Science, even by reputable practitioners, proceeds in fits, starts, and frequent excursions down blind alleys.

As example, in 1877 astronomers on both sides of the Atlantic observed things about Mars that had the potential for making SETI a done deal, a fait accompli.

In that year, the Red Planet had orbited into a particularly favorable opposition - bringing it closer to Earth than is almost ever the case - and making it a tempting target for new discoveries. In Italy, Giovanni Schiaparelli, director of Milan's Brera Observatory, thought he saw linear features furrowing Mars' russet face. His discovery was made using a telescope that is dwarfed by many amateur instruments today: its objective lens was a modest 8 inches in diameter. Schiaparelli provocatively referred to the features as canali, an ambiguous word choice that would keep eyeballs enthusiastically pressed against eyepieces for a half-century.

It was the American astronomer, Percival Lowell, who dug Schiaparelli's work the most. He figured that the canali were deliberate constructions, and he soon became the most literate advocate for the existence of a globe-girdling irrigation system on Mars. This spider web of canals was, Lowell averred, the engineering work of advanced beings whose homes were a mere 35 million miles from our own.

Ever since the Mariner 4 space mission, in 1965, we've known that the "canals" mapped by these two astronomers - every single one - were optical illusions. Nonetheless, Lowell, who died during the First World War, remained convinced to the end that he had charted the infrastructure of a vast, hydraulic civilization. It's noteworthy that the public only showed intermittent interest in Lowell's claims. It's also stupefying to contemplate how different our lives would be today had they been true.

Merely a few weeks before Schiaparelli's work was made public, an astronomer at the United States Naval Observatory in Washington announced a different discovery about Mars - one that would (eventually) also suggest the presence of nearby intelligence. Asaph Hall, using what was then the world's largest telescope - its lens was three times the diameter of Schiaparelli's - found two midget moons orbiting the Red Planet (they're so small, you could walk around either one in a day). This was, of course, an interesting story, since moons accompanying planets of the inner solar system are as rare as Swiss jokes. Mercury doesn't have any, nor does Venus. So a pair of satellites around Mars was a compelling astronomical discovery. The Washington Evening Star hailed the find as "Glorious news from the skies." Following a suggestion from Eton scholar Henry Madan, Hall christened his diminutive discoveries Deimos ("dread") and Phobos ("fear"). According to The Iliad, these were the two bad-boy sons of Mars (Ares, in Greek tradition) and Aphrodite. Nice names for dead rocks.

A lifetime later, something funny happened. Long-term measurements of the inner of the two moons, Phobos, showed that it was losing altitude, about ten feet per century. Its orbit was gradually decaying, much the way a low Earth-orbit satellite will gently spiral downward due to the delicate drag of residual atmosphere several hundred miles up.

In the wake of Sputnik, this analogy proved irresistible to some, in particular to Iosef Shklovsky, a Soviet astrophysicist. In 1958, Shklovsky worked the numbers, making estimates of the amount of martian atmosphere at the altitude of Phobos' orbit. He concluded that the drag on this moon could produce the observed rate of descent only if Phobos was a featherweight - a hollow, metal sphere with a skin only a few inches thick.

That was an astounding claim, and remember that Shklovsky was no dummy. (His book of a few years later, Intelligent Life in the Universe, enlarged and edited by Carl Sagan, was the first to comprehensively address the question of sentient cosmic life. It was an early guide book for SETI research.) But what Shklovsky was saying - and what others, including President Eisenhower's science advisor, believed - was that Phobos wasn't dead rock at all, but a space station welded together by advanced beings from a star system far, far away.

The public, to its credit, didn't riot in the streets at the prospect of an alien outpost nearby. And indeed, a dozen years later, researchers found that the descent of Phobos had been seriously overestimated, and its orbital change was most likely due to tidal friction with the Red Planet, not atmospheric drag.

A credentialed astronomer had falsely hailed a hollow world. But at least it was a mistake with precedent. At the end of the 18th century, Edmund Halley - no astronomical slouch himself - wrote a paper for the Royal Astronomical Society in which he claimed that the Earth was hollow! More precisely, he envisioned that our world consisted of three spherical shells surrounding a small, hard central ball, a construction that he reckoned could explain the observed changes in Earth's magnetic field. Indeed, according to Sir Edmund, not only were there massive amounts of lebensraum beneath our feet, but he suggested that these inner sanctums might even be inhabited. (The idea of a vasty void within the Earth is still popular among some people. None of these are geophysicists.)

Now the peculiar thing is that, despite the errors of the past, artificial, hollow habitats remain a possibility for putative aliens. In 1959, only a year after Shklovsky's speculations about Phobos, British physicist Freeman Dyson reasoned as follows: since we seem to have an insatiable hunger for energy, it makes sense to assume that aliens have the same. To satisfy their craving for kilowatt-hours, extraterrestrials who are centuries more advanced than we are will surround their sun with an orbiting flotilla of solar-cells, allowing them to capture a large fraction of their star's energy output.

It's become popular to envision Dyson's concept as a closed, complete energy-collecting shell around a star, perhaps built with the materials obtained by dismantling an outer planet - a so-called "Dyson sphere." However, a complete spherical shell has dynamical and structural problems, and is impractical. Dyson himself was thinking more along the lines of a very large number of individual solar collectors.

Finding a Dyson sphere, or more realistically, a Dyson swarm, would be just about as exciting as learning that Phobos was built of steel plates. It would tell us that someone out there has developed some pretty advanced technology. We can't hope to see any of these solar-collectors-on-steroids directly, but they might be uncovered by their waste heat, which would be radiated into space as infrared light. In the last dozen years or so, there have been several attempts to find such objects, usually by mining data from the Infrared Astronomical Satellite (IRAS) that mapped the sky in the 1980s.

Living high on the energy hog within a surrounding shell of hardware? It's an intriguing idea, although given our limited understanding of what advanced societies will do, it might be just another blind alley. Still, science proceeds by looking, and some researchers have. So far, no Dyson.