Peter Aczel published his first audio review in the 1950s. He reviewed amplifiers, turntables, speakers, cables, tape machines, and eventually compact disc players. He built his own equipment. He listened to other people’s equipment. He compared measurements to listening impressions across every format change that occurred in recorded sound during his working life.
In 2015, at the end of six decades of reviewing, he published a numbered list of what he had learned. The list contained ten items.
Item three: “The principal determinants of sound quality in your listening room, given the limitations of a particular recording, are the loudspeakers — not the electronics, not the cables, not anything else.”
He noted that a new amplifier would not change anyone’s audio life. It might produce a very small improvement, usually not, unless the old amplifier was badly designed. The basic sound of a system would remain the same. Only a better loudspeaker could change it.
He offered a guess as to why this finding had not prevailed among audiophiles: a new pair of loudspeakers presents a problem in interior decoration. Swapping amplifiers is simpler and more acceptable to spouses. The anticipation is just as high. The disappointment takes longer to arrive.
Item two: “The principal determinants of sound quality in a recording produced in the last 60 years or so are the recording venue and the microphones, not the downstream technology.”
The size and acoustics of the hall, the number and placement of the microphones, and the quality and level setting of the microphones would have a greater influence on the perceived quality of the recording than whether the signal was captured on analog tape, digital tape, hard drive, or direct-to-disk cutter. The proof: some classic recordings from the 1950s and 1960s sound better, more real, more musical, than average super-high-definition recordings made today.
In the signal chain from musical performance to ear, everything between the microphone and the loudspeaker is interchangeable. The two transducers — the points where air vibration becomes electrical signal and electrical signal becomes air vibration — determine the result. Aczel spent sixty years confirming this. The middle of the chain does not matter.
In 1895, Wallace Clement Sabine measured the reverberation time of the Fogg Lecture Hall at Harvard University. The hall had been built three years earlier and its acoustics were terrible — spoken words blurred into incomprehensible echoes. Sabine carried seat cushions from a nearby theater into the hall, one load at a time, at night, measuring how the reverberation changed as absorptive material accumulated. He derived the formula that bears his name: reverberation time equals room volume multiplied by a constant, divided by total absorption.
The formula requires knowing the absorption coefficient of each surface material — the fraction of incident sound energy absorbed rather than reflected. This is a number between 0 and 1. Zero means the surface reflects everything. One means it absorbs everything.
In 1913, the Riverbank Laboratories in Geneva, Illinois, built the first reverberation chamber specifically for measuring absorption coefficients. Other laboratories followed. By the 1930s, a standard method existed: place a sample of material in a specially designed room, measure the reverberation time with and without the sample, calculate the absorption coefficient using Sabine’s formula.
The measurements routinely produced coefficients above 1.0.
A coefficient above 1.0 means the material absorbs more sound energy than strikes it. This is physically impossible. It violates conservation of energy. A material cannot destroy more wave energy than it receives.
The explanation is known. Sabine’s formula assumes that sound energy in the room is perfectly diffuse — striking all surfaces equally from all directions. In practice, this never happens. Sound waves diffract around the edges of the test sample, making the effective absorbing area larger than the geometric area measured by a tape measure. The formula divides the total absorption by the geometric area, producing a number inflated by the edge effects.
The Acoustical Society of America published a standard test method for measuring absorption coefficients. The standard acknowledges the problem. It states: “Diffraction effects usually cause the apparent area of a specimen to be greater than its geometrical area, thereby increasing the coefficients measured according to this test method. When the test specimen is highly absorptive, these values may exceed unity.”
The standard has been revised repeatedly since its first publication. The problem has been analyzed in hundreds of papers across a century. In 2023, a review in Acoustics Today described it as “perhaps the most controversial and long-lasting debate in the history of acoustics.”
The absorption coefficients measured by this method continue to be used by architects and acoustic consultants worldwide. Concert halls, classrooms, offices, recording studios, and hospitals are designed using numbers that regularly exceed their own theoretical maximum. The buildings sound as intended.
In 1832, Adolphe Quetelet needed a way to compare body weight across people of different heights. Quetelet was a Belgian mathematician and astronomer. He was not a physician. He was studying the distribution of human physical characteristics across populations — what he called “social physics.”
He tried dividing weight by height. The result varied too much with height. He tried dividing weight by the cube of height. The result varied too much in the other direction. He divided weight by the square of height. The distribution was the most stable.
Weight divided by height squared. Quetelet published this formula as a statistical convenience for normalizing population data. He did not propose it as a measure of health.
In 1972, Ancel Keys, an American physiologist, analyzed data from over 7,400 men across five countries. He tested multiple indices for predicting body fat percentage. Quetelet’s formula performed best among the simple options — not because of any biological insight, but because the square of height happens to correlate better with body fat than height alone at the population level. Keys gave the formula a new name: body mass index.
In 1995, the World Health Organization adopted BMI as the global standard for classifying underweight, normal weight, overweight, and obesity. The cutoff points — 18.5, 25, 30 — were chosen based on statistical associations with disease risk across large populations.
BMI does not measure fat. It does not distinguish muscle from adipose tissue. A person with 8% body fat and significant muscle mass registers the same BMI as a person with 35% body fat and minimal muscle mass. It does not account for where fat is stored — visceral fat around the organs is far more dangerous than subcutaneous fat under the skin, but both contribute identically to the number. It does not adjust for sex, age, or skeletal frame. It was not designed by anyone who studied the body.
In 2026, BMI remains the standard measure used by physicians, insurance companies, public health organizations, and research studies worldwide to classify body composition. It was 194 years old and had never once measured what it claimed to measure.