Chapter 12: The Principle of Computational Equivalence

Section 10: Intelligence in the Universe

Messages to send [to extraterrestrials]

The idea of trying to send messages to extraterrestrials has existed since at least the early 1800s. The proposed content and medium of the messages has however steadily changed, usually reflecting what seemed to be the most significant human achievements of the time—yet often seeming quaint within just a few decades.

Starting in the 1820s various scientists (notably Carl Friedrich Gauss) suggested signalling the Moon by using such schemes as cutting clearings in a forest to illustrate the Pythagorean theorem or reflecting sunlight from mirrors in different countries placed so as to mimic an observed constellation of stars. In the 1860s, with the rise of telegraphy, schemes for sending flashes of light to Mars were discussed, and the idea developed that mathematics should somehow be the basic language used. In the 1890s radio signals were considered, and were tried by Nikola Tesla in 1899. Discussion in the 1920s led to the idea of sending radio pulses that could be assembled into a bitmap image, and some messages intended for extraterrestrials were probably sent by radio enthusiasts.

There is a long history of attempts to formulate universal languages (see page 1181). The Lincos language of Hans Freudenthal from 1960 was specifically designed for extraterrestrial communication. It was based on predicate logic, and attempted to use this to build up first mathematics, then science, then a general presentation of human affairs.

When the Pioneer 10 spacecraft was launched in 1972 it carried a physical plaque designed by Carl Sagan and others. The plaque is surprisingly full of implicit assumptions based on details of human intellectual development. For example, it has line drawings of humans—whose interpretation inevitably seems very specific to our visual system and artistic culture. It also has a polar plot of the positions of 14 pulsars relative to the Sun, with the pulsars specified by giving their periods as base 2 integers—but with trailing zeros inserted to cover inadequate precision. Perhaps the most peculiar element, however, is a diagram indicating the 21 cm transition in hydrogen—by showing two abstract quantum mechanical spin configurations represented in a way that seems completely specific to the particular details of human mathematics and physics. In 1977 the Voyager spacecraft carried phonograph records that included bitmap images and samples of spoken languages and music.

In 1974 the bitmap image below was sent as a radio signal from the Arecibo radio telescope. At the left-hand end is a version of the pattern of digits from page 117—but distorted so it has no obvious nested structure. There follow atomic numbers for various elements, and bitvectors for components of DNA. Next are idealized pictures of a DNA molecule, a human, and the telescope. All these parts seem to depend almost completely on detailed common conventions—and I suspect that without all sorts of human context their meaning would be essentially impossible to recognize.

In all, remarkably few messages have been sent—perhaps in part because of concerns that they might reveal us to extraterrestrial predators (see page 1191). There has also been a strong tendency to make messages hard even for humans to understand—perhaps on the belief that they must then be more scientific and more universal.

The main text argues that it will be essentially impossible to give definitive evidence of intelligence. Schemes that might however get at least some distance include sending:

• waveforms made of simple underlying elements;

• long complicated sequences that repeat precisely;

• a diversity of kinds of sequences;

• something complicated that satisfies simple constraints.

Examples of the latter include pattern-avoiding sequences (see page 944), magic squares and other combinatorial designs, specifications of large finite groups, and maximal length linear feedback shift register sequences (see page 1084). Notably, the last of these are already being transmitted by GPS satellites and CDMA communications systems. (If cases could be found where the sequences as a whole were forced not to have any obvious regularities, then pattern-avoiding sequences might perhaps be good since they have constraints that are locally fairly easy to recognize.)

Extrapolation of trends in human technology suggest that it will become ever easier to detect weak signals that might be assumed distorted beyond recognition or swamped by noise.

Image Source Notebooks:

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