The Taos Hum: The Sound Only Some Can Hear

Ask anyone in Taos, New Mexico, who hears it, and the description barely varies: a low, steady drone, somewhere between an idling diesel engine and a large appliance running in the next room, loudest at night, indoors, once the rest of the world has gone quiet. It doesn’t rise and fall with traffic. It doesn’t point toward any single building or road. For the people who hear it — by the best surveys, somewhere around two per cent of the town’s population — it isn’t a faint background irritation to shrug off. Sufferers have reported insomnia stretching for months, headaches, nosebleeds, and in a handful of cases enough distress that they left Taos for good. Stand in the same room as a “hearer” and you will very likely hear nothing whatsoever. The sound exists entirely inside a minority of skulls, and that asymmetry is the whole of the mystery.

Taos gave the phenomenon its name, but it was never alone. Bristol, England, had a hum epidemic in the late 1970s that put dozens of residents into doctors’ surgeries. Kokomo, Indiana, had one in the early 2000s. Windsor, Ontario, has had one running for over a decade. Call it what you like — the Bristol Hum, the Kokomo Hum, the Windsor Hum — the pattern repeats: a low-frequency drone, heard by a self-selected minority, absent from every recording made by every microphone anyone points at the air. What happened next in Taos is instructive, because for once the state didn’t ignore its citizens. It sent scientists.

The investigation that found nothing

By 1992 enough Taos residents had complained to their congressional delegation that Senator Jeff Bingaman and Representative Bill Richardson secured federal funding for a proper study. In 1993 a team drawn from the University of New Mexico, Sandia National Laboratories, Los Alamos National Laboratory, and the Phillips Laboratory at Kirtland Air Force Base, led by the physicist Joe Mullins, moved into town with seismographs, electromagnetic sensors, acoustic recorders, and a deliberately assembled panel of self-identified hearers to test against a control group. The point of the study was to find whatever the hearers were responding to, wherever that search led.

It found nothing that matched. The instruments picked up ordinary background noise, industrial hums from known local sources, seismic microtremors — all present, all explicable, none of them correlating with when the panel’s hearers said the drone was loudest. No unexplained electromagnetic signal turned up. No infrasound source large enough to explain a phenomenon reported across a wide area presented itself. Mullins’s team could not identify what the hearers were hearing, and they said so plainly in their report. That plain admission — “we cannot find it” rather than “there is nothing to find” — is the hinge the entire modern legend swings on, because an honest gap in the data reads, to a mind primed for pattern, exactly like a cover-up.

The fork: from a real secret to an unfalsifiable one

The theories that filled the gap borrowed real technology and stretched it past its actual capability, and that borrowing is worth taking seriously rather than dismissing outright, because part of it is true. The United States Navy really did operate a covert extremely-low-frequency transmission system, Project ELF, running from sites at Clam Lake, Wisconsin, and Republic, Michigan, from 1982 until its decommissioning in 2004, generating powerful electromagnetic signals in the 76-hertz range specifically because such frequencies can penetrate seawater to reach submerged submarines. That project was genuinely classified in its early years, genuinely powerful, and genuinely capable of producing effects at a distance that most people had no way to verify for themselves. It is a real, documented case of the government running exactly the kind of low-frequency system that hum theories imagine.

The fork happens at the leap from “such systems exist” to “such a system explains this specific acoustic sensation in this specific town.” Project ELF’s transmitters were in the upper Midwest, thousands of kilometres from New Mexico, tuned to frequencies engineered for underwater propagation rather than airborne hearing, and switched off entirely in 2004 — a full decade after Taos hearers had already been complaining. HAARP, the Alaska-based High-Frequency Active Auroral Research Program that gets folded into nearly every unexplained-signal theory going, studies the ionosphere using radio frequencies far outside the range of human hearing, from a facility over 2,700 kilometres away. Neither system, on the physics alone, can produce a localised low hum audible to a minority of residents in a small New Mexico town. The theory survives on its own unfalsifiability: a secret transmitter is by definition invisible to the instruments sent to find it, so every negative result reads back as confirmation of the secrecy rather than evidence the theory is wrong.

The journey: hums that got solved, and the ones that didn’t

Bristol’s case came first and set the pattern everyone since has followed. Complaints there began piling up through the late 1970s, enough that the city council and the Department of the Environment took an interest, and enough that the story made national newspapers by 1979 and 1980. Investigators tried microphones in bedrooms, seismographs on windowsills, and interviews with dozens of sufferers, and came away with a result Taos would echo over a decade later: no consistent external source, no shared direction, no correlation with weather, traffic, or the electricity grid. What the investigators did notice was a demographic skew. Hearers tended to be older, tended to be more often women than men, and tended to describe the hum as worst in absolute silence — in bed, at night, with double glazing shutting out the traffic that might otherwise mask a faint internal sound. That detail points toward the ear itself rather than the air outside the window.

The geophysicist David Deming, of the University of Oklahoma, gave the wider phenomenon its first serious academic treatment in a 2004 paper for the Journal of Scientific Exploration, cataloguing hum reports from Taos to Bristol to Auckland, New Zealand, and finding the same demographic pattern recurring at every location: hearers skewed older, complaints clustered indoors and at night, and standard acoustic instrumentation kept failing to register anything at the reported frequency. Deming leaned toward a real, if elusive, environmental or physiological source rather than mass delusion, and his paper is still cited in nearly every serious treatment of the subject since, largely because it treated the sufferers’ reports as data worth investigating rather than dismissing on sight. That even a peer-reviewed academic survey, working from hundreds of independently reported cases across several countries, settled on a plausible physiological guess rather than a demonstrated mechanism is itself part of why the mystery has outlasted every attempt to close it.

What makes the mechanism of this belief visible is that some hums, elsewhere, did get solved, and the solving was almost dispiritingly ordinary. The Kokomo Hum, which had residents of that Indiana city reporting the same nocturnal drone throughout the early 2000s, was traced in 2003 by the acoustical engineering firm Acentech to two identifiable industrial sources: a cooling fan at the DaimlerChrysler foundry and an air-handling unit at the nearby Haynes International plant. Once the specific equipment was modified, complaints dropped. The Windsor Hum, reported across the Canadian side of the Detroit River from around 2011, was linked by a University of Windsor research team led by the engineering professor Colin Novak, publishing findings around 2014, to blast-furnace operations on Zug Island, the heavily industrialised strip of land on the Michigan side of the river. Both cases had a real physical source, a real fix, and a mundane ending involving fan blades and furnace vibration rather than a listening device or weapon.

Taos never got that ending, and neither, fully, did Bristol. No factory, no furnace, no fan has ever been identified as either town’s culprit, and researchers who examined Bristol sufferers directly in the years after the initial panic increasingly focused less on an external sound source and more on the sufferers themselves, noting that some cases resembled a form of low-frequency tinnitus rather than an environmental noise at all — a signal generated somewhere in the auditory system itself, made audible only by the near-total silence of a quiet bedroom in a quiet town. That is an unglamorous, only partially satisfying answer: a genuine perceptual phenomenon, plausibly rooted in the ageing auditory system or in unusually low ambient noise, rather than a wilfully hidden signal. It resolves the mystery without resolving the discomfort of the two per cent who still, every night, hear something the rest of the room does not.

What the gap is really doing

The reason “possibly a subtle physiological phenomenon in older ears, in an unusually quiet town” satisfies almost no one is a matter of proportion. A sensation this disruptive — the insomnia, the headaches, the sense of being trapped with a noise nobody else will acknowledge — registers to the person suffering it as significant, and significant experiences want significant causes. Human cognition runs on a rough rule of thumb that big effects need big explanations; it is the same instinct that makes a president’s assassination feel too large to have been fired by one lone, mediocre gunman, and it functions the same way here. A tired auditory nerve is not a proportionate answer to months of ruined sleep. A concealed military transmitter is. The theory isn’t chosen because the evidence points there; it’s chosen because it is the only explanation on offer whose scale matches the suffering.

There is a second ingredient, just as important: an unfalsifiable cause cannot let its believer down. Every negative test the University of New Mexico team ran in 1993 could be filed, by someone already convinced, as evidence of a cover-up rather than evidence of absence, because the theory was built to accommodate exactly that outcome. A hum with a boring, findable cause — a fan, a furnace — can be switched off and the belief along with it, the way Kokomo’s was. A hum attributed to a secret and by definition undetectable system never runs out of road, because it was never staked to anything the world could actually show you.

There is also a quieter reward on offer to the hearer alone, worth naming honestly. To be one of the two per cent who can perceive what the instruments and the ninety-eight per cent around you cannot is, whatever the discomfort involved, a strange kind of distinction. It recasts a medical oddity as a private sensitivity, a heightened attunement to something real that the wider world is simply too blunt to register. That reframing costs nothing and offers real comfort, which is exactly why it recurs everywhere this pattern shows up, from hum sufferers to those who report unexplained illness clusters decades apart and continents away.

None of that makes the hearers of Taos dishonest, or their nights any less ruined. The two per cent who lie awake to a drone the rest of the town cannot register are reporting something real to them, whatever its ultimate source turns out to be, in the way that the Havana Syndrome sufferers a generation later reported symptoms that were entirely real regardless of whether a weapon caused them. What the Taos Hum actually demonstrates, forty years and one federal study on, is how a sound too small for an instrument to catch can still be too large for a mind to leave unexplained — and how, in that gap between sensation and proof, a listening public will always find room to hear an engine, a transmitter, or a shortwave signal broadcasting from somewhere it was never meant to be traced, before it will settle for hearing nothing at all.