≡ Menu

A Test Case for Astroengineering

Last year the New Frontiers in Astronomy & Cosmology program, set up by the John Templeton Foundation as a grant-awarding organization, dispensed three grants with a bearing on what Clément Vidal calls ‘Zen SETI.’ The idea of looking into our astronomical data and making new observations to track possible signs of an extraterrestrial civilization at work is not new, and yesterday we looked at Freeman Dyson’s early contribution. Carl Sagan and Josif Shklovskii are also among those in a lineage we can extend back at least to the early 20th Century.

The recent grants show a gathering momentum for extending SETI in new directions. The team of Jason Wright (Pennsylvania State) and colleagues Steinn Sigurðsson and Matthew Povich is embarking on a hunt for Dyson spheres, which if observed in a distant galaxy colonized by a Kardashev Type III civilization, should throw an unmistakable signature in the infrared. Could we find such an object in our data from WISE, the Wide-field Infrared Survey Explorer satellite?

Or how about Kepler? Lucianne Walkowicz at Princeton was a winner of one of the 2013 grants, looking for hints of technology — of artificiality — around distant stars. The third recipient was exoplanet hunter Geoff Marcy (UC-Berkeley), working with Andrew Howard (University of Hawaii) and John Johnson (Caltech) on data from Kepler. When Clément Vidal writes about SETI as an observing rather than a communications program (in The Beginning and the End (Springer, 2014), he gives a powerful boost to the principles behind such searches.

Vidal’s book is rich and densely textured, which is why what I’ve tried to do in the last few days is to extract a few core ideas in the area most related to what we do here on Centauri Dreams. The early chapters are primarily concerned with building a worldview that is consistent with the latest thinking in cosmology, and Vidal speculates as well not only on multiverse theories but on a role for life in the cosmos that includes cosmogenesis, the creation of new universes. Olaf Stapledon immediately comes to mind because Vidal’s ambitious lunge into new intellectual terrain reminds me so much of the British writer. He would be at home with Vidal’s ideas of a cosmological artificial selection, one that draws on and extends ideas originally put forward by another deeply creative thinker, Lee Smolin.

Black Holes and their Uses

What can we, for example, say about black holes in a SETI context? For one thing, they form what would surely be the most powerful gravitational lensing opportunity available. Claudio Maccone has written about the potential of the central black hole in galaxies like the Milky Way becoming surrounded by a swarm of observing stations aligned with targets throughout the universe. For that matter, the lensing of electromagnetic radiation around a black hole is so intense that a communications channel could be set up for intergalactic distances (waiting out the answer is a different problem).

But black holes offer more than this. As the densest known objects in the universe, they can meet the needs of a Type III civilization faced with a continuing demand to support its energy consumption. Vidal runs through the literature on the matter, starting with Roger Penrose, who imagined extraction of black hole rotational energy by injection of matter, and through other scientists (the bibliography is extensive and quite good) who worked out the specifics of drawing energy from rotating black holes. Another possibility: Collecting energy from gravitational waves generated when black holes collide, or actually manipulating the merger of smaller black holes. In recent days, Louis Crane has studied small black holes as a power source — these objects convert matter into energy (Hawking radiation) at high levels of efficiency.

There are computational uses for black holes that push us out to the boundaries of computer science in the form of theorized ‘hypercomputers’ that draw on relativistic effects to dilate time in the proximity of black holes. Vidal’s philosophical ideas of cosmological artificial selection draw on the prospect that a Type III civilization may learn how to use black holes to create entirely new universes. However we view such prospects, the idea here is that for a wide variety of reasons, black holes should be attractors for intelligence. Vidal wants to know what the observable manifestations of any of these uses might be. Would such things be detectable?

Energy Sources for Advanced Civilizations

But we don’t have to confine our search to black holes themselves. If extracting energy from the thin accretion disk around a rotating black hole may be one of the most efficient power sources we can imagine, we can also look for similar configurations around neutron stars or white dwarfs. A key question, then is this: Could a civilization harness its star’s energy with efficiencies that approach black hole densities? The interesting family of binary systems called X-ray binaries (because of their emissions in the X-ray electromagnetic spectrum) should, Vidal believes, intrigue us as one possible sign of an artificial astrophysical system.


Image: An artist’s impression of GRS 1915, which is thought to be an X-ray binary. The black hole sucks material off the companion star, which is heated by friction, emitting X-rays. Credit: Rob Hynes, from http://www.phys.lsu.edu/~rih/.

There are others, including a whole range of contact binaries where stars exchange matter and energy in complicated ways. Vidal’s assumption is that these binaries are natural objects, but he doesn’t want to rule out the possibility that in at least some, we may be seeing something else at work.

Let me quote the author on this:

Accretion is a ubiquitous astrophysical process in galaxy and planet formation, so we may object that all binaries may simply always be natural. But let me introduce an analogy. Fission can be found in natural processes, as well as fusion, which is one of the core energetic processes in stellar evolution. Yet humans seek to copy them, and would certainly benefit greatly from — always — controlling them. So it is not because a process is known to occur naturally that its use in a given case is not under intelligent control. In fact, the situation may even be more subtle. The formation of XRBs might be natural, but they may later be controlled or taken over by ETIs, just as a river flowing down a mountain is a natural gravitational energy source that humans can harness with hydroelectric power stations.

In other words, there is a wide variety of binary stars in which we find accretion disks forming that could provide useful sources of energy to an advanced civilization. Vidal creates the term starivore to describe a civilization that could ‘feed’ on stars. More specifically:

It is an extraterrestrial civilization using stellar energy (Type KII) in the configuration of a slow non-conservative transient accreting binary…, with the dense primary… being either a planet, a white dwarf, a neutron star, or a black hole.

And indeed, Vidal quotes Stapledon’s novel Star Maker by way of showing that the idea of energy extraction from binary systems is not new. A speculative scenario that grows out of this is where our own civilization may one day go. As our technologies function on smaller and denser scales and we continue to move up the Kardashev ladder, thus using more and more energy, we run out of energy even if we cover the Earth with solar panels. So we bring Earth closer to the Sun (a Stapledon notion) to get more energy, with our descendants now living as postbiological beings. Still we need more energy, so stellar engineers create active accretion from, the Sun, transforming what had once been human life into a starivore civilization.

The density of the evolved Earth now approaches that of a white dwarf, and the new binary resembles what we see in our data as a cataclysmic variable, a binary system with white dwarf component. Vidal:

If such binary systems are starivores, then we should find that the primitive versions of them extract energy from a star paired with a planet that is not dense compared to WDs, NS, or BHs. This would happen at a low accretion rate, so planetary accretion is one of the concrete predictions from the starivore hypothesis (and indeed planet-star interactions have recently been discovered…)


Image: An artist’s concept of the accretion disk around the binary star system WZ Sge. P. Marenfeld and NOAO/AURA/NSF.

Vidal’s hypothesis of starivores lets us see high energy astrophysics from an astrobiological point of view. Speculative? Of course, but Vidal is a philosopher for whom the play of ideas is as entrancing as the flow of notes in a Bach fugue. Rather than claiming the existence of starivore civilizations, he offers data on the wide variety of binary systems and the possibilities for energy extraction, with predictions about what we might see if such civilizations exist. A high energy astrobiology agenda is presented containing proposals for specific research. I do not have time this morning to go through the wealth of supporting argument but the book is well worth extended study.

Ultimately, the starivore idea is Vidal’s way of describing SETI’s new direction, a concrete example of how we can study objects in our data that may show the signature of extraterrestrial engineering. Building a robust scientific structure for such inquiries is at the heart of The Beginning and the End, whose principles are being played out and refined in the ongoing SETI searches mentioned at the beginning of this post. As with the original SETI work back to the days of Project Ozma, we can’t know what we’re going to find until we mount the actual search. Finding a Type II or III culture — or its remnants — would show us what intelligent life is capable of, while raising the familiar question of how long any technological species can hope to survive.


Comments on this entry are closed.

  • Hary R Ray October 30, 2014, 9:33

    First; calling this type of search “SETI” is not entirely correct. Seth Shostak and Jill Tarter prefer to call it “SETT”, or, the search for extraterrestrial technology. Secondly, it can be done by (VERY) ameture data miners (LIKE ME) just by surfing the net! CASE IN POINT: Every time a NEW “KOI” is posted with FOUR (NO MORE AND NO LESS) close-in planet candidates, I check the THREE orbital period ratios of ADJACENT candidates in hope of finding them to be either pi, e, phi, or SOME OTHER COMBINATION of those three numbers. Kepler data for short period orbits are usually accurate to EIGHT decimal points! If any ONE orbital period ratio were any of these three numbers, it would just be a coincidence, any two would be either an EXTREME coincidence, or a three sigma detection. ALL THREE WOULD BE A SIX SIGMA DETECTION of what I call “EOS”‘s, which stands for “Echos On Steriods; in other words, HUGE inflatable ballon-like structures which a KI civilization could create and maintain.

  • Christopher L. Bennett October 30, 2014, 10:06

    “Starivore?” Ouch. It should probably be “astrovore.”

  • Alex Tolley October 30, 2014, 14:15

    We definitely should be looking to see how to distinguish the natural from the artificial. I would add the list of examples – wild fires from camp fires. All of the examples given should be distinguishable by some feature[s] that separate the natural event from the controlled, artificial one. Sometimes it gets a bit tricky – e.g. Beaver dams. Without knowing about beavers, could one say it was built by human level intelligence, animal intelligence or just a natural result of logs jamming up in the river? What would be the stellar equivalents that separate natural events from astro-engineering? My sense is engineering on this scale will also require other, less ambiguous artifacts that may be detectable at distance. Much like campfires are constrained artificially, or may be inside caves, and may have pots and supports to help cook food, stellar engineering may also have such artifacts that indicate intelligent construction rather than a natural occurrence.

  • CharlesJQuarra October 30, 2014, 15:37

    Dyson swarms would be virtually indistinguishable from orbital debris disks from afar.

    Accretion disk would give a lot of energy cheaply, but it wouldn’t be the most efficient source of energy from a given amount of matter in the long term. The most efficient would have to be splitting the accretion mass into stars that are below the red dwarf threshold for buoyant convection in the star’s core. Why would such a civilization do hasty things as throw matter irreversibly into a black hole?

  • william October 30, 2014, 16:39

    Speaking of signal detection, would you Mr. Gilster, please explain to me what is all this about regarding the 550 AU (at least that’s what I think it is) point beyond our sun, which allows us to see what’s going on around other stars ? Do I have that right ? It’s something that you have mentioned many times in many of your different articles on here.

    As an aside, I just recently realized that you name this website after your original 2004 book called “Centauri dreams”; it just hit me recently. In regards to your 2004 book would you now have written it differently than you did then, knowing and learning all the things you have in the 10 years since that time ?

  • Andrew Palfreyman October 30, 2014, 17:10

    If astrovore (yes, better) cultures exist, we might detect them in, say, binary systems by noting an energy loss at odds with the raw dynamics.

  • Paul Gilster October 30, 2014, 19:30

    william writes:

    Speaking of signal detection, would you Mr. Gilster, please explain to me what is all this about regarding the 550 AU (at least that’s what I think it is) point beyond our sun, which allows us to see what’s going on around other stars ? Do I have that right ? It’s something that you have mentioned many times in many of your different articles on here.

    Here are two earlier entries that will, I hope, explain what’s going on as light is bent around a massive object:



    In short: mass curves spacetime, which means that light from behind a massive object can be ‘bent,’ or focused, at a point beyond the object. Light from behind the Sun can be bent and we find a natural focus at 550 AU and farther out, depending on wavelength, which we can exploit. But read the two articles above and see if they help explain what’s going on. If we can put a spacecraft exactly opposite, on the other side of the Sun, from what we want to observe, we can use this lensing phenomenon to study the object in question in great detail. At least, that’s the theory, and it has a history in astronomy of delivering on its promise, including detections of exoplanets.

    You also write:

    I just recently realized that you name this website after your original 2004 book called “Centauri dreams”; it just hit me recently. In regards to your 2004 book would you now have written it differently than you did then, knowing and learning all the things you have in the 10 years since that time ?

    Good question! I’m sure I would do a lot differently because so much has happened, especially in the exoplanet field, and also in the growth of interstellar propulsion research. The public seems more aware of interstellar flight as a possibility today, what with all the talk about habitable Earth-class planets being discovered at some point soon. I probably would have approached the topic with more interviews with people in the interstellar community and added a chapter or two on commercial space / crowdsourcing / Net-based communities supporting interstellar research, etc. Hard to know, of course, because any new edition would grow step by step, and so I don’t exactly know what I would see that I wanted to change. But maybe the above is a start. Thanks for the good questions!

    By the way, I consider Centauri Dreams (the website) to be a continuation of the book, so in a way I feel like I’ve been writing the book continuously since 2002, when I started writing Chapter 1.

  • Eric Hughes October 30, 2014, 20:02

    If it’s bad to mix English and Latin roots into the word “starivore”, it’s only less bad to mix Greek and Latin roots into the word “astrovore”. For those who would like to be coinage-practice correct, please use “stellavore”.

    For the record, the first time I saw that word, I winced. Not metaphorically, but on my face.

  • Rob Henry October 30, 2014, 21:31

    Charles Quarra writes “Dyson swarms would be virtually indistinguishable from orbital debris disks from afar.” That can’t be right can it?

    To me their should be no dust, yet very high levels of infrared radiation. High occultation, yet no planets. Very high reflection of light an AU or so out, yet no discernible single axis on which accretion disc could be based. Lastly, I thought we could expect lots of absorption from unnatural surfaces, such as those covered in a thin layer of metallic gold. I know very little about such matters, and wonder if anyone could put me right.

  • Shaun Steenkamp October 30, 2014, 22:42

    Hary, I believe that calling this work “SETI” is entirely accurate.

    If we examine the methods of SETI (in the conventional sense) you could mount a convincing argument that it is in actual fact SETC (Search for Extra-terrestrial Communication). You could apply this argument to almost any search method and you would never get one that you could call “SETI”. I mean, how do you search for intelligence specifically? You could argue that the search of intelligence is rather difficult for creatures here on Earth too (where it should be the easiest). For example, Dolphins clearly communicate, they haven’t got technology, but does that preclude them from being intelligent?

    The above raises a number of issues when searching for ET. For example, they might not build technology at all, they might not communicate at all, we simply do not know what ET might be doing. SETI itself makes an assumption about intelligence being a civilization that is advanced enough to send radio signals out into space. But, as we have seen over the past few posts here on Centauri Dreams, those assumptions, while being valid for their purpose, might not be the only assumptions we could make.

    Hence, whether we search for technology, communications or solar manipulation, all of these search methods ultimately return to the search for intelligence beyond our planet. Therefore, in my mind, SETI is the appropriate acronym for all of these endeavors.

  • G. Cecil October 31, 2014, 8:46

    The ‘elephant in the room of the blind’ is surely dark matter (DM) and its non-EM interactions, vastly more common than our baryons. Bumps in galaxy rotation curves even in sub-maximal disks hint at DM clumping. Upcoming lensing surveys may map DM sufficiently densely through nearby near-face-on disks to begin to trace out more than DM radial structure (which is controversial). The ‘eerie silence’ may be because we monitor a communications medium (EM radiation) that is a quaint side-band to chattering DMers.

  • Alex Tolley October 31, 2014, 10:02

    @Christopher L. Bennett, Eric Hughes.

    While “starivore” sounds horrible (wincingly so?), it is a little pedantic to say that “astrovore” is wrong because it mixes Greek and Latin. Astr- is a Latin root derived from the Greek – “aster”. We do have the Latin “astronomia” and “ad astra”. The only difference is that the Latin “stella” refers to one particular star, while astr- is used as the plural. Star eater could therefore mean “eating a particular star” -> stellavore, or generically eating stars -> astrovore.

    So astrovore seems as OK to me as stellavore.

  • Etienne November 1, 2014, 5:14

    To Eric Hughes, Christopher Bennett, Alex Tolley. I agree. We should coin right neologisms, yes, but beautiful ones too. The first time I read “starivore”, I did not like it at all. “Stellavore” is better but I propose “stellivore” instead, it sounds more like “carnivore”, “herbivore” and “omnivore”

  • Anthony Mugan November 2, 2014, 7:11

    I sometimes think we need to consider the policy context for how we (or another ETC) might handle the discovery of an emerging civilisation. We always assume signals will either involve clear, intentional signals or our detection of unintended signals such as Dyson Spheres etc. There may be a third general area so far unexplored in any serious way of intentionally ambiguous signals.

    I am using ‘ambiguous’ in the military sense here…

    to give a specific example let us take a scenario many thousands of years from now in which we discover an intelligent culture at around the level of (say) the Bronze age somewhere on the far side of the galaxy. This unique occurrence would have to be considered at a government, policy, level over and above any purely scientific considerations.
    Hopefully we would decide to simply study them from afar. Much could be done by remote sensing but there would be gaps. The absence of any radio communication would make studying their languages problematic without sending robotic AI special forces down to the surface equipped with various surveillance technology. Direct sampling of flora and fauna would also be highly desirable etc.

    Such operations would be risky in the sense the could be discovered leading to culture shock etc. They would be infrequent, carefully planned and hopefully successfully unnoticed using the latest stealth technologies. It may not be possible to be totally invisible at all times however, and, let’s face, people make mistakes and things can go wrong….

    As the millennia pass the culture may get close to the point where they may be ready to begin space exploration themselves. This would raise an interesting policy question. Do we leave them totally alone until they bump into us (we aren’t just going to withdraw as they expand) or broach the subject now – even though their society doesn’t seem ready for it…?

    An intermediate strategy of ambiguous operations may be desirable in this situation. Evidence could remain below the threshold needed for formal recognition but would begin to trigger awareness in groups that use lower thresholds for evidence such as the intelligence community (an approach which we all know has its risks – the Iraq WMD fiasco being a recent example). By this time our own military would no doubt be quite interested in conducting probing operations (in the fashion of reconnaissance missions during the cold war) to test out capabilities, tactics etc. This could, if carefully planned, fit the bill nicely….

    Such an ambiguous strategy would create considerable problems for SETI (or SETT) researchers. Unless something went badly wrong the operations would not present the level of evidence needed for a scientifically valid result. Even the occasional ‘foul up’ would be isolated odd events without replication.

    How could we test for something like an ambiguous strategy…this would be very difficult. There has been work done (e.g Puthoff, 2010, JBIS – also on the Earthtec website) on what effects a close approach by a craft using specific models of exotic propulsion might produce. Data mining of possible candidate events might be useful, but all thoughts welcome…

    possibly TBC as I think a bit more about this!

  • Anthony Mugan November 4, 2014, 4:59

    In terms of testing for an ambiguous strategy.

    At first sight this might seem impractical – relying on chance mistakes – but there may be a way to make testable predictions. Again this is a tentative idea and any thoughts or critique welcome.

    The proposal assumes Arthur C Clark was on the right lines and an advanced ETC may be deploying technology which appears to be ‘impossible’ or to contradict the limits of what we would currently understand. The assumption here is that there may be one or more paradigm changes to go in our understanding of physics and perhaps other sciences such as biology.

    Using what might be termed an ‘inverse anthropic principle’ it may therefore be possible to generate testable predictions. This would however be a long term game.
    a) Identify a set of possible candidate events
    b) Identify common features that imply fundamental developments in physics (e.g. characteristics that imply a specific sort of exotic propulsion or something unexpected in the biological sciences, if the candidate events are not just nonsense). From these generate predictions about scientific or technological developments that would be implied by these observations (this is the inverse anthropic principle).
    c) Wait…possibly a long time alas
    d) If the predictions come to pass go back over the data for the successful candidates and test them as far as possible – can all natural / mundane scenarios be ruled out?

    Probably not even then conclusive, but a pattern may begin to emerge which could inform future developments. One implication is that the candidate events would need to have a large dataset to go with them of observations using a range of quantitative and qualitative methods – this probably restricts us to looking from the mid 20th Century onwards unfortunately (as I assume such events would be rare if the occur at all).

    I acknowledge the sociological problem this approach would raise as it gets perilously close to various ‘off limit’ subjects, but so be it.

    I’m not going to suggest any possible candidate events or possible predictions here – too soon and not appropriate at this stage, but any critique would be appreciated to help refine my thinking on this.

  • kzb November 4, 2014, 9:16

    The funny thing is, there actually ARE low surface brightness galaxies with high IR emissions relative to their visible luminosity (this is conventionally explained by dust)

    A population of IR-excess stars with optical luminosities well below normal for their mass WERE apparently discovered in the MACHO searches in the 1990’s. I’ve seen no follow-up to this.

    Lastly and maybe a bit tongue in cheek: there ARE quasars with measured parallaxes and proper motions. Explained away by lensing. But maybe the idea we are staring up the jet pipes of interstellar craft needs revisiting…

  • ljk November 4, 2014, 13:28

    See galaxy NGC 5907:



    Astronomers claim it is a spiral galaxy dominated by red dwarf suns.


  • NS November 4, 2014, 14:25

    I think it was Kardashev who said that we might already be integrating what is actually alien technological activity into our “natural” picture of how the universe works. Michio Kaku’s anthill next to the freeway scenario also comes to mind…

  • DJK November 5, 2014, 3:47

    To Eric Hughes, Christopher Bennett, Alex Tolley
    Gentlemen, I appreciate your comments.
    Be that as it may, the term “astrovore” contains two Greek words. These are:
    “Astro” (αστρο) from αστήρ=star and “vore” from βορα=food derived from the verb βιβρώσκω which means to ” eat” “devour”.
    (See also: http://www.perseus.tufts.edu/hopper/text?doc=Perseus%3Atext%3A1999.04.0057%3Aentry%3Dbora%2F )
    Thus, the term “astrovore” is based on purely Greek words.
    Anyway, enough of my Hellenic phraseology in a fascinating topic dealing with astroengineering!!!

  • Wojciech J November 5, 2014, 18:35

    “ljk November 4, 2014 at 13:28

    See galaxy NGC 5907:”
    Thanks ljk! I always forget the name of this galaxy! It’s one of the possible detections besides ambiguous Dyson Sphere’s results.

  • drs November 12, 2014, 1:01

    “Accretion disk would give a lot of energy cheaply, but it wouldn’t be the most efficient source of energy from a given amount of matter in the long term. The most efficient would have to be splitting the accretion mass into stars that are below the red dwarf threshold for buoyant convection in the star’s core. Why would such a civilization do hasty things as throw matter irreversibly into a black hole?”

    H->He fusion releases 0.7% of mass as energy. H->Fe might yield 1%. IIRC, black hole accretion discs release about 40% of mass as energy. Feeding mass very slowly and cooling into a black hole, then waiting for the Hawking radiation, should I think be 100% conversion, though rather slow. If you’ve got anything close to total conversion, then stars are quaint and obsolete. Of course, that means you might still want to dismantle them all for long term survival. Then again, you’re not losing that much from the fusion per se; what you want to avoid is having lots of mass blasted away or worse, tied up gravitationally in white dwarfs or neutron stars. And it might be hard to keep gas reserves from falling back together.

    So we might expect to see a galaxy of stellar lifted red dwarfs, not for people living around such stars but simply as trillion-year mass reserves to be fed later into the smaller number of conversion engines.