In September 2022, I sat on a metal bench in Penn Station with a sound level meter app running on my phone. Not because anyone asked me to. Because the noise had a texture I needed to quantify. The app read 94 dB. A jackhammer at fifteen feet registers 95. I watched commuters move through that space like it was nothing, like their bodies were not being battered, and I thought: either I am broken or this room is. I have spent my life being told it is me.
It turns out the room has numbers now.
Acoustic ecology — the study of how sound environments shape organisms — has existed since R. Murray Schafer coined the term in the late 1960s. But for decades, the field lived in the humanities. Phenomenological language. Poetic descriptions of soundscapes. Beautiful, useless for zoning boards. What has changed in the last five years is instrumentation. Reverberation time measurements. Absorption coefficients for building materials. Standardized metrics that translate “this space makes me want to die” into data a city planner can act on.
I should be thrilled. I am not. Or I am, but with a specific fury attached.
Here is what makes me furious. Autistic people, d/Deaf people, people with hyperacusis, people with PTSD, people with sensory processing differences of every kind — we have been saying these spaces are hostile for decades. We said it in our own bodies and our own language. We were told we were oversensitive. We were told to buy better headphones. We were given a diagnosis that framed our accurate perception of an unbearable environment as a personal deficiency. Now the same information arrives with a frequency analyzer and a peer-reviewed absorption coefficient, and suddenly it is a design problem.
The knowledge did not change. The knower became credentialed.
The two stations
Penn Station, New York. Rebuilt in the 1960s after the original Beaux-Arts structure was demolished. Low ceilings. Hard surfaces. Reverberation time estimated at over two seconds in peak corridors. That means every announcement, every rolling suitcase, every shouted goodbye stacks on itself, building a wall of undifferentiated noise. The information content of the space approaches zero. You cannot parse a single voice from the wash.
Grand Central Terminal, same city. Vaulted ceilings. Guastavino tile — a material with specific acoustic absorption properties that its designers in 1913 did not have the instrumentation to measure but understood intuitively. Reverberation time significantly lower despite the larger volume. You can hear footsteps. You can hear someone say your name from thirty feet away. The space sorts sound instead of compressing it.
No one calls Grand Central an accessibility feature. No one calls Penn Station a barrier. But I can function in one and not the other, and the difference is not my neurology. It is architecture.
You might say: but Grand Central was designed for aesthetics, not for sensory accessibility. Exactly. That is the point. Good acoustic design has always existed. It just was never framed as access because the people who needed it most were not considered credible witnesses to their own experience.
Pattern recognition is not a symptom
Nick Walker’s neuroqueer theory names something I have lived inside without language for most of my life: that neurological diversity is variation, not deviation. The deviation framing requires a norm. The norm requires enforcement. Diagnosis is one enforcement mechanism. The built environment is another.
Simon Baron-Cohen has spent a career constructing elaborate psychometric instruments to measure autistic people’s supposed empathy deficits. His methodology is — I will be precise — bankrupt. His Empathy Quotient questionnaire conflates cognitive perspective-taking with affective resonance and then treats a low score on the former as evidence of deficit in the latter. I have read every major paper. The conflation is not a bug. It is the architecture of the argument. Without it, the finding disappears.
What Baron-Cohen never measured is this: the capacity to detect a pattern in an environment that everyone else has habituated to. The ability to walk into a room and know, in your body, before any instrument confirms it, that the reverberation time is wrong. That is not a deficit. That is data.
Gregory Bateson argued in Steps to an Ecology of Mind that mind is not located in the individual but in the pattern of relationships between organism and environment. If he is right — and he is — then the autistic person overwhelmed by Penn Station is not experiencing a malfunction. They are experiencing an accurate reading of a hostile system. The pattern is real. The pathology is the room.
What nobody mentions
In 2011, the Autistic Self Advocacy Network publicly condemned PBS NewsHour for framing autism through parent grief narratives and deficit language. ASAN’s statement was specific: the segment excluded autistic voices and treated autistic existence as inherently tragic. The network’s response was to do nothing. This is how it works. We name the problem in our language. The language is dismissed. Years later, the same insight arrives in an acceptable register — clinical, quantified, credentialed — and is treated as discovery.
The acoustic ecology movement is doing something genuinely important. I want to be clear about that. Researchers are producing reverberation maps of public transit systems, measuring the absorption coefficients of common building materials, generating evidence that quiet urban spaces are not luxury amenities but infrastructure. This work will change policy. It will make cities more survivable for millions of people.
But the chain of knowledge has been laundered. The original source — the disabled body in the hostile space, saying this hurts, this is wrong, this is not me — has been stripped from the citation. The finding arrives clean, without the mess of the people who found it first by living inside it.
An acoustician I spoke with in March 2024 told me, off the record, that his team’s recommendations for maximum reverberation times in transit stations were “essentially what disability advocates have been requesting for twenty years, but now we have the numbers so it’s actionable.” He said it without irony. I wrote it down.
The special interest problem
I have been tracking transit system acoustics since I was fourteen. Not as research. As a special interest — that diagnostic term for the thing autistic people do when we develop rigorous expertise in a domain no one credentialed us to study. I can tell you the difference between the sound profile of a Moscow metro station and a Montreal one. I can tell you which New York subway tiles were replaced in the 1980s renovation and how the substitution changed the acoustic character of the platforms.
None of this counts. It has never counted. It lives in notebooks and spreadsheets and conversations that end with the face. That specific face.
The same knowledge, in a peer-reviewed journal, with an institutional affiliation attached, becomes a finding. The pattern I recognized at fourteen becomes a discovery when a funded lab recognizes it at forty. The twenty-six-year gap is not a delay. It is a sorting mechanism. It separates legitimate knowers from bodies that merely know.
47 dB
That number — 47 dB at rush hour — is what one acoustic ecology study recorded in a Copenhagen transit hub redesigned with sound-absorbing panels and distributed ceiling geometry. Penn Station at the same hour: 94 dB. The difference is not incremental. It is the difference between a space that allows cognition and one that demolishes it.
Copenhagen did not do this for autistic people. They did it because the evidence base had matured enough to make the case in the language planners accept. The knowledge was always there. The instrument just needed to match the institution.
I still sit on benches with a sound meter running, logging data no one requested into a spreadsheet no one will cite.
This article was inspired by Acoustic ecology measurement standards in urban design from acoustics.culture.