The emptiness was the experimental design itself.
In December 1995, an astronomer named Robert Williams chose to spend ten precious days of the Hubble Space Telescope's time staring at what appeared to be nothing — a pinhead-sized void near the Big Dipper — and in doing so, revealed roughly three thousand galaxies hiding in the dark. The decision was contested, the risk was real, and the telescope itself was still recovering its reputation from a flawed mirror. What emerged was not merely a stunning image but a new way of seeing: the deliberate act of looking at emptiness as a method for uncovering the deep past of the universe.
- Robert Williams staked ten days of the newly repaired Hubble on a blank patch of sky, defying senior colleagues — including John Bahcall, who had published doubts and visited Williams personally to urge him to stop.
- The telescope's reputation was already fragile after its mirror flaw humiliated NASA publicly, meaning a failed gamble on apparent nothingness could have been a second, devastating blow.
- Over 342 exposures and more than 100 hours of combined light-gathering, the image slowly filled with roughly 3,000 galaxies — many of them twisted and irregular, showing the universe's chaotic youth billions of years ago.
- Williams released the data immediately to the entire scientific community, forgoing the usual proprietary period, which multiplied the discovery's impact far beyond his own team.
- The method became the template: southern Deep Fields, Ultra Deep Fields, and now James Webb images are all descendants of that first contested stare into the void.
In December 1995, the Hubble Space Telescope fixed its gaze on a patch of sky near the Big Dipper's handle — a region so small it would disappear behind a pinhead held at arm's length — and held it there for ten consecutive days. When the exposures were combined, roughly 3,000 galaxies emerged from what had appeared to be empty space.
The emptiness was not accidental. Robert Williams, director of the Space Telescope Science Institute, had chosen the field with precision: far from the dust-choked plane of the Milky Way, free of bright foreground objects, and positioned so Earth would not interrupt the telescope's line of sight. The apparent void was the experimental design — a cleared stage on which the faint and ancient could finally appear.
The risk was as much political as scientific. Hubble time was fiercely contested, and the telescope was still raw from public embarrassment over its flawed mirror. A servicing mission had only recently restored its vision. John Bahcall, one of Hubble's founding champions, visited Williams more than once to argue against the plan, having even published skepticism that the telescope would uncover new galaxy populations. Williams proceeded anyway, describing himself simply as a risk taker.
The resulting image, built from 342 exposures totaling over 100 hours, revealed not just familiar spirals and ellipses but strange, contorted galaxies — the distant ones, whose light had traveled billions of years. Their irregular shapes fit a theoretical model in which small galaxies collide and merge across cosmic time to build larger ones, offering direct visual evidence of the universe's long structural history.
Williams then made a second consequential decision: he released the data immediately to the entire astronomical community, waiving the proprietary period that would normally have given his team exclusive access. Much of the science that followed came from researchers who had played no part in taking the image.
The method itself became the inheritance. Southern Deep Fields, the Hubble Ultra Deep Field, and now the James Webb Space Telescope's even deeper surveys all descend from that first contested stare. What began as a gamble against expert advice became the standard template for reaching further and further back into cosmic time.
In December 1995, the Hubble Space Telescope pointed at a patch of sky near the handle of the Big Dipper—a region so small it would fit behind a pinhead held at arm's length—and stared at it for ten consecutive days. When the exposures were assembled, astronomers found roughly 3,000 galaxies packed into that apparent void. The decision to do this was not obvious. It was contested. And the astronomer who pushed it through had to do so against the explicit objections of senior colleagues who thought the whole thing was a waste of time.
The emptiness was not accidental. Robert Williams, then director of the Space Telescope Science Institute, did not stumble upon a blank field and decide to look at it. He and his team selected this particular patch of sky because it met a very specific set of requirements. It had to sit far from the crowded plane of the Milky Way, where dust and nearby stars would block the view. It had to be free of bright objects whose light would overwhelm anything faint. And it had to be positioned so that Earth would not repeatedly swing into the telescope's line of sight. An empty-looking field satisfied all these conditions. The emptiness was the experimental design itself. The point of aiming at apparent nothing was to clear away the foreground clutter so that the faint and distant could actually be seen.
What made this a genuine risk was not the science—it was the politics. Hubble time was scarce and fiercely fought over. Every hour of observation was claimed by multiple competing teams. Worse, the telescope's reputation was still recovering from a public humiliation. The flaw in its primary mirror, discovered after launch in 1990, had made it a symbol of failure. A servicing mission had recently installed corrective optics that fixed the problem, but the wound was fresh. Spending ten days of a newly repaired and heavily criticized telescope on a blank piece of sky was the kind of decision that would look catastrophic if it returned nothing at all. John Bahcall, a central figure in Hubble's creation, visited Williams more than once to urge him to abandon the idea. Bahcall had even published a paper around the time of Hubble's launch arguing that the telescope would not discover a new population of galaxies. Williams proceeded anyway. He later described himself, without elaboration, as a risk taker.
The image that emerged was built from 342 separate exposures, with a combined exposure time of more than 100 hours—roughly ten times longer than a typical Hubble observation. The first analysis counted roughly 1,500 galaxies. Later work raised that figure to around 3,000. They were not all the same. Alongside the familiar spiral and elliptical shapes were galaxies that looked irregular, disturbed, and oddly contorted. Many of these distorted ones were the distant ones, light that had traveled billions of years to reach the telescope, showing galaxies as they had been in the universe's youth. The European Space Agency noted that the image reached back to an era when many galaxies had not yet assembled most of their stars.
This mattered beyond the sheer novelty of the count. The twisted and irregular appearance of the distant, younger galaxies fit neatly with a theoretical picture in which small galaxies collide and merge over cosmic time, gradually building larger ones. The Deep Field did not prove this model on its own, but it provided direct visual evidence of it across billions of years of cosmic history.
What transformed the gamble from a single bold result into something much larger was a decision about how to share the data. Williams and his team released the Deep Field observations to the entire astronomical community without the usual proprietary period that would have given them exclusive access for months or years. Any researcher could work on it immediately. Much of the science that followed came from people who had no part in taking the original image.
The method itself then became standard practice. Staring for extremely long periods at carefully chosen blank fields was repeated and extended: a southern Deep Field in 1998, the Hubble Ultra Deep Field in 2004, and more recently the deep images from the James Webb Space Telescope, which can see galaxies from even earlier in cosmic time than Hubble could reach. What began as a contested gamble became the template for how astronomers would look deeper into the universe.
One caveat deserves mention. The Deep Field is a single narrow sample of the universe, and the often-quoted estimates of the total number of galaxies in the observable universe are extrapolations from this sample and others like it, not a direct count of every galaxy that exists. What the 1995 image actually settled was narrower and more solid: that a patch of sky chosen specifically for appearing empty was, in fact, full of distant light. The question it handed forward was how much further back the same method could be pushed, and that question is still being answered.
Notable Quotes
John Bahcall visited Williams more than once to urge him to abandon the idea, having published a paper arguing that Hubble would not discover a new population of galaxies.— Source material on Bahcall's opposition
Williams later described himself, without much elaboration, as a risk taker.— Robert Williams, director of the Space Telescope Science Institute
The Hearth Conversation Another angle on the story
Why choose an empty patch at all? Why not just point at something that already looked interesting?
Because interesting-looking regions are crowded with nearby stars and dust from our own galaxy. That glare drowns out everything faint and distant. The emptiness was the whole point—it cleared the foreground so the faint light could register.
So Williams was betting that nothing would turn out to be something.
Exactly. But the real bet was institutional. Hubble time was precious and contested. If he pointed at a blank sky for ten days and found nothing, it would have looked like a catastrophic waste of a newly repaired telescope that was already embarrassed.
And Bahcall, who helped create Hubble, told him not to do it?
More than once. Bahcall had even published a paper saying Hubble wouldn't find a new population of galaxies. Williams went ahead anyway. He called himself a risk taker, but he didn't elaborate on why.
What did the image actually show about how galaxies form?
The distant galaxies looked twisted and irregular, not like the neat spirals and ellipses we see nearby. That fit a theory that small galaxies collide and merge over time to build larger ones. The Deep Field gave that theory direct visual support across billions of years.
Did Williams keep the data to himself?
No. He released it to the entire astronomical community without any proprietary period. That was crucial. Much of the science came from researchers who had nothing to do with taking the original image.
So the method became standard?
It did. Other deep fields followed—southern, ultra-deep, and now James Webb is pushing even further back. What started as one contested gamble became how we look at the universe.