You are sitting in the public gallery of a debating chamber, maybe thirty feet above the floor, watching a soft-spoken member rise to speak. The microphone picks her up, technically. The words travel to the ceiling-mounted speakers, bounce off the stone, arrive at your ears. And yet something dissolves in transit: the sense of command, the feeling that the room has decided to listen. The member with the baritone two rows along never loses that feeling. The chamber, for its part, never notices it has a preference.
The acoustic design of a parliament shapes which voices its amplification systems favour, through a combination of geometry, material choices, and the physics of human speech. This isn't conspiracy. It's an engineering problem that architects and sound designers have historically solved by optimising for a particular kind of voice, without quite meaning to.
The room before the microphone
Every chamber has a native acoustic. A behaviour it exhibits before a single cable is run or a loudspeaker mounted. Hemicycle chambers, like the French Assemblée nationale or the European Parliament's main hall, scatter sound in arcs and create long reverb tails. Adversarial chambers, like Westminster's House of Commons, are long rectangular boxes with parallel hard walls that produce strong early reflections. Both shapes create a baseline sonic environment that microphones and amplifiers are layered on top of, not replacing the room's character but wrestling with it.
The critical variable is the fundamental frequency of the human voice. Male voices in conversational speech typically sit between 85 and 180 Hz; female voices between 165 and 255 Hz. The overtones that carry intelligibility and presence extend upward from there, into the 2,000 to 4,000 Hz range. A chamber built from dense stone, hardwood panelling, and plaster reflects mid and high frequencies efficiently but absorbs very little bass. That sounds neutral. It isn't. Low-frequency voices, which already carry further in open air, get a natural acoustic boost from room reinforcement. Higher-pitched voices, sitting in frequencies that the room's surfaces handle less predictably, depend almost entirely on the electronic system to remain intelligible.
So when a sound engineer tunes the amplification, they are calibrating against a room that already has a thumb on the scale.
The microphone placement problem
Picture two members of a hypothetical legislature, both given equal speaking time. One is a tall man who projects from the chest, filling the chamber with sound at around 120 Hz before the microphone does anything. The other is a shorter woman whose fundamental sits closer to 210 Hz, with the clarity of her speech carried in the upper harmonics. Both stand at the same lectern. Same distance from the same cardioid condenser microphone.
The microphone has what engineers call a proximity effect: it boosts low frequencies for sources that are physically close, like a bassline turned up simply because the speaker is near. The taller speaker, whose voice is already low, benefits twice over. The higher-pitched voice, which needs those upper harmonics to project authority, gets less proximity boost and more competition from the room's reflected energy, which smears fine detail. Engineers compensate with EQ, but the compensation is set once for the room, not adjusted speaker by speaker. The system is tuned to a centre of gravity that doesn't sit at the centre of human vocal diversity.
This isn't hypothetical physics. Acoustic consultants working on legislative refurbishments in Canada and Australia have documented exactly this asymmetry when reviewing legacy sound systems, finding that intelligibility scores (measured by a standard called the Speech Transmission Index, or STI) consistently ran lower for higher-frequency voices in unrenovated chambers than in purpose-rebuilt ones where microphone arrays were placed at variable heights and the DSP processing was tuned per seat position.
The assumption worth challenging is that amplification solves the problem. It doesn't. Amplification makes everyone louder. Intelligibility is a different thing entirely, and the gap between those two words is where the politics lives. A voice can be loud and still be acoustically disadvantaged by a room that smears its particular frequency signature, turns its consonants to mush in reflected sound, or positions its microphone for a different body type. Volume is the easy fix. Clarity is the hard one, and most historic chambers never bothered.
The chambers that have taken this seriously, among them the Scottish Parliament's debating chamber at Holyrood, designed with adjustable acoustic baffles and distributed speaker arrays calibrated to multiple vocal profiles, demonstrate that the problem is solvable. It requires treating the room as a political instrument. Which is precisely what it is, and always has been.
A legislature that cannot hear all its members with equal clarity is not merely an acoustic inconvenience. It is, in a quietly structural way, a legislature that amplifies some arguments and muffles others before anyone has said a word.