From io9 via SETI: “How SETI Will Understand Messages Broadcast by an Alien Intelligence”

SETI Institute


George Dvorsky

Karl V. Jansky Very Large Array

Imagine the day when we finally receive a signal from an extraterrestrial intelligence, only to find that there’s a message embedded within. Given that we don’t speak the same language, how could we ever hope to make sense of it? We spoke to the experts to find out.

Communication with Extraterrestrial Intelligence, aka “CETI”, is the branch of SETI concerned with both the transmission and reception of messages between ourselves and an alien civilization. Scientists have been trying to detect signals from an extraterrestrial intelligence (ETI) since the 1960s, but haven’t found anything.

Allen Telescope Array
SETI Institute’s Allen Telescope Array

SETI@home, a public distributed computing project running on BOINC software and using data from the Arecibo Observatory
Arecibo Observatory
Arecibo Observatory


At least not yet. If and when we do receive a signal, whether it be an intercepted transmission or a deliberate attempt to get our attention, we’ll be tasked with deciphering an alien message. It could prove to be a monumental task, but it’s a problem with no shortage of solutions.

Natural or Unnatural Signals?

The first challenge will be to recognize an incoming alien signal. This may prove easier said than done.

When pulsars were first discovered, for example, their eerily precise spectral flashes convinced some scientists that we were actually looking at some sort of alien beacon. And in 1977, the 72-second-long Wow! signal was likewise interpreted as extraterrestrial in nature. More plausibly, it was just a natural, continuous signal, or some human-instigated artifact.


These episodes aside, most SETI researchers agree that an alien signal will be unambiguous.

“Due to the random motions of the particles that are ultimately at the source of natural electromagnetic emission, these emissions tend to get spread out in frequency or in time,” says Andrew Siemion, a PhD candidate in astronomy at SETI-Berkeley. “Technology, on the other hand, is capable of producing very fine time and frequency structure. We can use this fine structure to distinguish between natural and ‘unnatural’ sources.”

Siemion says it’s important to keep in mind that our knowledge of physics and the cosmos isn’t complete, and it’s conceivable that there are some natural processes that could mimic the types of signals we look for in SETI experiments.

“But discovering these would be great as well,” he told io9.

According to Douglas Vakoch, Director of Interstellar Message Composition at the SETI Institute, we should actively look for signals that stand out from the cosmic static as distinctly artificial.

“The radio signals created by nature are spread out on the radio dial,” he says. “We’re looking for narrowband signals at one place on the radio dial.”

Overcoming the Language Barrier

Given that an ETI would most assuredly “speak” a different language than any found on Earth, it’s fair to ask how we could ever hope to overcome such a barrier.

How SETI Will Understand Messages Broadcast by an Alien Intelligence

Linear A tablets. (Credit: University of Houston/CC BY 3.0)

And indeed, linguists are already struggling with this issue as it pertains to Linear A—an undeciphered writing system used in ancient Greece. It’s not immediately obvious if we’ll ever crack the code of this ancient language. Likewise, if we ever intercept an unintentional alien message, say something akin to a radio or television transmission, we may never be able to decipher the message, save for any visual or acoustic information gleaned from the broadcast.

But it would likely be a different story if the message was intentional.

“If an advanced civilization did want to communicate with us, they would probably choose to base their communication on something we have in common, such as the fact that we live in the same physical universe,” says Siemion. “They might use the properties of astrophysical objects, like pulsars, quasars or the shape of our galaxy, as a first step at teaching us their language.”

Siemion says that an advanced ETI, if they were fairly close to us, say within 40-50 light-years, might actually know quite a lot about us. They may have already taken it upon themselves to decipher parts of our early radio and television transmissions. If this is the case, Siemion says it may be very easy for them to communicate with us in a way we understand.

Speaking in Math

Alternately, we can forgo arbitrary symbolic communication altogether and use the logic of math as a communication medium. As Vakoch says, the ability to communicate mathematics will allow aliens to communicate virtually anything that can be quantified.

“One of the most basic parts of mathematics is counting,” Vakoch told io9. “When we think of counting, we usually imagine counting 1, 2, 3, and so on. But there are other ways to count as well.”

For example, Vakoch says we could tell an ETI about the Fibonacci series by starting with the simplest numbers, zero and one, and then add them together, and then repeat the process over and over, adding the last two numbers in the series. Zero plus one is one, one plus two is two, and so on until the Fibonacci series is obvious.

Spencer Greenberg from Ask a Mathematician says it shouldn’t be too hard for an ETI to develop a signal that, if we received it, could tell us it was created by another intelligent life.

To understand why, Greenberg considers how we ourselves might construct a signal if we wanted aliens to notice that we’re intelligent. To that end, he devised a (somewhat oversimplified) approach that assumes an ETI would have developed the notions of binary encodings of integers (which is by no means an overreaching assumption).

Talking in code: Greenberg says we could pulse our signal by emitting a relatively high frequency, and at other times emitting a relatively low frequency. Each high section of our signal could represent the digit 1, and each low section, the digit 0. “With this mechanism in place, it’s easy to transmit in binary,” says Greenberg.

By sending out pulses in binary, Greenberg says we could let the receiver know how many bits we’re using to represent a single number. After settling on the number of bits per group (e.g. 16), we could communicate our system by simply starting the message off with counting.

So for instance, if we wanted to signal that we are treating groups of 16 bits as a single number, we could transmit all the binary numbers from 0 to 65,535 in order, each of which would be represented by 16 binary digits. Therefore, our transmission would start 0, 1, 2, 3, 4, and so on, which in binary, with 16 bits per number, would be:

0000000000000000, 0000000000000001, 0000000000000010, 0000000000000011, 0000000000000100, etc.

Greenberg says that instead of sending each of these numbers a single time through, we might actually want to send each sequence of 16 bits a fixed number of times in a row (say, 100 times each) to provide for redundancy. That way if our signal gets corrupted in transit, it’s still easy to correct any mistakes that are introduced.

It should be obvious to an alien receiver that we are sending digits in groups of 16, with each 16 digit block representing a number. That would allow us to transmit any number we please (so long as it’s between 0 and 65,535) by representing it in the next 16 digits of our binary code.

At that point there would be plenty of options for what to send to prove that we’re reasonably intelligent. We could transmit all the prime numbers from 2 up to 65,521. We could also send triplets of numbers where the third number in the triplet is equal to the first two multiplied together, or we could send pairs of numbers that are twin primes. We could even convey mathematical formulas by creating special symbols like an equal sign.

Other ways to Communicate


In 1974, scientists transmitted a message into space consisting of 1,679 bits, arranged into 73 lines of 23 characters per line. Called the Arecibo Message, it consisted of the Arecibo telescope itself, our Solar System, DNA, a stick figure of a human, the biochemicals of terrestrial life, and other things.

Such messages aren’t perfect or overly sophisticated, but they can convey simple concepts, like our location relative to our Sun, and our physical appearance. Clearly, an ETI could send or transmit a similar pictorial message.

The Pioneer message: So simple even an alien could understand it. (Credit: NASA)

Mathematics could also be used to send algorithmic messages. These systems, such as CosmicOS and logic gate matrices, use a small set of math and logic symbols to form the basis of a simple programming language that an alien receiver could conceivably run on a virtual machine. Algorithmic messages, if complex enough, could actually be used to convey advanced concepts—and even signs of intelligence—if run on a sufficiently advanced computer system.

Binary logarithms represent a microarray of expression data for 8,700 mouse genes. (Credit: Louis M. Staudt/National Cancer Institute/Public domain)

As for natural language processing, we’re still a long ways off from having the ability to make sense of arbitrary symbols. But Laurance Doyle, a member of the SETI institute, is using math to do just that. Doyle is trying to use information theory—a branch of math that looks at the structure and relationships of information—to separate binary code from random 0’s and 1’s. The idea is to find linguistic substance within undefined symbols, whether they be written or oral, and an associated grammar and syntax. Fascinatingly, Doyle’s work is being applied by marine biologists in an effort to crack the dolphin language code.
Hmm, What to Talk About…

Assuming that an alien civilization wants to reach out to us and say “hello”, it’s reasonable to wonder what else they might want to say to us.

Vakoch says the most important thing an alien civilization could communicate to another is their intention to make contact.

Siemion says ETIs might offer tips on existential dangers to humanity, both intrinsic threats, like biological weapons or artificial superintelligence, and extrinsic threats, such as asteroid impacts or a looming nearby supernova.

“Some people believe that technological progression and increased altruistic tendency go hand-in-hand—that is, that the more advanced a civilization becomes, the friendlier they get,” says Siemion. “Personally, I don’t think we have any evidence of that. In fact, I think we have evidence that the contrary may be the case.”

The Risk Factor

Indeed, this could be a rather dangerous exercise. We run the risk of receiving and translating a malign message, such as a trojan horse that contains a kind of computer virus, or the seeds of our own destruction.

And as we’ve written before at io9, the effort to deliberately transmit messages to aliens—an endeavor known as METI—needs to be seriously re-considered. Our efforts to reach ETIs could alert an evil force to our presence.

Alien Outpost

“Sending messages of our own creation to try to make any possible aliens aware that we exist, is an incredibly risky proposition,” says Greenberg. “Sure, they might be friendly, but then again, they might not—and that’s a big risk to take,” says Greenberg. “Attempting to wake up a force more powerful than ourselves that we do not understand is simply not a good plan.”

Vakoch says that concerns of alerting an ETI to our presence are too late.

“Any civilization with the ability to travel between the stars would already know we’re here from our accidental leakage radiation,” he says. “So a sufficiently advanced extraterrestrial might already have picked up ‘I Love Lucy.’ But they still don’t know that we’re attempting to communicate with them. That’s the most important reason for us to transmit powerful, intelligible signals to other stars—to let any intelligence out there know we’re ready to make contact.”

See the full article here.

Please help promote STEM in your local schools.


Stem Education Coalition

SETI Institute – 189 Bernardo Ave., Suite 100
Mountain View, CA 94043
Phone 650.961.6633 – Fax 650-961-7099
Privacy PolicyQuestions and Comments