Publications Topics:

Books

ASP Conference Series

Monograph Publications

IAU Publications

Books of Note

Purchase through the AstroShop

Journals

Publications of the ASP (PASP)

Magazines

Mercury Magazine

Archive

Guidelines for Authors

Order Mercury Issues

Mercury Advertising Rates

Newletters

The Universe in the Classroom

ASP E-mail Newsletters

Special Features

Astronomy Beat

Contact Us

Mercury, March/April 1999 Table of Contents

Douglas A. Vakoch
SETI Institute

In the absence of knowledge of physical and cultural clues, communication between two species can be almost impossible.

Almost.

An Historical Overview of Interplanetary Messages

Directly showing pictures for interplanetary communication

The earliest speculations about communication with extraterrestrial intelligence (CETI) involved contact with the inhabitants of other bodies of our Solar System, either our Moon or other planets. Astronomical theories in the nineteenth century made plausible a belief in the prevalence of life on non-terrestrial worlds circling our Sun. When the focus turned to the possibility of communicating with the potential denizens of these worlds, their relative closeness to the Earth made it conceivable that signals could be detected through optical telescopes.

Assuming that other intelligence could visually observe Earth, and given that nineteenth century astronomers relied on optical telescopes to survey the heavens, it is understandable that early proposals for messages emphasized visible signaling. One popular type of proposal involved displaying meaningful figures on a side of the Earth facing the target moon or planet. In these plans, huge diagrams would be etched on large expanses of land. For example, a visual representation of a right triangle could be shown, with a square attached to each side of the triangle to illustrate the Pythagorean theorem diagrammatically (Figure 1).

Figure 1. In an early proposal to communicate with inhabitants of the Moon, geometrical concepts would be shown directly. For example, the Pythagorean theorem could be illustrated visually during the daytime by clearing vast expanses of forest in Siberia to show the areas surrounding a right triangle (right side of figure). During the night, canals dug into the Sahara desert in the shape of a circle could be filled with kerosene (left side of figure). When lit, the flames would provide a pictorial signal of our existence. Illustrations courtesy of author.

By clearing gargantuan stretches of forest in Siberia, such geometrical concepts could be illustrated for viewers watching the lighted side of the Earth and, by creating large canals in the Sahara Desert filled with lighted kerosene, a similar signal could be sent from the dark side of the Earth (Figure 1). Among the early proponents of directly displaying pictures to communicate with extraterrestrials was the illustrious mathematician Karl Friedrich Gauss, who in 1826 was attributed with suggesting such an approach for communicating with potential selenites - inhabitants of the Moon. As interplanetary spacecraft became conceivable in the early twentieth century, the same notion of directly showing pictures was proposed for vessels bridging the space between worlds. A century after Gauss's proposal, rocket pioneer Robert Goddard suggested that interplanetary craft might bear metal plates inscribed with geometrical shapes and astronomical objects to initiate CETI.

Encoding pictures for interplanetary communication

In addition to proposals for sending pictures directly, others advocated sending messages that would not be comprehensible without a reconstruction of the message format. For example, in 1869 the polymath Charles Cros suggested presenting pictorial information as sequences of numbers (Figure 2). In this plan, several series of numbers would be sent. As a clue to reconstructing the picture from these series of numbers, the sum of the numbers in each series would be the same. Recipients successful in decoding the message would first convert these series into strings composed of beads (or their equivalents) of two different colors, with the number of beads of a certain color equal to the transmitted number for that part of the series. Finally, if these strings of beads were aligned one under the other in the order they were received, the reconstruction of the picture would be complete.

A similar scheme was advanced in 1920 by H. W. Nieman and C. Wells Nieman. Rather than transmitting series of numbers, however, the Niemans suggested signaling with a series of "dots and dashes" sent either by "wireless or light" (Figure 2). Each dot or dash could be represented by a bead, with dots and dashes represented by different color beads. Since each series would have the same number of combined dots and dashes, the recipients would have a clue that they could be aligned one under the other, as in Cros's proposal.

Figure 2. A pictogram representing a square can be described in several ways. Cros suggested transmitting several series of numbers, with the sum of each series being equal to the same number, as an aid to decoding. In this example, the total of the numbers in each series equals 11. When the series are "stacked," a two-dimensional representation is created. Nieman and Nieman proposed a similar approach, using signals of two different durations ("dots and dashes"), which are represented as beads in black and white in this reconstruction. Illustration courtesy of author.

A slightly different approach was offered by Francis Galton in 1896. Instead of encoding pictures as sequences of discrete units ordered in a two-dimensional array, Galton favored starting with an introduction to mathematics, and only later building up to pictorial representations of objects by defining their outlines. Galton compared his "picture-writing" to embroidery in which each "stitch" composing the outline of an object would be defined in terms of its length and direction.

Unlike Gauss's plan, the proposals of Cros, the Niemans, and Galton add the requirement of having the recipients reconstruct the format of the message. As can be seen from the few proposals mentioned so far, even when there is a goal of portraying information in a two-dimensional array, there are additional questions about exactly how the information should be encoded, and thus also how it would be reconstructed.

Messages for Interstellar Communication

Directly showing pictures for interstellar communication

Ideas similar to those proposed in the nineteenth and early twentieth centuries were independently espoused by scientists and engineers working in the 1960s on what had become proposals for interstellar (as opposed to interplanetary) communication. With diminished estimates of the likelihood of intelligent civilizations elsewhere in our Solar System, the search for extraterrestrial intelligence (SETI) moved to other stars. (While the acronym "CETI" refers to either sending messages to extraterrestrials or receiving signals from them, the more recent term "SETI" typically refers only to the more passive listening strategy.) As with messages proposed for communication within our Solar System, we see two options for communicating with pictures at interstellar distances:

• Directly, without encoding them; or
• Indirectly, using messages that require reconstruction, but that would be designed to facilitate decoding.

The first option-showing pictures directly-is exemplified by the messages borne on the Pioneer and Voyager spacecraft. Although these craft were not intended primarily for CETI, each carries a message to any extraterrestrials who might intercept them in their travels beyond the Solar System during the coming millennia. Each Pioneer spacecraft carries a plaque that is inscribed with several diagrams (Figure 3), including one of the spacecraft as it makes its way through the Solar System (with planetary diameters indicated as multiples of hydrogen wavelengths in binary numbers), another showing the same spacecraft in greater detail, and an outline drawing of a human female and male.

Figure 3. On the Pioneer plaque, an outline of the Pioneer spacecraft is seen behind the figures of two humans. At the bottom of the plaque, the same spacecraft is shown in a smaller scale as it passes through the Solar System on its journey from Earth (the third planet from the Sun). The figure with two connected circles in the upper left-hand corner represents a hydrogen atom's spin transition. The left central figure of fifteen converging lines shows the Earth's location in time and space in relation to prominent pulsars. Illustration courtesy of author.

Similarly, the Voyager spacecraft each bear messages, some engraved on the metal coverings of the recordings and others encoded on the record disks themselves. Some of these engravings are the same as those on the Pioneers, such as the diagram showing the location of Earth and the time of launch by reference to several pulsars. Others are unique to the Voyager record cover, for example, the outline drawing illustrating how the stylus and recording are to be used together (Figure 4). For the external diagrams on the Voyager, as well as the pictures on the Pioneer plaques, the format need not be reconstructed. As with Gauss's plan of portraying complete diagrams on the face of the Earth, both the Pioneer and Voyager spacecraft carry direct pictorial messages.

Figure 4. The cover over the Voyager recording bears two figures (upper left) that provide instructions on how to place the stylus on the recording. Diagrams in the upper right-hand corner illustrate the signals produced by playing the record. As on the Pioneer plaque, representations of the Earth's position and of a hydrogen atom are also included. Illustration courtesy of author.

Encoding pictures for interstellar communication

Each Voyager spacecraft also bears a recording encoded in a format that must be understood by the recipients before the recorded message will be intelligible. The protective covering over each recording is inscribed with pictorial instructions about how to place the stylus on the record disk, and then how to turn the recording, resulting in "playing" the pulses encoded on the disk (Figure 4). In addition, an etching indicates how to align a series of pulses next to one another, in a manner reminiscent of that suggested by Cros and the Niemans. As a confirmation that the correct format is used, the first image to be reconstructed from the recording-a circle-is also shown on the protective cover. The first part of the recorded message, which consists of pictorial information, quickly moves from a description of a numbering system to diagrams of atoms and molecules, and then on to a wide variety of pictures of the world as we know it.

But what if we do not have physical contact with a spacecraft that bears messages? What if we must overcome the distance between stars with electromagnetic radiation, for example at radio or optical frequencies? The vast interstellar distances separating transmitter and receiver make proposals like Gauss's impossible. No longer is it possible to show directly pictures of objects or concepts, at least not using pictorial representations.

Figure 5. A hypothetical stocky biped is shown at the bottom center of this pictogram created by Frank Drake in 1962. The fictional being's solar system is also depicted, with the system's star in the upper left-hand corner and its nine planets along the left side. An oxygen atom is shown in the upper right-hand corner, with its central nucleus surrounded by eight electrons. Similarly, a carbon atom is represented at the top center with six electrons orbiting its nucleus. Illustration courtesy of author. (cf. with Figure 7)

The standard response is to use a strategy similar to that proposed by Cros, the Niemans, and Galton. Such was the approach taken by astronomer Frank Drake in 1962 when he constructed a two-dimensional "pictogram" similar to one we might some day receive from extraterrestrials (Figure 5). As a clue to decoding the message, it consists of 551 bits of information. The only factors of 551 are the prime numbers 19 and 29, which are the lengths of the sides of the message. When properly formatted, the message shows-among other things-a picture of the hypothetical species sending the message, a diagram of its solar system, and pictorial representations of carbon and oxygen atoms to indicate elements important for this extraterrestrial biochemistry. Electrical engineer Bernard M. Oliver, who constructed a similar message with this format, justified the use of chemistry for inter-species communication because, "The structure of atoms does not depend on who studies them."

When Drake actually transmitted a message from the Arecibo radio telescope in 1974, he constructed a pictogram similar to his earlier creation, but with more emphasis on terrestrial biochemistry (Figure 6). Similarly, in my own work in the 1970s, I proposed sending sequences of pictograms containing both pictures and words. When properly reconstructed, such series of interrelated pictograms could convey more than isolated pictograms.

Figure 6. The top three rows of the message sent from the Arecibo telescope in 1974 list the numbers from one to ten in binary notation. This forms the basis for numerically describing the chemical structure of DNA, which is also illustrated iconically by the helical form at the center of the message. Immediately below the helix is a figure of a human being, and below that, a representation of our Solar System, with the third planet (Earth) displaced toward the human figure. Illustration courtesy of author.

The Incommensurability Problem

While it may be true, as Oliver noted, that the structure of atoms does not depend on who studies them, scientific models of atoms may be very much influenced by the characteristics of the scientists who construct these models. And when two scientists differ in biology, culture, and history as much as humans and extraterrestrials would differ, these models of reality may vary considerably. This view that extraterrestrials and humans may have such divergent ways of conceptualizing the world that there can be no mutual understanding is referred to as the Incommensurability Problem.

At the core of the Incommensurability Problem is the view that no intelligent species can understand reality without making certain methodological choices, and that these choices may vary from civilization to civilization. As philosopher of science Nicholas Rescher explains, extraterrestrials could well have very different ways of doing science than humans. If extraterrestrials have markedly different biologies and live in considerably different environments than humans, they may well have different goals for their science. In addition, they could have radically different criteria for evaluating the success of their science. Rescher notes, "Their explanatory mechanisms, their predictive concerns, and their modes of control over nature might all be very different." Finally, their means of formulating models of reality might differ drastically from ours. For example, their analog of arithmetic might not even be quantitative, but comparative. Similarly, their conventions for picturing things could be very different from those used by humans.

Critiques of Pictorial Representation

Philosopher and historian of science Michel Foucault contends that our reliance on science is based on studying the visible characteristics of objects and was by no means a necessary development. Rather it reflects a belief that originated in the seventeenth century that true knowledge must be acquired from sight. This emphasis on vision led to eliminating other senses as potentially valuable sources of scientific information. If Foucault is correct, our reliance on visual information for scientific reasoning might have taken a different turn, with instead our other senses providing important data for science. Thus even without raising the question of whether extraterrestrials will be able to see, we may be wise not to overestimate the importance of pictorial representations for them.

But is all of this merely armchair philosophy that has no bearing on real intelligent beings? To be more concrete, let's take a closer look at contact between alien cultures here on Earth: interactions between humans from different cultures. According to Jamake Highwater, "We do not all see the same things. Though the dominant societies usually presume that their vision represents the sole truth about the world, each society (and often individuals within the same society) sees reality uniquely."

As we consider the pictorial representations in pictograms for CETI, how certain can we be that our particular conventions will be understood by extraterrestrials? To exemplify this problem, artificial intelligence expert Michael A. Arbib showed how Drake's message from 1962 might be misinterpreted if it were read upside down by an extraterrestrial very different in form from humans. In this orientation, the pictures of carbon and oxygen might be mistaken for a creature with six legs, and the picture of the biped might be seen as a communication satellite (Figure 7). As Arbib summarizes his point: "While there may be some chance of Drake's message being deciphered by an intelligence that expects any biped-like shape to be an intelligent being, it is very unlikely indeed to succeed with six-legged but large-brained creatures with tails."

Figure 7. Frank Drake's message from 1962 might well be misinterpreted by a race of large-brained hexapods. This message is an upside-down copy of the pictogram shown in Figure 5. Illustration courtesy of author.

Because we cannot be certain of the nature of any recipients of our messages a priori, it may prove difficult to construct pictures that will be unambiguous. No matter how careful we are, to some extent extraterrestrial viewers of our pictograms may project characteristics from their own species-specific experiences onto our messages.

What about more "objective" means of pictorial representation, such as photographs? Is it not true, as Stanley Cavell maintains, that "Photography overcame subjectivity in a way undreamed of by painting - by removing the human agent from the task of reproduction"?

Joel Snyder and Neil Walsh Allen claim that this is not the case. Instead, they contend that photography does not succeed in providing a fundamentally more objective image than that obtained in other visual arts such as drawing and painting: "An image is simply not a property which things naturally possess in addition to possessing size and weight. The image is a crafted, not a natural, thing. It is created out of natural material (light), and it is crafted in accordance with, or at least not in contravention of, natural laws. This is not surprising. Nor is it surprising that something in the camera's field will be represented in the image; but how it will be represented is neither natural nor necessary."

To support this view, Snyder and Allen include (among other examples) a discussion of photographing an object in motion. If we wish to photograph a person running, for instance, our options as photographers include the following ways of depicting motion:

• A blurred body against a stationary background
• A distinct body against a blurred background (by "panning")
• Both body and background "frozen" by the camera, with motion implied by the relative position of parts of the body

Although people accustomed to these conventions for depicting motion could easily interpret each possibility, an extraterrestrial with different conventions might have considerably more difficulty.

Semiotics: A General Theory of Signs

When we think of interstellar messages in terms of classical information theory, there is no innate relationship between the form of the message and the content borne by the message. Instead, once the information of the message is decided upon, an efficient means of encoding it is sought. In this approach, there is a purely arbitrary connection between content and form of the message. If instead we consider messages from a semiotic perspective, we have a wider range of possibilities for relating form and content. Semiotics is the general study of signs, where a sign is something that represents something else, the signified. For example, the words "the chair" might represent the object you are currently sitting on, if indeed you are now sitting. In this case, the words "the chair" are a sign standing for the signified, in this case, a material object. One of the tasks of semioticians is to categorize signs according to the ways that the sign and the signified are related to one another. In the case of the association between the sign "the chair" and its signified object, this relationship is purely arbitrary. The sign for this object could equally well be "the glumich." Thus, in this example there is a purely conventional association between the sign and the signified. In semiotic terms, when the association between sign and signified is completely arbitrary, the sign is referred to as a symbol. With symbols, there is no intrinsic connection between the form of expression (the sign) and the content that is expressed (the signified).

There are, however, alternatives to the purely arbitrary connection between sign and signified that is seen in symbols. One of these alternatives is the icon, which is a sign that bears a physical resemblance to the signified. For example, the profile of the man on a modern American quarter is an icon for a specific man who was the first President of the United States. We can also represent the same man with the symbol "George Washington" (Figure 8). In this case, the image of Washington is an icon because it physically resembles the signified (a particular man with certain facial characteristics). With icons, the form of the message reflects its content.

Figure 8. With a symbol, there is no physical resemblance between the sign (in this case, the words "George Washington") and its referent (a particular man with this name). By contrast, an icon is physically similar to its referent. For example, the icon of George Washington on the quarter resembles the appearance of a specific man who was President of the United Sates. Illustration courtesy of author.

Icons can also be used when the signified is less concrete. For example, the semiotician Ferdinand de Saussure noted that the concept of justice is sometimes portrayed by the scales of justice. In this case, the scales are an icon, because there is similarity between the sign (the scales that balance two weights) and the signified (the concept of justice, which involves a balance between transgression and punishment).

Iconic Representations of Chemical Concepts

As we return to the pictorial messages previously suggested for CETI, we can see that they relied heavily on icons. For example, in his message from 1962, Drake pictorially represented the Bohr model of the atom, in which electrons are visualized as bodies moving around a more massive nucleus. In addition to detailing the chemical composition of deoxyribonucleic acid in his 1974 pictogram, Drake also included an iconic representation of the double helix, thus illustrating the structure of this macromolecule. Similarly, the etched messages aboard the Pioneer and Voyager spacecraft convey notions of time and distance in terms of characteristics of the hydrogen atom, which is portrayed by an iconic line drawing.

As we have seen, pictorial representations may not be the best way to overcome the Incommensurability Problem. Yet there is something very promising about using icons, signs that bear a similarity to what they represent. To help understand what it would be like to include nonpictorial icons for CETI, it is helpful first to realize that icons are not specific to the visual sensory modality. Rather, it is possible to have a sign that physically resembles the signified in a nonvisual way. For example, the fly Spilomyia hamifera beats its wings at a frequency very close to the wing-beat frequency of the much more dangerous wasp Dolichovespula arenaria. As a result, when one of these flies is in the vicinity of a group of these wasps, the fly gains some immunity from attack by insect-eating birds. The fly's mimicry of the wasps occurs within the auditory modality; it is not attacked by would-be predators because it sounds like the wasps. In short, the fly's defense strategy is based on producing an auditory icon, in which the fly's wing-beating (the sign) physically resembles the wing-beat of the wasps (the signified).

In the same way, icons could function in any sensory modality. Given that we are not sure which sensory modality will be primary for extraterrestrials, a sign for interstellar communication that is not reliant on any particular sensory modality would be preferable. The critical difference between the iconic approach to communicating chemical concepts that I discuss next and the standard iconic approaches to CETI is the electromagnetic signal itself that acts as the sign, in this case for an object that it resembles. Thus electromagnetic radiation is used as an iconic representation, allowing a direct communication of chemical concepts without encoding the message into a format specific to a particular sensory modality. In short, we are able to fulfill Gauss's goal of directly communicating messages, but now over interstellar distances. Thus, the content of a transmitted interstellar message can be shown directly through its form.

To understand how we might use icons to communicate chemical concepts, it is important to recall that each chemical element can be characterized by a particular pattern of frequencies of radiation that it gives off. This characteristic emission spectrum results from the transition of electrons between orbitals. Specifically, as an electron moves from a higher to a lower orbital, the atom releases radiation at a set frequency. For any given transition, the frequency of radiation is always the same because energy is released only in certain specifiable quanta, or discrete units.

An iconic approach to directly conveying chemical concepts could involve transmitting signals at multiple frequencies, either simultaneously or sequentially (see "Traversing the Galactic Darkness," p.14). The frequencies of these transmissions would correspond to a few of the emission lines that together best characterize each element. For example, to transmit an icon representing hydrogen using infrared wavelengths, we could sequentially transmit narrowband signals at the following wavelengths: 1875.1 nanometers (nm), 1281.8 nm, 1093.8 nm, 1004.9 nm, and 954.5 nm. These are the wavelengths of the photons that are emitted as electrons move from one orbital to a lower level (Figure 9). Specifically, photons having these wavelengths are emitted when electrons end up at the orbital level with quantum number n = 3, after having moved down from five higher orbital levels. These signals would be of very long duration and would be repeated many times to increase the likelihood of their being detected. This would be particularly important because of the wide range of frequencies these signals would cover.

Figure 9. Each chemical element shows a characteristic pattern of frequencies that results from its emission of photons. When an electron moves from a higher to a lower orbital, a photon of a specific wavelength is released. For example, when a hydrogen atom's electron moves from the fourth orbital to the third, radiation is emitted with a wavelength of 1875.1 nanometers. Hydrogen can be represented iconically by transmitting a series of signals at the frequencies corresponding to these orbital transitions. This figure illustrates some of the wavelengths associated with the Paschen series, in which electrons start or end at the third orbital. Illustration courtesy of author.

Using this approach, rather than sending a signal with a purely conventional structure, the structure of the signal would physically resemble the concept being communicated. A range of concepts could be shown with this method, from notions of energy levels and orbital transitions to more complex concepts like chemical reactions. And although I have emphasized phenomena associated with electronic transitions, the same approach could be used for other transitions, for example, between vibrational and rotational energy levels of molecules. Relying more on iconicity that does not involve additional coding of the message may help address the Incommensurability Problem. The message's recipients are pointed directly toward the phenomena of interest, and not toward our models of these phenomena. Rather than providing a pictorial representation of, for example, the Bohr model of the atom, we de-emphasize particular models of atoms and attempt to show atomic phenomena directly.

The Partial Conventionality of Icons

Thus far we have spoken of a sign as if it is in a simple dyadic relationship with its signified. By this analysis, icons seem superior to symbols for establishing communication with extraterrestrials about whom we know very little because there is a natural connection between the sign and that which is signified. This contrasts with the conventional relationship between sign and signified in symbols.

But if we examine the problem from a more complete perspective, things are not so simple. In reality, the sign and the signified are in a triadic relationship between the sign, the signified, and the interpreter of the relationship between the sign and the signified. Thus, the similarity that exists between an icon and its referent does not exist independently of the intelligence perceiving this similarity. Although in iconicity there is a natural connection between the sign and the signified, this connection cannot exist without intelligence to observe the connection.

Ultimately, the problem of iconicity is that similarity is in the eye of the beholder. And because we do not know what extraterrestrials will be like, we cannot be sure that what to us seems an obvious similarity will be seen as such by intelligence with a different biology, culture, and history. Thus, judgment of similarity is not purely objective, but is influenced by a variety of factors that impact conventions of interpretation.

I have attempted a partial solution by de-emphasizing models of atomic structure and focusing more on the phenomena by which we have constructed our models. But we cannot be absolutely certain that even if extraterrestrials conceive of the material basis of reality in terms of atomic and molecular structure, they will necessarily attend to the phenomenon of spectral emissions to construct their models. As much as we try to present the phenomena in themselves, it is difficult to bracket - or even identify - the presuppositions that we make when identifying the phenomena we assume to be of universal significance. But with each attempt that we make, we will increase our chances of constructing messages that will be understood by extraterrestrials.

Notes:

1 This article originally appeared in SETIQuest, Volume 4, Nos. 1 and 2. (c) 1998 by Helmers Publishing Co., Inc. All rights reserved. Reprinted with permission of copyright holder. SETIQuest is no longer published. On the Web, visit www.setiquest.com to order individual issues or the entire set of 16 SETIQuest issues published from 1994-8. A different version of this article appeared in SPIE Proceedings, 1996, Volume 2704, pp. 140-9.

2 The title of this article was inspired by Harlow Shapley's book, The View from a Distant Star: Man's Future in the Universe, Basic Books, New York, 1963.

DOUGLAS A. VAKOCH is a social scientist at the SETI Institute in Mountain View, California, where he conducts research on the cultural aspects of SETI and composes interstellar messages. He can be reached via email at dvakoch@seti.org.