The search for extraterrestrial intelligence is a scientific investigation into possible communication signals originating from an extraterrestrial civilisation. It originated in searches and subsequent discussions between a few interested individuals around forty years ago. These early forays resulted in more systematic searches for microwave signals, and also in the proposal of Drake’s Equation, a method for estimating the number of communicating extraterrestrial civilisations. The searches have been centred on a certain part of the microwave spectrum, in ever-narrower bands as technology increases to allow it. They can be either targeted, allowing greater sensitivity, or sky surveys which increase the area searched though with less precision. Other methods have been proposed, some of which consider the technology that an advanced civilisation may have developed, while others are based on our own progression. Search programs may be categorised as either dedicated, directed or shared. Current programs, due to lack of funds, generally fit into the directed or shared category, though they still manage intensive searches. One of the most innovative is the SETI@home program, which utilises volunteer’s home computers to process data. Other programs such as Optical SETI are developing into exciting new search methods which may supercede the microwave searches currently underway. The problems facing the search for extraterrestrial intelligence are manifold, but to date they have been managed well and current interest levels suggest a bright future.
The scientific search for extraterrestrial intelligence (SETI) has its origins in a seminal paper by Giuseppe Cocconi and Philip Morrison in 1959, called ‘Searching for Interstellar Communications’, in which they argued for the use of microwave communication between interstellar civilisations. At much the same time (and independently) Frank Drake, a young radio astronomer, conducted ‘Project Ozma’: a single-channel scan in the direction of two nearby sun-like stars, using an 85-foot antenna. Interest continued to grow after these initial excursions into ‘alien territory’, with the Soviet Union and the United States of America taking turns at leading the search for extraterrestrial intelligence. The hopes for success grew in the 1970’s after a comprehensive analysis of SETI by the NASA-commissioned ‘Project Cyclops’ (History of SETI 1999). Continuing advances since then in the technology required for searching has resulted in a rapidly evolving field in which the chances of success seem to be improved by an order of magnitude each year. The various programs currently running are, however, for naught if there is no one out there. Estimates of the number of interstellar civilisations which are capable of broadcasting a signal, are therefore an important element of the search. The accepted equation for estimating this value is called ‘Drake’s Equation’.
In 1961 Frank Drake proposed an equation for expressing the number (N) of observable civilisations in the Milky Way galaxy (Billingham & Tarter 1993, p. 248):
N = R * fp * ne * fl * fi * fc * L
R is the rate of star birth in our galaxy,
fp is the fraction of stars that have solar systems,
ne is the average number of Earth-like planets in each solar system,
fl is the fraction of those planets on which life forms,
fi is the fraction of life-bearing planets which produce an
fc is the fraction of intelligent species that are capable of
L is the average lifetime of a communicating civilisation (in years).
Estimates of N by experts, based upon this equation, vary from just one (we are alone) to thousands, if not millions. The reason for this huge discrepancy is the difficulty inherent in quantifying each element of the equation. While some are now reasonably certain (for example, R is now generally thought to be about 1, although it may have been as high as 5 in the early history of the galaxy (Schilling 1999), others are not so easy. Examining fi exposes the problem – we currently have only one example: ourselves. Some believe that humanity is nothing special, and that the value is therefore close to one; others think that we are the product of numerous coincidences, and that the value is almost 0. Similarly, pessimists may say that in view of our own self-destructive tendencies L may be quite small; others believe that an advanced civilisation may be able to continue for millions of years. Thus it is evident that each ‘unknown’ in the equation is capable of either creating a bottleneck, which substantially reduces N, or opening a floodgate, which improves the chances that we are not alone. It is important to note that this figure is not an estimation of the number of planets with life on them, or even the number of intelligent species. N is equivalent to the number of intelligent civilisations capable of communicating across interstellar distances (Billingham & Tarter, p. 248). For a recent (2003) debate on the Drake Equation, see this article on Astrobio.net. Though the Drake Equation remains a theoretical quandary, the question of whether we are alone or not may be answered in the meantime by a more practical method: listening.
The most prevalent strategy used in SETI to this point has been the scanning of the microwave radio spectrum for narrow-band signals. The band of frequencies from 1 to 10 gigahertz has the least amount of interference from both cosmic background noise and our own atmosphere, and hence has been the focus of search strategies (Horowitz et al. 1986, p. 526). Narrow-band signals are searched for because they are characteristic of an artificial transmission: the narrowest naturally occurring microwave frequency is about 300 hertz, produced by interstellar masers. Therefore searching with a higher resolution than this is the best way to detect a signal from an extraterrestrial civilisation. Also, the narrower a signal is, the more efficient it is to send, hence narrow-band signals would seem the obvious choice for broadcasting (Billingham and Tarter 1993, p. 262). However, searching for such narrow frequency bands amongst a 10-gigahertz spectrum (which, it must be remembered, is also only a small portion of the overall microwave spectrum) is a huge task. To search the whole spectrum in 1 hertz bands would, for example, take 10 billion channels per point source observed. For this reason, many searches have centred on certain ‘magic frequencies’. The most predominant is the area around 1.42 gigahertz, the emission line for hydrogen – considered by many to be the most probable ‘standard interstellar frequency’ (MacRobert & LePage 1999). As search technologies have improved the spectrum covered has increased. The area from 1.42 up to 1.64 gigahertz (the emission line for Hydroxyl, OH), known as ‘the waterhole’ due to it’s boundaries (H and OH = H2O), has recently been the focus of some searches (Hoang-Binh 1985, p. 493).
Even with the advances in technology, it is obvious that large parts of the spectrum remain unchartered. Unfortunately, scanning the whole spectrum at this point in time can only be achieved through a wide-band survey, which offers a lower chance of success. Only through further improvements in technology will a narrow-band search of the entire spectrum be remotely feasible. For this reason searches are separated into sky surveys and targeted searches (Cullers et al 1985, p. 38). Sky surveys are methodical sweeps of the entire sky visible from each telescope location. If a signal is strong enough and in the correct frequency band, this search should pick it up. Targeted searches concentrate on certain points in the sky – the ‘good suns’, some 1000 stars similar to our own sun (type F, G and K) which lie within 100 light-years of Earth (Billingham and Tarter 1993, p. 264). Obviously, the targeted searches can be conducted much more thoroughly than sky searches, however, they rely on signals coming only from these points. Detected signals themselves could be of two different types. Either they have been intentionally sent to us by an extraterrestrial civilisation, expressly designed to get our attention, or they could be ‘leaked’ (that is, we have eavesdropped on them). The Earth is currently leaking signals itself, the most powerful being military radar (Papagiannis 1985, pp. 269-270) – some radio and television transmitters are also broadcasting strong signals (thankfully this is the carrier signal, not the content). SETI aims to pick up either of these two types of signal originating from an extraterrestrial civilisation by listening to as many frequency bands as is possible. Overall however, microwave scans still constitute the proverbial ‘needle in a haystack’ search, although continual advances in technology and processing may soon make this less of an issue.
Other methods of detecting extraterrestrial civilisations have been proposed. In the 1960’s Freeman Dyson raised the possibility of advanced civilisations harnessing enormous amounts of power through the use of so-called ‘Dyson Spheres’ (Dyson 1963, pp. 111-113). He hypothesised that the mass of a planet such as Jupiter could be distributed in a spherical shell revolving around the sun at 2 AU, which would be capable of exploiting the solar radiation striking its inner surface. Working upon this hypothesis Dyson called for SETI to be extended to the infrared range in an attempt to detect such a construction, a proposal which the Soviet Union acted upon in the 1960’s (with no success). There are some sound arguments against the feasibility of Dyson Spheres though (Papagiannis 1985, pp. 268-269). However, an important aspect of Dyson’s work is that it opens our eyes to possibilities outside of our limited experience. Another more accepted alternative search method is termed ‘Optical SETI’. Only a couple of years after Cocconi and Morrison’s initial paper on microwave communication, an alternative method of using lasers was suggested by Schwartz and Townes (1961) in their paper ‘Interstellar and Interplanetary Communication by Optical Masers’. At the time lasers were in their infancy, as opposed to the relative ‘maturity’ of microwave communication, so the area remained unexplored. In recent years, however, a number of projects have been initiated which are grouped as ‘Optical SETI’. The idea is that a high-intensity pulsed laser and a moderately sized telescope can form an efficient interstellar mode of communication (Optical SETI Homepage 1999). Although radio waves travel further, lasers have been recognised as the superior mode of direct interstellar communication (Kingsley 1999) – this method assumes that we are the target of an intentional attempt at signalling. Thus, searches for these kinds of pulses from extraterrestrial civilisations utilise modified telescopes, fitted with a device to recognise such pulses. The search is generally conducted through the visible part of the spectrum, however some have concentrated on the infrared and ultraviolet sections. Optical SETI is one of the most exciting new directions amongst the various SETI programs currently in progress.
Tarter (1985, pp. 271- 280) classifies SETI programs under three headings; directed, shared or dedicated. Dedicated programs are those in which the sole function of the equipment is SETI: obviously the favoured choice but in reality very rare. Directed programs are those that control the acquisition of data (eg. pointing the telescope): not as ideal as a dedicated program, but more realistic. Shared programs are those in which the data acquisition is shared with another user, or which re-uses existing data collected for another purpose. The most comprehensive directed search so far has been ‘Project Phoenix’, a privately funded search which literally rose from the ashes of a NASA program (committed to both a sky survey and targeted search) which lost funding in the early 1990’s. The program utilised a truck trailer filled with custom-built equipment and travelled to whichever radio telescopes it could secure time on (MacRobert & LePage 1999) – at times being lucky enough to get time on such massive installations as the Arecibo Telescope in Puerto Rico. Project Phoenix purchased the NASA program’s equipment and discontinued the wide-band sky survey element of the project, concentrating on the targeted search instead. It was able to listen to more than two billion channels between 1.2 and 3.0 gigahertz at a resolution of 0.7 hertz per channel, easily the most impressive search by any existing SETI program (Project Phoenix General Information 1999). Its greatest weakness was that due to it being a directed search, it was only running for 5 percent of the year (assuming perfect conditions). Project Phoenix has just concluded, but unfortunately found no signals from extraterrestrial sources.
Another program, ‘Project SERENDIP’, has got around some of the problems faced by Project PHOENIX by ‘piggybacking’ an extra receiver onto an existing radio telescope. This shared search is therefore capable of, in theory at least, running continuously, although it cannot control where it is aimed (due to it being the secondary user). Project SERENDIP is thus a sky survey that listens to 168 million channels, each 0.6 hertz wide, in a 100-megahertz band focused on the magic frequency of 1.42 gigahertz (MacRobert & LePage 1999). The weakness in the SERENDIP program is that there is no real-time follow up; that is, if a signal is found the telescope cannot be directed back to the point immediately to check whether it is genuine, it must be done elsewhere later. One problem that SERENDIP has addressed ingeniously however, is how to process huge volumes of data on narrow-band signals – by using the power of distributed computing.
As technology advanced, SETI projects were able to scan growing numbers of channels through the use of MCSA’s (Multi-channel spectrum analysers (Cullers et al 1985, p. 39)): compare Project Ozma’s single channel versus Project Phoenix’s two billion. The problem soon became not that enough channels were available, but that there wasn’t enough computer processing power to analyse each channel for a genuine signal. This problem has been solved in part recently by the SETI@home program, which utilises the power of distributed computing. Volunteers download a program for their home computer which installs itself as the default screensaver. While the computer is not being used the program processes a ‘work unit’ of SETI information; once done, it uploads it to the SETI@home site and downloads a new work unit. Each work unit represents 107 seconds of listening time and only 10 kilohertz of bandwidth: every 256 work units covers SETI@home’s 2.5 megahertz search band (centred once again on the magic frequency of 1.42 gigahertz) for this time period (MacRobert & LePage 1999). As every 107 seconds represents a strip of sky 0.1 by 0.3 degrees in area, it is obvious that a lot of volunteers are necessary. Fortunately for SETI@home, they have had at least over four hundred thousand volunteers in their program (MacRobert & LePage 1999).
There are a number of other programs running as well, though not on the scale of SERENDIP, SETI@home and Project Phoenix, such as META (Million-channel Extra-Terrestrial Assay), META II and BETA (Billion-channel Extra-Terrestrial Assay). There is also amateur participation, such as Project BAMBI (Bob and Mike’s Big Investment) and Project Argus, a coordinated effort organised by the SETI League. The most revolutionary programs at the moment, however, are probably the various Optical SETI projects. COSETI (Columbus Optical SETI) is conducting a targeted search of hundreds of stars for both narrowband laser signals as well as pulsed signals at visible wavelengths (MacRobert & LePage 1999). The COSETI Observatory has continually argued against the proposition that optical SETI is scientifically illogical (Kingsley 1999). The Director of COSETI, Stuart Kingsley, states on his website his belief that laser transmitters are the superior form of interstellar communication. This view is gradually gaining acceptance – other Optical SETI programs are being carried out at the University of California and also at Harvard University.
Nevertheless, orthodox SETI continues with the ‘sequel’ to Project PHOENIX. The SETI Institute, working with funds donated by Microsoft co-founder Paul Allen, is constructing their very own radio instrument, the Allen Telescope Array (ATA). The ATA can be used 24 hours a day for simultaneous observations by radio astronomers and SETI researchers. The technique used by the ATA is novel: hundreds of “satellite” dishes will be wired together to form an instrument that’s the equivalent of a single, massive, 100 meter-wide radio telescope.
Despite the many approaches currently being undertaken in SETI, and the optimism of those involved, there remain many problems in this area of research. One of the largest, and not at all scientific, is the problem of funding. The ‘giggle factor’ inherent in SETI discussion remains a difficulty for those pursuing government funding, with politicians deeming it trivial and unworthy (SETI Faces Uncertainty on Earth and in the Stars 1992). All programs at the moment are funded either by educational institutions or through donations. The success of SETI@home argues that the public is interested in supporting searches however, so perhaps this situation will change (indeed, Paul Allen’s recent involvement may be a watershed). The lack of funding contributes to another problem: budget-oriented shared programs lack real-time follow up on promising signals. Without this feature we may always be left wondering whether a signal that cannot be re-acquired was a missed chance of detecting an extraterrestrial signal. Also, the growth of interference from Earthly technologies makes the job of listening in more difficult each year. For microwave searches, a major problem also lies in the ‘cosmic haystack’ feel (Billingham and Tarter 1993, p. 255) which is inherent in scanning narrow-band signals within a very wide total bandwidth. This problem may be alleviated as search technology advances and computing power increases though.
Finally, some problems are always going to be inherent in any attempt we make to communicate over long distances. If we detect a signal from a source 100 light years away, by the time of receiving it the civilisation from which it originated may no longer exist. Communication therefore, will probably not be between individuals, but between civilisations. And perhaps we are wrong with our assumptions: we have selected certain forms of communication and certain frequencies as ideal, whereas advanced civilisations elsewhere may use technology outside of our sphere of thinking. Alternative writer Terence McKenna once pointed out this anthropocentric problem
To search expectantly for a radio signal from an extraterrestrial source is probably as culture bound a presumption as to search the galaxy for a good Italian restaurant.
In response though, SETI enthusiasts may quote pioneers Cocconi and Morrison (1959), who said at the very beginnings of SETI:
The probability of success is difficult to estimate; but if we never search, the chance of success is zero.
In the past forty years, the search for extraterrestrial intelligence has taken great leaps forward in both theory and practice. Drake’s Equation has become a standard method of working out the number of communicating civilisations (although still debated hotly), and perhaps more importantly, has lent a legitimacy to the SETI effort as a scientific process. Considering methods of communication which might be employed by an extraterrestrial civilisation has led to interesting theoretical discussions of technologies which may be employed. On the practical side, evidence of improvement is obvious when comparing the capabilities of Project Ozma and Project Phoenix (Project Phoenix matched Project Ozma in a fraction of a second of operation). At the rate technology is changing, we soon may be able to achieve truly comprehensive searches of the microwave spectrum. Also, new developments such as Optical SETI and SETI@home offer new and exciting ideas to the search. Many problems still face SETI programs, but strong organisation and a desire to succeed may be enough to ensure their survival. The only question remaining, perhaps truly the only question, is whether anybody is out there.
Billingham, John & Tarter, Jill 1993, ‘SETI: Search for Extraterrestrial
Intelligence’, in Space and its Exploration, eds Arnauld E. Nicogossian,
Stanley R. Mohler, Oleg G. Gazenko, Anatoliy I. Grigoryev, American Institute
of Aeronautics and Astronautics Inc., Washington.
Cullers, D.K., Linscott, I.R., & Oliver, B.M. 1985, ‘Signal Processing in
SETI’, Communications of the ACM, vol. 28, no. 11, pp. 37-48.
Cocconi, G., & Morrison, P. 1959, ‘Searching for Interstellar Communications’,
Nature, vol. 184, no. 4690, p. 844.
Dyson, Freeman J. 1963, ‘Search for Artificial Stellar sources of Infrared
Radiation’, in Interstellar Communication, ed. A. G. W. Cameron, Benjamin,