Chapter 12: The Principle of Computational Equivalence

Section 10: Intelligence in the Universe

[History of] SETI

First claims of extraterrestrial radio signals were made by Nikola Tesla in 1899. More widely believed claims were made by Guglielmo Marconi in 1922, and for several years searches were done—notably by the U.S. military—for signals presumed to be coming from Mars. But it became increasingly accepted that in fact nothing beyond natural radio emissions such as whistlers (see note above) were actually being detected.

When galactic radio emission was first noticed by Karl Jansky in 1931 it seemed too random to be of intelligent origin. And when radio astronomy began to develop it essentially ignored extraterrestrial intelligence. But in 1959 Giuseppe Cocconi and Philip Morrison analyzed the possibility of interstellar radio communication, and in 1960 Frank Drake used a radio telescope to look for explicit signals from two nearby stars.

In 1965 a claim was made that there might be intensity variations of intelligent origin in radio emission from the quasar CTA-102—but this was quickly retracted. Then in 1967 when the first pulsar was discovered it was briefly thought that perhaps its precise 1.33730113-second repetition rate might be of intelligent origin.

Since the 1960s around a hundred different SETI (search for extraterrestrial intelligence) experiments have been done. Most use the same basic scheme: to look for signals that show a narrow band of frequencies—say only 1 Hz wide—perhaps changing in time. (The corresponding waveform is thus required to be an almost perfect sinusoid.) Some concentrate on specific nearby stars, while others look at the whole sky, or test the stream of data from all observations at a particular radio telescope, sometimes scanning for repetitive trains of pulses rather than single frequencies. The best current experiments could successfully detect radio emission at the level now produced on Earth only from about 10 light years away—or from about the nearest 10 stars. The detection distance increases like the square root of the signal strength, covering all 10^11 stars in our galaxy when the signal uses the total power output of a star.

Most SETI has been done with specially built systems or with existing radio telescopes. But starting in the mid-1990s it became possible to use standard satellite receivers, and there are now plans to set up a large array of these specifically for SETI. In addition, it is now possible to use software instead of hardware to implement SETI signal-processing algorithms—both traditional ones and presumably much more general ones that can for example pick out much weaker signals.

Many SETI experiments look for signals in the so-called "water hole" between the 1420 MHz frequency associated with the 21 cm line of hydrogen and the 1720 MHz frequency associated with hydroxyl (OH). But although there are now practical constraints associated with the fact that on Earth only a few frequency regions have been left clear for radio astronomy I consider this to be a remarkable example of reliance on details of human intellectual development.

Already in the early 1960s it was suggested that lasers instead of radio could be used for interstellar communication, and there have been various attempts to detect interstellar optical pulses. Other suggested methods of communication have included optical solitons, neutrinos and as-yet-unknown faster-than-light quantum effects.

It is sometimes suggested that there must be fundamental limits to detection of radio signals based on such issues as collection areas, noise temperatures and signal degradation. But even existing technology has provided a steady stream of examples where limits like these have been overcome—most often by the use of more sophisticated signal processing.

From Stephen Wolfram: A New Kind of Science [citation]