Sending a probe to every star in the galaxy would mean building hundreds of billions of spacecraft. But what if you only had to build one? A von Neumann probe would travel to another star system, mine asteroids or moons for raw materials, build copies of itself, and send those copies on to further stars. Each new generation would do the same. The numbers would grow exponentially, and the galaxy would be explored with almost no further effort from home.
Self-reproducing machines
The idea takes its name from the Hungarian-American mathematician John von Neumann. In lectures at the University of Illinois in 1949 he worked out, in abstract mathematical form, how a machine could build a copy of itself, including a copy of its own instructions. His work was completed by Arthur Burks and published in 1966 as Theory of Self-Reproducing Automata. It anticipated, in mechanical terms, how living cells copy themselves using DNA.
A design
In 1980 Robert Freitas published a detailed design for a self-reproducing interstellar probe, based on the British Interplanetary Society’s Project Daedalus starship study of the 1970s. His probe would travel at 12% of the speed of light and land a seed factory that, over about 500 years, would grow into a full industrial base, mine fuel from a gas giant, and then build a new probe about every 500 years.
Even at modest speeds, the exponential spread is fast on cosmic timescales. Estimates suggest that probes travelling at a tenth of light speed could visit the whole Milky Way in as little as half a million years. In 2013 Arwen Nicholson and Duncan Forgan showed that using stars’ gravity for slingshots could cut exploration times by up to a thousand times.
Where are they?
That speed is exactly the problem. The galaxy is more than 13 billion years old. If any civilisation had built von Neumann probes, they should have reached every star system long ago, including ours. In 1980 the physicist Frank Tipler used this argument to claim that extraterrestrial intelligence does not exist. Others reply that civilisations might choose not to build them, or that we have not looked carefully enough. The question sharpens the Fermi paradox: if the galaxy could be filled so easily, why does it look empty?