Science’s great paradox: we don’t know what we don’t know until we look
- There are a great many puzzles humanity has sought to solve throughout our history, and scientific investigation has proven to be the best problem-solving tool of all, over and over again.
- But whenever we build a novel scientific instrument capable of probing the Universe as never before, we don’t just improve our knowledge in expected ways.
- The ability to discover something entirely new and unexpected, perhaps revealing a surprise about the workings of nature itself, is why the pursuit of the unknown must never cease.
Whenever we perform science at the frontiers — probing the Universe, at some level, in ways, with instruments, or at precisions that we’ve never achieved in all our prior interrogations of it — there’s an incredible puzzle that arises. On the one hand, we design and build our tools and experiments to be sensitive to things that we strongly suspect ought to be out there, but that we haven’t yet been able to confirm or refute with concrete evidence. When we built the James Webb Space Telescope (JWST), for example, we knew we were going to break many records: for the most distant galaxy, for the youngest supermassive black hole, for the earliest galaxy cluster to form in the Universe, and much more.
After all, we designed it with capabilities that would render it uniquely equipped to surpass the limits of all prior observatories: from the space-based Hubble, Spitzer, and WISE telescopes to the ground-based Very Large Telescope, Keck, Subaru, Gemini, and Magellan Telescopes, and even the current world’s largest optical telescope: the Gran Telescopio Canarias. And yet, on the other hand, those very capabilities enable it to make the biggest discoveries of all: discoveries that we couldn’t have even foreseen would be there to make, as nature’s imagination frequently surpasses our own.
Here, using JWST as an example, we can see the importance of investigating the Universe in ways that fundamentally take us beyond all previous limitations. The lessons we frequently learn simply by looking, humble us into realizing how ignorant we were, previously, of what we didn’t even realize we didn’t know.
Back when JWST was first being conceived, the idea was relatively simple and straightforward. We had designed, built, flown, and conducted science with space telescopes previously, and for a long time at that. After all, going to space gives you a huge advantage over astronomy conducted on the ground: you no longer have Earth’s atmosphere to contend with. Beyond that, however, there are two important physical characteristics, in general, that astronomers care about whenever they’re building a new space telescope for investigating our reality.
- The size of your telescope’s primary mirror. Size isn’t necessarily everything in astronomy, but it truly does matter for a few reasons. The total collecting area of your telescope is what determines your light-gathering power, which in turn determines how faint the objects you can detect are. With larger telescopes, you also achieve higher resolutions, as your telescope’s resolution is determined by the number of wavelengths of light that can fit across your primary mirror’s diameter. And with larger telescopes, you can image your targets and their details in less time than you can with smaller telescopes. It’s a winning bet, all-around.
- The range of wavelengths your observatory is sensitive to. We have to remember that light isn’t merely restricted to the visible light that our eyes can see, but rather extends to shorter (gamma-ray, X-ray, and ultraviolet) and longer (near-infrared, mid-infrared, far-infrared, microwave, and radio) wavelengths. Only visible light, radio, and a tiny bit of infrared and ultraviolet light can be observed from Earth; all other wavelengths require going to space, and carry along different pieces of information than we can see with visible light alone.
You then have to design and build instruments that make the most of the light you collect: to maximize the science output of your observations. For JWST, that meant a suite of four instruments, each with unique, complementary capabilities to the others.
- The Near-Infrared Camera, or NIRCam, is the primary imaging camera aboard the James Webb Space Telescope, and it is ideal for seeing through the interstellar dust that blocks most of the light that’s visible to human eyes.
- The Near-Infrared Spectrograph, or NIRSpec, is specialized for measuring the cosmic “fingerprint” of atoms and molecules that are present in any astrophysical object.
- The Mid-Infrared Instrument, or MIRI, contains both a camera and a spectrograph, and can reveal planets, comets, asteroids, warm interstellar dust, and even protoplanetary disks around newly forming stars. It probes the longest-wavelength light that Webb can see: up to 28 microns, or some 40 times longer than the maximum wavelength human eyes can see.
- And the Fine Guidance Sensor/Near Infrared Imager and Slitless Spectrograph, or FGS/NIRISS, helps point the telescope and will detect, characterize, and measure the atmospheres of exoplanets.
Each one of these instruments has many observing modes that it can work in, and by using these instruments appropriately for the science goal in question, astronomers can reveal details within objects of interest that we’ve never seen before.
Sure, its views of the Universe are gorgeous and detail-rich, and when astronomers crave those new details, JWST is an incredible tool to show us what’s there beyond the limits of all previous observations. In particular, its:
- high-resolution,
- long-wavelength capabilities,
- that reveal fainter objects than ever before,
- at cooler temperatures and greater distances than have ever been seen before,
have led to an incredible wealth of new knowledge. As a result of these capabilities, we fully expected that many new cosmic records would be set.
Prior to JWST, the most distant galaxy that we knew of was GN-z11, which we had measured, to the best of our abilities, to have emitted the light we’re observing today back when the Universe was just 407 million years old. That’s impressive! But now, in the JWST era, GN-z11 isn’t even in the top 10 of most distant galaxies known; it’s been surpassed by 10 other galaxies, all of which have been discovered, measured, or both with JWST. GN-z11 turns out to be a little older than expected (from ~430 million years after the Big Bang, as JWST revealed), while the youngest and most distant of the ten at present, JADES-GS-z14-0, had its now-observed light emitted just ~285 million years after the Big Bang.
However, at the same time that JWST stripped GN-z11 of its status as a record-breaking galaxy, it sent that same galaxy to the top of another record-breaking chart: the chart for most distant supermassive black hole ever discovered so far. Back in the Hubble era, supermassive black holes were seen early on, and many of them seemed bigger than theorists were expecting. The earliest ones had been found when the Universe was just 670 million years old, suggesting that perhaps these black holes grew very big very quickly in the early Universe, raising questions as to just how early these objects appeared in cosmic history.
Then, when the JWST era arrived, we began to find black holes that were still supermassive at even greater distances.
- In July of 2023, JWST researchers from the CEERS collaboration reported the discovery of CEERS 1019, which contains a black hole of 9 million solar masses from a cosmic age of 570 million years.
- Then, a few months later, a galaxy known as UHZ1 was found, back from when the Universe had an age of just 470 million years, with a rather large black hole mass of 10-100 million solar masses.
- And galaxy GN-z11, as was shown in January of 2024, contains a supermassive black hole of 1.5-3.0 million solar masses, from an era that was just 420-440 million years after the Big Bang.
These records, of course, weren’t the only ones to fall.
Plenty of other cosmic record-breakers came along with the newfound, unrivaled capabilities of JWST. It also found:
- The most distant red supergiant star, known as Quyllur, by looking in the field of the most massive early galaxy cluster known, El Gordo.
- The most distant object ever found to create a gravitational lens, a whopping 17 billion light-years away, with a picture perfect Einstein ring around it, formed by the lensed background object.
- The farthest cluster of galaxies ever discovered, so early on in cosmic history that it can only be considered a protocluster of galaxies, as it hasn’t completed formation, just a mere 650 million years after the Big Bang.
- The most distant galaxy to display the brightest emission line of all, of the n=2 to n=1 transition in hydrogen (the Lyman-α line), at just 330 million years after the Big Bang.
And the list goes on. In addition to its impressively beautiful, detailed views of the Universe, and these (and other) record-breakers as well, however, none of these are among the most important discoveries that JWST has revealed about the Universe.
That’s because none of them were unexpected; none of them were complete surprises. None of these discoveries taught us something about the Universe that we hadn’t imagined or anticipated that we’d find. But there are indeed (at least) four major discoveries — purely serendipitous discoveries, that only occurred because we dared to look at the Universe as we’ve never looked at it before — that have truly shown us aspects of the Universe that were total surprises. Because of them, our cosmic picture is forever changed.
We discovered JuMBOs — Jupiter-mass binary objects — within the Orion Nebula.
When JWST peered into the Orion Nebula, the closest large star-forming region to Earth, we knew it would reveal details that we had never seen before. Its high-resolution infrared capabilities meant it could reveal fainter, cooler objects than we’d ever seen before, while simultaneously peering through the dust-rich regions that would obscure those objects to optical telescopes. We knew that stars would be forming inside this region, and that many objects would exist that hadn’t become stars yet: either because they were protostars still in the process of formation, or because large, hot, more massive stars had blown away the material that was attempting to accumulate onto sub-stellar objects.
But when JWST found those substellar objects, or objects that were more like free-floating Jupiters than they were like stars or brown dwarfs, they got a shocking surprise. Of all the free-floating planets — or planets without parent stars — that JWST was able to directly detect, a whopping 9% of them were locked up in binary systems. It led, immediately, to a host of new follow-up questions.
- How did these binary systems come to exist?
- Were they somehow ejected from a planet-forming system together, as pairs?
- Are they examples of “failed binary stars” that didn’t form quickly enough but that remained bound together even as the surrounding material was photoevaporated?
These are questions we only know to start asking now, because prior to these JWST observations, we didn’t even know that binary Jupiter-mass objects were something that would, or could, exist in our Universe.
JWST uncovered the nature of a long-period gamma-ray burst, and it was merging neutron stars.
For decades, gamma-ray bursts had been divided into two fundamentally different categories: short-period bursts, which lasted 2 seconds or fewer (often much less), and long-period bursts, which lasted more than 2 seconds (often hundreds of seconds or more). We suspected that the short-period bursts came from near-instantaneous cataclysms, like merging neutron stars, and that was confirmed in a 2017 observations that revealed a kilonova event in gamma-rays, gravitational waves, and the afterglow that revealed ejected heavy elements. But 70% of gamma-ray bursts are of the long-period variety, and what were they?
Well, on March 7, 2023, two NASA observatories — Swift and Fermi — detected a long-period gamma-ray burst that endured for about 200 seconds, definitively in the long-period class. It was also an excessively bright burst: the second most energetic gamma-ray burst observed since they were first observed more than 50 years ago. Shortly thereafter, JWST observed the location where the burst occurred, and did so twice: 29 and 61 days after the burst occurred. Spectroscopy revealed the copious production of several very heavy elements, including tellurium, tungsten, and selenium. The only place known to create these elements is neutron star-neutron star mergers: the same events that trigger short-period gamma-ray bursts.
Perhaps all gamma-ray bursts, short or long, have the same type of origin, and we still don’t know why or how. That’s the new frontier, brought about only because of the incredible capabilities of JWST.
Supermassive black holes must arise independently from the stellar component of the galaxies they occupy.
The earliest black holes we were finding, even before JWST, posed a chicken-and-egg question: could these supermassive black holes have grown so big, so fast, so early on, if they only arose from the corpses of the first stars that formed in the Universe? Or could they have formed earlier than, and independently of, the stars that came to be in the Universe? In other words, did the earliest stars serve as the seeds for supermassive black holes? Or was it the other way around: did the earliest black holes form separately, serving as seeds for early galaxies, and then stars formed around them?
In late 2023, the chicken-and-egg question was answered for these supermassive black holes. With an assist from gravitational lensing in many cases, galaxies with very low stellar masses — as low as right around 10 million solar masses worth of stars — were copiously discovered. In particular one faint, lensed galaxy, UHZ1, was right at that low value, with only about 10 million solar masses for its stellar mass. But when the X-rays were measured from this galaxy, a mass estimate for the supermassive black hole could be obtained, where it weighed in at 9 million solar masses. For this young, distant object, whose arriving light was emitted a spectacular 13.2 billion years ago, the black hole was just as massive as the galaxy itself.
In other words, it couldn’t be that these supermassive black holes formed from stars; it must’ve been that (at least some) supermassive black holes formed independently of, and earlier than, the stars within those galaxies did. It’s an incredible discovery that we couldn’t have anticipated without having built JWST to explore the Universe.
An “asteroid belt” and a “Kuiper belt” might not be defining features in most planetary systems.
This might be the biggest surprise of all. When we think about our own Solar System, we have:
- inner planets,
- an asteroid belt,
- outer planets,
- a Kuiper belt,
- and then an outer Oort cloud.
In pretty much every system where stars had planets, we expected to find something very analogous. But with the very first star that JWST looked at that was known to have a dusty debris disk around it — something that could reveal the presence of belts — we got a rude awakening. The nearby star Fomalhaut was seen to not only have an asteroid-like belt and a Kuiper-like belt around it, but a third “intermediate” belt, too!
Why was this?
The mystery would only deepen when we looked at our second nearby star with a dusty debris disk: Vega. How many belts would it display? Would Vega have two belts? Three? It turned out that the answer to the question was another shock: zero. Vega’s disk appears smooth and almost featureless, with no belts at all, and only one very slight “faint ring” in the disk, indicating a lone planet that was at most just ~40% the mass of Neptune, with no other discernible features at all.
These latter discoveries are truly revolutionary: we didn’t even know to look for them, and yet when we did look, JWST was able to reveal them to us almost immediately.
Why is that? It’s because we built JWST and dared to observe the Universe in a way we never had before: with unprecedented power, resolution, and capabilities. As a result, there are now a number of ways that, since it began science operations in 2022, JWST has fundamentally changed the way that we view our Universe.
The story must not stop here: not with JWST in astronomy, and not for all of the ways we interrogate and investigate the scientific nature of our reality. There are tremendous holes in the scientific suite of observatories we have in space, and it will take a bold and ambitious vision for NASA science to bring us to the next level. Complementarily, there are numerous ground-based facilities that the scientific community is trying to build to improve what we know about the Universe: the Giant Magellan Telescope, the Thirty Meter Telescope, LIGO II, the next-generation Very Large Array, a new generation of CMB experiments, and much more.
Whenever we dare to look at the Universe as we’ve never done previously, there’s a chance that we’ll find something surprising, novel, and wondrous. Sometimes, those discoveries surprise even the most imaginative among us, and teach us lessons we couldn’t have predicted beforehand. When people ask about the value of engaging in a new scientific endeavor, they often want to know — with some sort of guarantee — what we’ll discover. But that’s not what makes science so exciting. The things we “knew” JWST would discover aren’t its greatest finds at all. We simply can’t know what “we don’t know” until we look and find out. It may be nothing. Or it could be game-changing. That’s the power, joy, and risk, all at once, of our scientific endeavors.
