How to Measure the Power of Alien Civilizations Using the Kardashev Scale

We have yet to make contact with an extraterrestrial civilization. If they're out there — and surely they must be — we haven't the foggiest idea what they might be like. Or do we?

Given what we know about the universe and our own civilization, we should be able to make some educated guesses. And in fact, several decades ago, a Russian astrophysicist came up with a classification system to describe hypothetical aliens. Here's how the Kardashev Scale works.

Top image by Steve Burg.

The scale was devised by Nikolai S. Kardashev, a Soviet-era cosmologist who is still active today. Though he's 81, Kardashev works as the deputy director of the Russian Space Research Institute at Moscow's Russian Academy of Sciences. During the 1950s, while both his parents were in Stalin's slave labor camps, he became an astronomy student at Moscow University's Mechanics and Mathematics department. His primary interest was in astrophysics and the theoretic potential for wormholes, but he also shared a fascination with the search for extraterrestrial intelligence (ETIs).

It was around this time that Frank Drake launched Project Ozma — a pioneering attempt to locate ETI's by scanning the sky for radio emissions. Accordingly, Kardashev began to wonder if a good number of alien civilizations might be millions of years ahead of us, and if so, what their radio signatures might be like. Just how "loud," he surmised, could alien transmissions truly get?

This prompted Kardashev to write his seminal 1963 paper, "Transmission of Information by Extraterrestrial Civilizations." In it, he proposed a simple numbering system — from one to three — that could be used to classify hypothetical alien civilizations according to the amount of energy at their disposal. More specifically, he wanted to quantify the power available to them for their radio transmissions.

Today, Kardashev's scale has been expanded and re-interpreted to include more than just the capacity for communications technology. Astrobiologists and cosmologists now use the scale to simply describe the amount of energy available to an ETI for any kind of purpose. As a result, the scale is often used to speculate about the kinds of technologies and existential modalities that characterize advanced civilizations.

Here's how it works.

Kardashev Type I

In his paper, Kardashev wrote that a Type I civilization would be at a "technological level close to the level presently attained on the Earth, with energy consumption ~4 x 10¹⁹ erg/sec." That's about 4 x 10¹² Watts.

Kardashev's initial intention was to describe a civilization not too far removed from our own (again, for the purpose of rating its communicative capacities) — but one that has yet to exploit all of the solar system's resources (i.e. a pre-stellar ETI).

A Type I is typically associated with a hypothetical civilization that has harnessed all the power available to it on its home planet. As physicist Michio Kaku has said, it's a planetary scale civilization that can "control earthquakes, the weather — and even volcanoes." It will have taken advantage of every inch of space, and build "cities on the oceans."

For a civilization to attain Type I status, therefore, it needs to capture all of the solar energy that reaches the planet, and all the other forms of energy it produces as well, like thermal, hydro, wind, ocean, and so on.

More radically, Type I status would only truly be achieved once the entire planet is physically reconfigured to maximize its energy producing potential. For example, the entire mass of a planet could be reconstituted to take the form of a massive solar array to energize a civilization's power-hungry machinery.

Quite obviously, we are not a Type I civilization (at least not by this re-imagining of Kardashev's original description). Not even close. But Kaku predicts that we'll get there eventually, perhaps in a century or two.

But it could happen sooner if computational growth continues at its current breakneck pace (see Moravec, Kurzweil, and Bostrom). Hypothetically speaking, an artificial superintelligence (SAI) could get started in about three to four decades (either unilaterally, or by design).

Kardashev Type II

The next step is a big jump. And indeed, each increment of the Kardashev scale is an order of magnitude greater than the last.

Pre-dating Moore's Law and Kurzweil's Law of Accelerating Returns, Kardashev noticed that the rate of humanity's energy consumption was increasing steadily. He wrote, "...the annual increase in this energy expenditure is placed at 3-4% over the next 60 years, on the basis of statistical findings." Consequently, he predicted that, in about 3,200 years, "the energy consumption will be equal to the output of the Sun per second...i.e. 4 x 10³³ erg/sec."

This led him to speculate about a Type II civilization. For an ETI to reach K2, it would need to capture the entire energy output of its parent star.

The best way to achieve this, of course, is to build a Dyson Sphere.

Conjured by Freeman Dyson in 1959, this hypothetical megastructure would envelope a star at a distance of 1 AU and cover an inconceivably large area of 2.72 x 10¹⁷ km², which is around 600 million times the surface area of the Earth. The sun has an energy output of around 4 x 10²⁶ Watts, of which most would be available to do useful work.

It's difficult to predict when we ourselves could become a Type II, but physicist Stuart Armstrong says we could start the project in a few decades. And once underway, it would be subject to rapidly escalating construction speeds (fleets of robots would be powered by the newly-constructed portions of the Dyson shell).

With all this energy, an advanced civilization — probably one that's postbiological in nature — would use it to power its supercomputers and fuel its other endeavors (like interstellar colonization waves).

Kardashev Type III

Which leads to the next increment in the scale. Kardashev described a Type III like this: "A civilization in possession of energy on the scale of its own galaxy, with energy consumption at ~4 x 10⁴⁴ erg/sec." Needless to say, that's a tremendous amount of energy — somewhere between 10³⁶ Watts to 10³⁷ Watts (give or take a few).

Every inch of a K3 galaxy would be colonized, with every scrap of matter — and all its billions of stars — exploited for energy. From the perspective of an outside observer, a galaxy occupied by a K3 civ would appear completely invisible, save for the heat leakage which would register in the far infrared (around 10 microns in wavelength).

It would take a civilization anywhere from 100,000 to a million years to transition itself from a Type II to a Type III. Even at modest speeds, it wouldn't take a civilization very long (from a cosmological perspective) to completely colonize a galaxy.

From our vantage point, this would look like a hole in a galaxy, or an inexplicably large swath of open space.

Take the Boötes Void, for example, a huge chunk of the universe that's almost completely devoid of stars and galaxies. Speculatively speaking, this could be a large portion of the universe that has been overtaken by K3 civilizations.

 Interestingly, Fermilab's Richard Carrigan has argued that we should look for signs of extraterrestrial civilizations not in our own galaxy, but in neighboring galaxies. His idea is that we should look for civilizations that are transitioning from Type II to Type III. These colonization waves would look like a massive bubble that's spreading outwards from the originating star.

It's imaginable that a super-civilization would begin a wave of colonization that spread out to neighboring solar type stars from its home base. Each offshoot would "astro-form" the colonized planetary system by constructing a Dyson sphere around the host star.
Carrigan envisions seeing "Dyson bubbles" in nearby galaxies. These would be clusters of Dyson spheres that enclosed a grouping of stars colonized by a Type II Kardashev civilization. The logic is that after you've built a backyard fence you can start to conceptualize building the Great Wall of China and still hope to gain perspective on the process, Carrigan writes.
These would be detected as anomalous dark voids in a galaxy's disk. When these voids were observed in infrared light they would glow brightly with the heat radiation from the surfaces of Dyson spheres. This would show that they are not that simply voids where solar-type stars are conspicuously missing.

A good candidate for such a search would be the Andromeda Galaxy, which is only 2.5 million light years away. At most, we'd be glancing back a couple of million years into the past, which is not significant from a cosmological perspective.

What would an advanced civ do with all this energy? Well, if many futurists are to be believed, flipping one's and zero's. A Type II and III civilization may be completely based in digital substrate.

Kardashev Type IV? V?

Though Kardashev never went past a Type III, others have taken his idea to the next level. A Type IV would be an ETI (or merging groups of ETIs) that has harnessed all the power of a galactic supercluster, and a type V would — you guessed it — have the entire power of the universe at its disposal.

Unfounded assumptions?

While the Kardashev scale offers considerable food for thought, it is not without its problems.

First and foremost, and stating the obvious, no empirical evidence exists indicating the presence of K2 or K3 civilizations in our galaxy and/or galactic neighborhood. In fact, the Fermi Paradox — what's been dubbed "The Great Silence" — would indicate that civilizations never become migratory, thus making a Type III very unlikely. If Kardashev civilizations exist, we should expect to see large swaths of neighboring galaxies "disappear" from the visual spectrum — yet we do not.

We haven't found any Dyson spheres, either. But that doesn't mean they don't exist. Dysonian SETI is largely underway — an attempt to find the "gaps" in the stars.

Another problem with the Kardashev Scale is the assumption that advanced civilizations have an insatiable appetite for energy. No doubt, a K3 civ seems a bit excessive. It's not a stretch to suggest that a Type II civilization might be as far as these things go. Even a Type I for that matter. Ultimately, it all comes down to the consumptive needs of an "end stage" civilization — one that has successfully adapted to postbiological, post-SAI existence.

Alternately, civilizations may choose to avoid these trajectories, either to honor some kind of Prime Directive, or for self-preservational purposes.

Indeed, turning a galaxy into a massive supercomputer may be the last thing an advanced civilization wants to do. ETIs may have other desires and goals that preclude it from this kind of intergalactic imperialism.

But we don't know for sure. So in the meantime, let's be sure to keep listening and looking.