First world?

30 years ago, astronomers found a planet where it shouldn’t be — and made history

It got weird, fast.

Updated: 
Originally Published: 
Illustration of the black widow pulsar. This is a pulsar - a rapidly rotating neutron star - discove...
MARK GARLICK/SCIENCE PHOTO LIBRARY/Science Photo Library/Getty Images

The James Webb Space Telescope, Penn State exoplanet astronomer Alexander Wolszczan says, is built on a foundation of past data and future expectations — much of what the massive space telescope will reveal combines both what scientists expect and what they hope for.

“I'm always fascinated, as far as new instruments are concerned, in serendipity,” Wolszczan tells Inverse.

Wolszczan knows all about serendipity in science. Thirty years ago, on January 22, 1992, Wolszczan and his colleague David Frail published a paper confirming the existence of a planet outside our Solar System for the first time: PSR B1257+12 B, later named Poltergeist.

Long before Wolszczan and Frail’s discovery, scientists believed there must be exoplanets out there — and they were right. We now know of nearly 5,000 exoplanets, and major space telescope missions make studying them in detail part of their reason to be.

But Wolszczan’s discovery came after after five decades of mistakes. Here’s the kicker: Wolszczan wasn’t even looking for exoplanets.

“It was completely accidental,” he says.

In the three decades since, Poltergeist and its star system remain one of the most odd of all exoplanetary systems discovered.

NASA illustration showing what conditions in PSR B1257+12 might look like.

NASA/JPL-Caltech/R. Hurt (SSC)

Accidental exoplanet astronomer

In the late 1980s and early ‘90s, Wolszczan was studying pulsars — powerfully magnetized, rotating, neutron stars that pulse with radiation at exact periods.

“I was a pulsar hunter, trying to find interesting ones to be used to do other kinds of physics, as precise clocks,” he says.

Pulsars are so accurate in their periodic ping that astronomers use them as beacons or aides in the search for gravitational waves and dark matter. But due to their violent formation in the heart of supernovae explosions, no one suspected them to be a haven for planets.

Wolszczan was then the resident pulsar radio astronomer at the now defunct Arecibo observatory in Puerto Rico. Maintenance stopped outside astronomers from using the large dish for two months in early 1990, giving him the time he needed to search for millisecond pulsars beyond the Milky Way’s galactic disk. These neutron stars had been predicted to exist out there, but astronomers had not detected them — yet.

“I found rather quickly two pulsars out of the galactic plane,” Wolszczan says, one of which, PSR B1257+12, about 2,300 light years away in the Virgo constellation, appeared to have some irregularities in its timekeeping.

Exoplanetary false starts

Other astronomers were hunting exoplanets around the time Wolszczan stumbled on one. In 1988, a team of Canadian astronomers announced the discovery of Gamma Cephei Ab, a Jupiter-mass (ish) planet orbiting an orange giant star about 45 light years from Earth.

The “discovery” was retracted in 1992. The team didn’t have enough data to sustain their claim... but it re-discovered, for real, in 2003, when another group confirmed Gamma Cephei Ab’s existence.

Importantly, the Canadian team hunted exoplanets using the radial velocity method, which involves carefully watching the spectra of light emitted by stars over time for a wobble in their motion. The wobbles are indicative of the tug of an orbiting planet.

It’s a difficult technique, requiring long and precise observations. At the time, as astronomers had yet to discover an exoplanet, skepticism hampered scientists’ ability to detect them, Lisa Kaltenegger, a Cornell astronomer and director of the Carl Sagan Institute, says.

“The expectation was that planets around other stars that we could detect would need an incredibly long time to detect,” Kaltenegger tells Inverse.

“Or much more precise measurements, which were not yet available.”

Pulsar precision

Wolszczan, of course, wasn’t using the radial velocity method, but he was paying very close attention to the pulsing behavior of PSR B1257+12.

“When you discover a new pulsar, typically, what you do is you time it for some period of time,” he says.

This lets you know its precise rotation period. Knowing PSR B1257+12 was likely part of a binary system — a pulsar in mutual orbit with a white dwarf or similar star — Wolszczan looked for signs of the companion star’s gravitational influence on PSR B1257+12’s pulse.

“That changes because of the wobble of the pulsar around the center of mass. So you look at the varying delay of the arrival time of the pulsar signal,” he says.

“It's like a clock that doesn't really behave exactly the way it should. It sort of goes late and early, late and early, periodically.”

The thing was, PSR B1257+12 was swinging earlier and later than should be the case if its companion was a white dwarf star. It quickly became apparent to Wolszczan that he was dealing with other objects in the system. Unlike the radial velocity method, the variations in pulsar timing were exquisitely sensitive to alteration by the presence of other massive objects, he says, and it's now known to be sensitive enough to detect asteroid mass objects.

“It had to be at least two low mass objects orbiting the pulsar,” Wolszczan says. “It required masses that were terrestrial or to account for the slight wobbling of the clock, which was in the range of milliseconds.”

Wolszczan and Frail reported the detection of Poltergeist and its companion Phobetor in the 1992 paper. Both are about four times the mass of Earth. In 1994, Wolszczan and Maciej Konacki announced a third planet, Draugr, which remains one of the smallest exoplanets ever reported — just twice as massive as the Moon. Until an IAU contest wrapping up in 2015, astronomers simply called the planets PSR B1257+12 A, B, and C.

NASA made several “tourism” posters for distant planets — and this one of Poltergeist and its siblings as “zombie” world speaks both to their harsh nature and their post-supernova formation.

NASA-JPL/Caltech

Mixed reception

Serendipity may have smiled on Wolszczan, and the mathematics were inescapable, but the human response to his discovery was somewhat bifurcated.

“One common reaction was that if you can make planets around a neutron star, then planets have to be everywhere,” he says. “Let’s just go and intensify our efforts to find them.”

But some people gave a cooler reaction.

“The other one was like, well, that’s interesting, quite exotic,” Wolszczan says, “but this is not really what we’re looking for.”

Astronomers were looking for worlds around Sun-like stars, not certain-to-be sterile rocks orbiting a neutron star that is constantly bathing them in lethal radiation. No one is entirely sure how Poltergeist and its companions formed. One theory says they could be the congealed remains of PSR B1257+12 shredding passing stars.

“I was listening to a talk at MIT by a NASA scientist who was there giving a visiting lecturer in the physics department,” Lisa Messeri, an anthropologist at Yale University who has studied exoplanetary scientists, tells Inverse. “And he made it very clear that in his estimation that the pulsar planet — and his language is very striking to me — could not be imagined as a place.”

A few years later, the discovery of 51 Pegasi b, the first planet around a Sun-like star, overshadowed Poltergeist for such scientists — though it met with its own controversy initially, given the history of erroneous detections of planets around other stars. It was also weird in and of itself, orbiting its star in just four days.

The game was to look for “normal” stars, stars like our Sun, and planets more like our own.

“That was like, let’s just treat it as a historical fact and ignore it and just continue doing our stuff,” Wolszczan says.

He continued — today, Wolszczan is the director of the Penn State Center for Exoplanets & Habitable Worlds.

Poltergeist’s legacy

Exoplanet science has progressed tremendously since 1992. In 1999, astronomers successfully used the transit method — watching stars for dips in their brightness as planets pass in front of them — to discover HD 209458 b or Osirus. Since then, the technique has become incredibly fruitful for exoplanet hunters.

“About 70 percent of the exoplanets known so far have been discovered by transit methods,” Kaltenegger says.

NASA’s Kepler and TESS missions used the technique, and the James Webb Space Telescope will even be able to study transiting exoplanets to suss out details of their atmosphere and find planets with Earth-like atmospheres.

And what about Wolszczan’s pulsar detection method? As precise as it is, shouldn’t it have revealed many such pulsar planets by now?

That’s the thing, according to Wolszczan. “I’ve been engaged with a number of long-term surveys, which were to search for millisecond pulsars and then look for planets around them,” he says, and while they’ve found a few objects around neutrons stars, they’ve found nothing like the PSR B1257+12 system. “That is the disappointing part of the whole story.”

It remains a mystery why the first confirmed exoplanet appears to be part of a one-of-a-kind system. But then, exoplanet science is still relatively young and full of surprises. Star systems once assumed to be entirely hostile to life may yet prove to provide it a narrow niche.

“Last year, I was part of the discovery of a gas planet around a more calmly generated dead stellar husk, a white dwarf,” Kaltenegger says. If planets can survive the demise of their star and persist until it settles down in its white dwarf afterlife, she says, “a new kind of planet is now on our cosmic horizon, which could harbor life.”

This article was originally published on

Related Tags