≡ Menu

The End of Super-Civilizations

Is energy consumption a good way to measure a civilization? The Soviet astronomer Nikolai Kardashev thought so, using it as the basis for his famous classification scheme. A Type I civilization could harness the energy resources of its home world, while a Type II could use its own star’s entire energy output. A Type III, the most exotic of all, could tap the energy of an entire galaxy, making it a plausible SETI target if we assume we can identify its exotic activities for what they were.

But some are questioning whether energy consumption is the best marker for looking at possible extraterrestrial cultures. Zoltan Galantai (Technical University of Budapest) notes that expecting vast energy use may simply be the marker of an adolescent technology, one that assumes all possible futures will look something like our own present extrapolated forward. He points out as well that there is no fast correlation between energy consumption and the spatial growth of a civilization.

If that one draws you up short (and it did me, for a time), think of it this way: a civilization might explore and populate its entire solar system without ever attaining Kardashev Type I status. For that matter, interstellar colonization might be feasible in some scenarios without becoming Type I. Consider a gradual expansion into the Kuiper Belt and Oort Cloud, followed by slow, millennial migration through Oort debris all the way to the cometary field that presumably surrounds the Centauri stars. Perhaps an entire galaxy might be colonized without achieving Type III, a defensible position given our lack of examples.

Galantai suggests an alternative metric for civilization’s growth, one based on risk. It’s much closer to home, because in its early stages it applies to civilizations that have not left the surface of their home planets. Galantai Type I is unable to survive a local disaster (think of the Anasazi, unable to overcome the loss of their forestry resources). Type II is a culture unable to survive a regional catastrophe, such as a super volcano or an ice age. Type III is threatened only by global disasters, such as climatic collapse or an impact by a sufficiently large asteroid or comet (or, of course, a combination of such events).

And so on up the scale to Galantai Type IV, in which we’ve colonized the Solar System but are still vulnerable to such things as a nearby supernova explosion, and finally to Type V, in which we move out into the galaxy, becoming all but indestructible by any one event but subject to species individuation as the vast distances between the stars separate us into separate enclaves. “Not homo sapiens,” writes Galantai, “but their descendants will spread, and the traditional idea of ‘civilization’ will become meaningless.”

There is even a theoretical Type VI, in which a civilization has colonized its entire universe. Here the author quotes Freeman Dyson’s 1979 question (in an paper in Reviews of Modern Physics), asking whether “…converting matter into radiation and causing energy to flow purposefully on a cosmic scale, we could break open a closed universe and change the topology of space-time.” Galantai’s take:

Supposing that the known laws of physics will not change in the future, it seems to be enormously improbable that even a super civilization will be able to connect the distant parts of our Universe and will be able to harmonize its endeavors to use the energy sources. But there is nothing to prevent a civilization populating galaxies after galaxies if they have enough time, while regarding their technologies, there is not an inevitable difference between those who conquered ‘only’ a galaxy and those who inhabited the entire Universe.

It’s quite an interesting notion, one that suggests not only that galaxy-spanning cultures are unlikely after Kardashev Type I, but that an advanced civilization may well conquer its own solar system and nearby stars before ever reaching the technical level needed to construct a Dyson sphere. The upshot: “An advanced civilization has to expand (as [Krafft] Ehricke points out), unless they want to die out because of a planetary cataclysm. So it is reasonable to use a scale which is based on an intelligent race’s spatial expansion, since it can show their ability to survive a disaster.”

The paper, available online, is Galantai, “After Kardashev: Farewell to Super Civilizations,” and Centauri Dreams‘ take is that whether the author is right or wrong, it’s about time someone questioned the ‘inevitability’ of large scale structures as the markers of advanced civilizations. The universe is too curious a place to lock ourselves into any single future, no matter how entertaining or plausible our preconceptions.

The complete Dyson reference above is “Time Without End: Physics and Biology in an Open Universe,” Reviews of Modern Physics, Vol. 51, No. 3 (July 1979), pp.447-460, with abstract here.

Comments on this entry are closed.

  • hiro February 1, 2007, 15:10

    It might be possible for human beings to expand their habitats to the entire galaxy. However, the idea about changing the topological structures of the universe sounds more like friction to me.

  • Darnell Clayton February 1, 2007, 15:59

    Colonizing the Universe (Type VI Civilization) would be an incredible (yet eternal) feat for our race.

    Although we are currently on the verge of colonizing our own solar system (or beginning to at least) there will always be one potential threat that could eliminate the entire race, no matter how many billions are born: war.

    Think of Star Wars or Star Trek’s dominion war. Hopefully, by the time we begin exploring our solar playground, we will be beyond such things, but I am not holding my breathe.

  • Ron S February 1, 2007, 20:16

    If the objective is survival by spreading the civilization’s bets to two or more locations, which is cheaper/easier: locate, travel to, modify and inhabit an exoplanet; same for a ‘dead’ local planet (think Mars); or, construct and inhabit an artificial and self-contained ecosystem that does not depend on any planet or star?

    Anyone ever work through this?

  • george scaglione February 2, 2007, 16:12

    good comments but one great problem i see in colonizing such a large area is simply this – how could we possibly do it? ! i mean consider: distances of thousands or hundreds of thousands or millions of LIGHT YEARS, that is from our present point of view sitting here in february of 2007 for all PRACTICAL purposes an infinite distance !! i mean we do think about traversable wormholes warp drive etc. but ,how far are we away from those?no offense to anybody at all but what do we really have? old space shuttles with which i think we are pushing our luck even now! or in the case of other spacefaring nations,space capsules which more closely resemble mercury or gemini . colonize the galaxy or universe? respectfully… i do not think so. your friend george ps but having said all that – i dont mean either that vast progress will not eventually be made ! breakthroughs may come in new branches of physics which we do not yet suspect! all the best g

  • Brian Wang February 2, 2007, 16:41

    I see various flaws in the first papers reasoning.

    The energy scale is worthwhile as one method categorization. Just because we do not understand how certainly things are acheived does not mean we can use a particular measurement scale.

    With equal nanotech, quantum and energy efficiency for using energy then in general more energy is more capability. There could be modifiers in terms of usable force or effective force.

    The discussion about modifying space time to achieve a universal civilization can also be applied to create KIII civilizations. Modify the energy output of the star so that its energy output is a billion times greater but is used up sooner. Use various resources on a stellar scale to achieve greater power levels. Then the civilization could be localized and yet still have higher power levels. After they use it up they would use other stars. Stars could be shifted or a region of high star density could be chosen. Super-civilizations might manipulate black holes.

    What is the threat the lower energy and dispersed civilization could face that could still destroy them ? Not just natural disasters. A higher energy density and more technologically advanced civilization. The higher energy density civilization could come and apply more concentrated force and roll-up the lower energy density and dispersed civilization. Example, 168 spanish versus the Incas or the conquest of N America.

    Energy and size are not the only metrics there can also be information size. Ability to master various forces.

    But it is clear that we do not know what those advanced civilizations will be capable of or how they will operate and how detectable they are. If we believe that we are approaching a technological singularity. The super-civilizations could be ten singularities away. Another measure that can be applied in hindsight. When your tech is advanced enough you can layout the different levels of tech. Animals, primitive, agrarian, industrial, atomic/computer, nanotech/biotech, fusion/quantum/interplanetary etc…

    Since we do not know, but can apply relatively open ended scales as classifications (but not predictions) it is still useful.

  • ljk February 2, 2007, 16:48

    It Takes a Giant Cosmos to Create Life and Mind

    James N. Gardner


    A new book, The Intelligent
    Universe, proposes that the universe
    might end in intelligent life, one
    that has acquired the capacity to
    shape the cosmos as a whole.


    Foreward to The Intelligent Universe

    Ray Kurzweil



    The explosive nature of exponential
    growth means it may only take a
    quarter of a millennium to go from
    sending messages on horseback to
    saturating the matter and energy in
    our solar system with sublimely
    intelligent processes. The ongoing
    expansion of our future
    superintelligence will then require
    moving out into the rest of the
    universe, where we may engineer new
    universes. A new book by James
    Gardner tells that story.


  • Adam February 3, 2007, 1:42

    Hi Larry

    Part of Kurzweil’s “vision” is that we’re Alone in the Cosmos and that exponential growth will cause us to convert the Universe into computronium after we circumvent light-speed via wormholes.

    Dunno. Assumes traversible wormholes and that we’re NOT the first intelligences in the Cosmos. Both might be wrong.

    Hey Larry, why don’t you have a blog instead of huge email mail-outs on all the news you’ve trawlled?


  • andy February 3, 2007, 5:43

    I do like the way that nanotech and AI have become the new all-powerful gods. The powers ascribed to them by the transhumanist/Singularity movements suggest that there is nothing you cannot do once you have them. I doubt nanotech/AI etc. will be able to do anything like what the Singularity people suggest they will.

  • Administrator February 3, 2007, 9:16

    Hey Adam, knock it off. If Larry starts his own blog, I’ll have lost one of my best resources! For that matter, now that you’ve got Crowlspace.com up and running, I hope that doesn’t mean you’ll be scarce around here either.

  • Adam February 3, 2007, 15:22

    Hi Paul

    No chance of that. Though I might end up repeating what I say here back over there – albeit slightly edited. I realised the other day just how much stuff I post on other people’s blogs… ;-)


  • bigdan201 February 6, 2007, 17:32

    the kardachev scale is alright, but it really is a little simplistic. it might not be in a civs best interest to farm an entire galaxy for energy.. although id expect individual stars to be used that way.

    there are better markers than energy consumption.. ability to survive and continue is definitely one of them

  • Timothy J Mayes March 8, 2007, 15:52

    How much energy you have to expend determines the potential for
    large scale beamed power propulsion, and the potential for breeding
    and storing large amounts of antimatter to fuel antimatter rockets.

  • ljk June 19, 2007, 11:17

    Atom trap is a step towards a quantum computer

    NewScientist.com news service June 17, 2007


    A device that can hold hundreds of
    atoms in a 3D array, and image each
    one individually, may be an
    important stepping stone towards
    developing a quantum…


    Nanotubes could make superstrong chips

    NewScientist.com news service June 17, 2007


    In a new patent, Intel reveals how
    nanotubes’ strength and
    heat-dissipating properties can be
    used to to extend chip life by
    reducing thermal stress and
    increasing resistance to impact


    Game designers test the limits of artificial intelligence

    Boston Globe June 17, 2007


    Videogame developers are attempting
    to create more intelligent, more
    realistic characters, writing
    software that better mimics human
    intelligence. “A lot of the most
    interesting work in artificial
    intelligence is being done by game
    developers,” says Bruce Blumberg,
    senior scientist at Blue Fang Games
    and formerly a professor at MIT’s
    Media Lab….