for life should help us in identifying systems where life is less likely. That’s worth keeping in mind as we move beyond easy assumptions about life’s ubiquity and dig into realities that may make it more unusual than we thought.
The paper is Rathsam et al. (2026). Planet engulfment in the chemically anomalous HD 129171/HD 129209 pair. Astronomy & Astrophysics, 710, A236 (full text). The interesting paper on beryllium’s persistence is Soares et al. (2025). Assessing the processes behind planet engulfment and its imprints. Astronomy & Astrophysics, 693, A47 (preprint). The paper on the composition of our Sun is Meléndez et al. (2009). The peculiar solar composition and its possible relation to planet formation. The Astrophysical Journal Letters, 704(1), L66–L70. Abstract.
Wow – that’s a wide binary system. I haven’t found orbital parameters in an online search, but maybe they’re co-moving stars or in a very wide orbit.
There was a recent article: Planetary engulfment as a solution to solar-model discrepancies and its implications for planetary systems
https://academic.oup.com/mnras/article/551/4/stag1527/8787825
that suggests that the Sun’s Li depletion could be explained by the Sun engulfing a 5-10 Earth mass planet.
Maybe eating planets is a common dietary practice among stars.
The triple alpha process in stars uses Be and this is important in certain size stars like G class stars, since we can agree the size of the star does matter if we use an exact Earth twin and a similar solar system as our as the gold standard for the evolution of life that is Gas giants in the outer solar system and rocky inner planets.
I don’t get the accretion idea. All the inner planets came from the same protoplanetary gas cloud so some of the elements in the star is the same as those in the planets. Cosmic spalation must also occur in the gas cloud before the star formed. Consequently the refractory elements might easily be in only one star and not the other considering which star has the strongest gravity and that’s where the heavy elements went. This is nothing unusual. Furthermore, it don’t think it is good to start with a model that is different than our own system which is a double star system and any principles must only apply to double star systems.
Double star systems can be both favorable and unfavorable for the formation of planets. If the double stars are less than one AU apart, then they can have planets, but if too much greater say the Sun Saturn distance, then that would disrupt an Earth like system. Furthermore, double star systems at at a Sun Saturn distance both stars grab all the angular momentum and therefore gas and dust to make planets so these system are barren, no planetary offspring. Kippenhahn 1983. If the two stars are far enough apart then the probability of planet formation goes up. It has yet to be proven five sigma that Alpha Centauri A and B have planets. They may be further enough apart. Consequently, a lot of double star systems might not even have planets. I agree that the chemical elements in a star can help predict if a system has planets, but this alone is not enough and other contingencies take precedence such as the type of star and size of the solar system. I agree that our solar system indeed might be rare, but how rare? We could have at least ten of these to several hundred in our galaxy.
My position is more towards a weak Gaia hypothesis. Once we have the right environment which is a solar system exactly like ours with gas giants in the outer solar system and rocker inner planets and an Earth moon system in the life belt and planet similar size to our own, I think the probability of like evolving including intelligent life is one hundred percent. There are a lot of contingencies, but when they are met, I think we can assume this.
We tend to bang on about Earth been the cradle of life but what if it is not. What if moons like Europa are responsible for creating life and then it is redistributed via impacts and breakup of those moons. Moons like Europa are very good at the very least of incubating life from the very beginning of their formation where as the Earth has had some very violent periods.
Hi Paul.
A very interesting article.
I used to be of the view, that were orbiting an average star on an average planet in a very common system. However over the last few years I really have changed my view. G type stars are only 6-7% of the stars in the galaxy, and then a much smaller percentage seem to have solar systems like out own. The good points you make above along with the paper only seem consistent with my own thinking.
Cheers Edwin
If our solar system is not typical, what is the typical solar system that we have found so far?
6.000 confirmed exoplanets orbit around K and M stars which is most of the exoplanets. Copilot AI. We need to look at mostly G stars to answer your question.
Encountering this paper on disrupted star systems is akin to picking up a stone composed of many embedded pebbles and assessing what “shape” it is. It is very complicated research about what could have happened or can happen to an evolving star system from a planetary point of view. And perhaps what signature of a planetary system’s fate can be detected in the features of such a star. Reasoning beyond or critical review looks like a daunting task.
But what with the discovery of exoplanets anyway, it is not clear to me how this changes our circumstances. When “hot” Jupiters were discovered in the 1990s, it was inferred that most star systems would look like these cases, disregarding the fact that they were among the most detectable objects with 1990s technology. Yet correspondingly, from this vantage point, hot exoplanets with jovian mass are simply part of the interstellar terrain .
HD 129171 and HD 129209 have a notable difference, so to speak. Additional on line data about the stars are as follows:
HD 129171 HD 129209
————————————————————————————————-
MS Color Yellow-White Yellow-White
Right ascension: 14h 40m 18.513s 14h 40m 18.513s
Declination: 30° 26′ 37.102‘’ offset?
Parallax: 18.176 18.176
Distance: 179.447 ly 55.0 pc (same)
Mass: 115 % M Sun 109 % M Sun
radius : 120 % R Sun 130 % R Sun
Temperature: 5964 K 1.03 T Sun 5940 K | 1.03 T Sun
Age: 4.8 giga years | 1.04 Sun 5.1 giga years | 1.11 Sun
While this data set indicates that the two stars are very close, I have been unable to
locate binary system elements indicating that these two stars evolved together.
And judging by the age and mass estimates, there just might be alternative
explanations for the differences between the two presumed binaries such as
they are intrinsically different due to mass and age: controversy as to whether they are binaries. But to reiterate, the paper’s contention is that the distinction between the two stars is that one swallowed the elements that would have made up terrestrial planet “bedrock” and that it would be commonplace.
Since so many of our arguments for duplication of life in other star systems are based on examining our own circumstances, it is small wonder that the odds look nearly insurmountable. At various levels, though, we can identify common processes or bio-precursors that were necessary to establish life here and could have been extent in a number of “there’s”, even shared in nebulae surrounding newly forming stars. If a star or stellar system is not identical to ours there would be variation in results, but not necessarily that which precludes terrestrial planets. And with the case study above, I am not clear whether it is reducing the chances for life on planets as it is suggesting that some star systems will be so sterile they will not have planets. Clearly a number of them do and the planets are made of something.
But one problematic take-away to the probabilistic argument is that our circumstances were nearly the best possible culminating in our world as it exists now. And perhaps about to go bust for all the reasons enumerated in the news. Not only are we living in the “best of all possible worlds”, espoused by Leibniz and ridiculed by Voltaire, but nearly the best “instant” as well. Essays might address where these two philosophers would stand about positive or negatives of these past 2 or 3 centuries. It does seem akin to our pondering life’s existence around stars beyond that of Earth’s. Tools for forecasts based on sampling, but still inadequate…
But as a bottom line to this line of conjecture, is Earth’s current biosphere nearer to the lower or upper limit for several life related features?
What with the distribution of MS stars favoring red dwarfs vs. yellow, sampling/detecting over distances is much more favorable for the red dwarfs with their faster planetary periods. And planets more massive than the Earth are easier to detect. In other words validation of “Earths” around yellow suns takes longer than reds and a less frequent occurrence. The analysis above suggests that we might not find as many as we hope, but it will require more effort to establish this than with red dwarf stars due to sparsity, distance and orbital periods. Plus now we have to worry about the planet or their building blocks recycling into a sun-like star. Limited warranties?
So, despite our certain discomfort on more massive exoplanets with deep atmospheres, one could just as well look at Earth as fulfilling minimum environmental requirements for life beyond the microscopic. After all, while we have good odds now on detecting some ancient and departed life on Mars, but the epoch of life on Mars was likely shorter than that on Earth demonstrated thus far. A number of factors could be weighed to explain current Earth atmospheric pressure, but exactly what is magic about 1000 millibars at the surface other than it is good for us to have it after 4.5 billion years? On oceanic worlds, life could level out at pressure and depth based on a different evolutionary process.
Just trying to explain away our return call difficulties.
Though at a moment like this, if the chips seem down, William Tenn won the 1959 science fiction Hugo Award for a short story, titled “The Hell Bound Train”. Optimism in a different sense as remarked at the story’s end: It was not the destination, but the ride and the company.
Its important to keep in mind that binary systems are extremely common, indeed, probably about half of all stars are members of binary systems. Solitaries, like our own sun, seem to be the exception.
However, binary separation is highly diverse. Some binary pairs are actual in physical contact with one another. Others are so far apart no orbital motion around a common center of mass can be detected, the only reason they are even suspected to be binary is that they have identical proper motions and radial velocities. And of course, the degree of separation tends to influence our ability to establish their binary status. In other words, even professional binary star catalogs are victim to that old bugabear of astronomy–selection effects, and are of little use in drawing general statistical inferences about these systems.
It is pretty safe to say that once a star is determined to be a binary then it is understood both members of the pair formed at about the same time, and probably from the same parent material, that is, the chemical composition and isotopic abundances in each star started out the same way and are similar. But even this can be misleading. The evolution of a star is determined primarily by its mass, more massive bodies evolve much faster than light ones. And more evolved stars have had more complex nucleogenesis histories in their cores, or may have convective episodes in their histories. In other words, two stars may have formed from the same material at the same time, but their spectra may look very different today.
From the viewpoint of astrobiology, close binaries may play a disruptive role in planetary orbits due to tidal disruption, and the more massive star will more rapidly evolve, perhaps into a sterilizing nova or planetary nebula phase. And of course, the effects of these catastrophes will be highly dependent on their initial separation and the progress of biological evolution on the surviving star.
Having said that, I suspect Life (with a capital L) is resilient and adaptive and it will have a good chance of arising and persisting in a stable star system whose frisky, more massive binary brother is far enough away to not cause trouble. Whether we have enough data now to draw statistical conclusions about Life probabilities in these systems is something else again.
As for the appearance of “intelligence”, well, that’s another question entirely.
@Abelard, @Geoffrey
If the discussion is really about what fraction of planets are inhabited (Earthlike), then, from the perspective of the Drake equation, it is these 2 terms:
ne = the average number of planets that can potentially support life per star that has planets.
fl = the fraction of planets that could support life that actually develop life at some point.
By allowing for a narrower definition of ne (reducing the probability), the fl term will approach 1.0. The combined probability will be constant. Essentially, what the scientific and media perspectives become is the search for “Earth 2.0”, with the artistic expression as an image of a “blue planet” with land of a different geography than Earth. Whether that surface land is “green” depends on whether we accept Earth 2.0 as being in an earlier stage of evolution with primarily bacteria or single-celled algae, or whether the land is supporting evolved multicellular plants. The popular imagination prefers a verdant world with different biomes, from grasslands to forests, that supports metazoa, preferably charismatic animals.
While a naturalist would like to see a world populated with plants and animals that will have phenotypes that both mimic terrestrial forms and totally different forms that were never found in the terrestrial fossils, a molecular biologist might be more interested in whether the fundamental biology of exoplanet life is similar to, or very different from, the molecular biology of terrestrial life. Spectroscopic evidence might provide [strong] hints about the question of the fundamental biology, and very high-resolution remote imaging might please the naturalist; only a sampling mission with probes able to collect and analyze samples will satisfy the molecular biologist, who will want those missions to sample a greater number of living worlds.
Hello Paul—Thanks for yet another stimulating, thought-provoking essay. It’s fascinating to see the ways in which accepted notions of stellar and planetary system evolution are being upended by research such as this, which seems to indicate that chaos, rather than stability, may be the norm. Although it’s not in the same category as planetary engulfment, it was interesting to read another paper, by Stephen R. Kane et al., on “Tidal Demise: The Evolution and Fate of a Hypothetical Venusian Moon,” The Astrophysical Journal 1009:1 (Sept. 2026) [https://iopscience.iop.org/article/10.3847/1538-4357/ae9d6c], to find yet more evidence for how instability, or the lack thereof, may determine the chances for the emergence of life in planetary systems.
One line in your essay gave me pause, and I reread it several times to let its impact sink in: “More broadly speaking, though, any reminders of the factors that make our Solar System adaptable for life should help us in identifying systems where life is less likely.” It seems to offer an alternate paradigm for studying exoplanets—namely, what conditions make, or may make, “habitability” unlikely to occur?” I hope you’ll expand on that statement, either in replying to my comment or, perhaps, in a follow-up essay.
Yes, you’re reading my mind, John. I do hope to expand on that statement soon. But I need to do some more thinking about just where I want to go with that essay. More in the near future.
Not too far off topic, the origins of life, or rather the specific events that generate or modify life, are some of the events that should be modeled in efforts such as the Drake Equation and/or other ways of estimating the probability of life arising. One such event has happened only once or twice in the entire history of earth. That’s a pretty low probability of happening, even in what are proven conducive environments. I’m talking about the Paulinella, article here:
https://www.nytimes.com/2026/09/26/science/motel-science-discovery.html?unlocked_article_code=1.EVE.0kH2.voFUp-whzScj&smid=url-share
How many other exceedingly rare events led to the proliferation and evolution of life on this planet? What are the probabilities that any one of these might occur somewhere else, and even more interesting, what are the probabilities that a similar sequence of such events might have occurred elsewhere?
To me it seems, even in deep time and deep space, to be a rare thing to have occurred, and it therefor, to me, indicates a low probability of the contemporary existence or proximate existence of intelligent life (or any life?).
Anthropocentrism may go even deeper. We may stop looking for another Earth and still keep looking for ourselves. The question may not only be how rare the conditions for life are, but how much our definition of “life” determines what we are capable of recognizing.
Well said.