The Photo Is the Exception, Not the Standard
Ole Eggers Bjælde, astrophysicist at Aarhus University, is plain about the numbers. "There are literally millions of black holes in our Milky Way galaxy. And we have maybe observed a little more than a handful of these." The ratio is roughly a million to one. And yet the existence of those unobserved millions is not in serious doubt.
The reason is that the evidence for them does not depend on seeing them directly. It depends on observing what they do. A black hole can be detected when it is pulling in surrounding gas, heating it to millions of degrees and producing x-ray emissions that telescopes can measure. It can be detected when its gravity measurably alters the orbit of a nearby star. In both cases, what you are observing is the effect, not the object. The object is inferred from the effect.
This is not a workaround or a fallback position. It is the normal operation of physics. We cannot observe electrons directly either. We observe their effects. We cannot observe the interior of the Earth directly. We infer it from seismic waves. The entire history of particle physics is a history of inferring the existence of things from the traces they leave in detection equipment. Indirect observation is not a concession to the limits of technology. It is how knowledge about unobservable things is actually built.
The Physics Does Not Wait for the Camera
Bjælde describes his own position with notable honesty: "If someone came and told me that these objects exist and I didn't have the background and I didn't know anything about it, I would probably not believe them. I would probably say, okay, in Sci-fi, sure. But not in reality."
He knows, from the physics, that they are real. The equations that predict black holes follow from general relativity, which has been tested to extraordinary precision in other contexts. The x-ray signatures of accreting black holes have been observed and catalogued for decades. The orbits of stars near the centre of the Milky Way are only explicable by the presence of an enormous dark object at their gravitational focus. The evidence converges from multiple independent directions, none of which involve directly photographing the object.
The 2019 image confirmed what the physics already predicted. That confirmation matters, not because it established the existence of black holes, but because it tested the specific prediction of what the shadow of an event horizon should look like when observed from the outside. It matched the model. That is a meaningful result. It is just not the first result. The knowledge came first.
Why This Matters Beyond Astronomy
The instinct to demand visual proof is not irrational. It has a sensible origin. In everyday experience, seeing something is one of the most reliable ways of knowing it is there. But everyday experience is calibrated to a particular scale, a particular range of energies and distances and timescales. Outside that range, the instinct becomes unreliable.
Black holes are not the only case. The same structure applies to dark matter, to the Higgs boson before it was detected at the LHC, to the existence of Neptune before it was observed (predicted from perturbations in Uranus's orbit), to gravitational waves before the first direct detection in 2015. In each case, the knowledge preceded the direct observation by years or decades. In each case, the knowledge was real.
The more general point is that science builds knowledge through multiple converging lines of evidence, and direct visual observation is one line among many. When those lines converge consistently from independent directions, the conclusion is well-established, regardless of whether a camera has ever pointed at the object in question.
The Millions We Will Never See
Most of the black holes in the Milky Way are silent. They are not pulling in gas, not perturbing nearby stars in measurable ways, not producing any signal that current or near-future technology could detect. They are simply there, dark and inert, in the vast empty spaces of the galaxy. We know they exist because the physics of stellar evolution tells us what fraction of stars are massive enough to produce them, and we can count the stars. The inference is not uncertain. It is well-constrained.
Bjælde is direct about what this means for the future of detection: "Maybe we need to come up with a new method of detecting those black holes. That would also be hugely interesting." The honest position is that we will probably never observe most of them. The knowledge that they exist does not require it.
What does require the observations is the next layer of understanding: what exactly happens at the singularity, whether the theory of quantum gravity changes the picture inside the event horizon, how the Milky Way's history of stellar formation shaped the current distribution of black holes through the galaxy. Those are questions that need more data. The basic existence of the objects is settled.
The gap between the one black hole we have photographed and the millions we know are out there is not a gap in knowledge. It is a gap in resolution, in the current limits of detection technology. Those limits will move. The knowledge will not wait for them.