We’ve compiled the crash course in everything we’ve been talking about. Will it make music sound better? Maybe, but only because you know what is happening a little bit better now.

Twenty-two words cover everything (almost) that we have shipped out in newsletters.

None of this is required. You do not need to know what a tritone is to get chills off a Black Sabbath riff. Or why some bass lines feel thicker than others. But there is a particular pleasure in finally having a name for a thing.

Most of these started as a newsletter. Where they did, we have linked it.

The pickup. The little upbeat that lands before the downbeat.

Composers have used it for centuries to make a room lean in.

It is why a phrase can begin on the "and" of beat four of the previous bar, so the new bar is already three syllables deep before the snare hits. Andre 3000 lives there. Earl Sweatshirt's SOME RAP SONGS is almost entirely anacrusic, which is how he manages to sound like he is mumbling and rewarding close listening at the same time.

The production version is pre-roll silence. A bar of nothing dropped in right before the chorus. Longer journey, harder landing.

How to hear it: count yourself in out loud on any track with a strong vocal entry. If the words start before your "one", that is an anacrusis.

The full stop. The whole band cutting out at once.

Also called the grand pause. Not a gap in one instrument. Everything, at the same time, on purpose.

James Brown was a master at it. The band stops dead and waits while he decides when to bring them back. The crowd loses it every time, because at that point the music has stopped being something you are listening to and become something you need. Anticipation turned all the way up.

There is a world of difference between an accidental silence, of which there are plenty on record, and a timed one. A gap is an accident. A caesura is a producer who knows you are holding your breath and decides to make you hold it longer.

How to hear it: watch who claps in the pause at a gig. Those people have not heard the song enough times to know where the space falls. It catches us all from time to time…

→ Fermata

Sound triggers an involuntary color experience.

A subtype of synesthesia, which is the broader condition where one sense involuntarily triggers another. You may have heard of grapheme-color synesthesia, where letters and numbers each have a fixed color. Chromaesthesia is the same wiring pointed at timbre, notes, chords, and whole tracks.

Franz Liszt used to tell his orchestra to play bluer, or not so rosy. He was not being purposely vague or obtuse. He was seeing color as he conducted and trying to get the music to match. His musicians thought he had lost it. He had not.

Amy Beach had a color for every key. F sharp was red. D major was yellow. C major was white. Not roughly. Exactly. Every time.

Pharrell describes seeing color as geometric shapes. Frank Ocean has talked about the color of specific chords. Billy Joel says his moods have colors and his music follows them. To all of them it is just how music works.

Here is the part that matters for the rest of us. You do not need synesthesia to hear in color.

A 2013 study in the Proceedings of the National Academy of Sciences found that non-synesthetes show consistent cross-modal associations between music and color, driven largely by shared emotional response. Faster music in a major key pulls yellows and warm tones. Slower music in a minor key pulls blues and dark greens. Consistently. Across participants. Without any of them realizing they were all doing the same thing.

Think about the last time you called a track dark. You were not describing the frequency response. You were describing a color the music made you think about.

How to hear it: close your eyes, play thirty seconds of anything, and note the first color that arrives. Don’t overthink it, or try to match the color and the sound. It’ll just happen.

→ Chromaesthesia

"The devil in music." The medieval name for the tritone.

See tritone. The nickname is the interesting part and it has been misunderstood for about five hundred years.

Nobody in the medieval church believed the interval summoned anything. It was kept out of sacred composition because it doesn’t resolve, and unresolved tension read as morally unserious to ears trained on consonance. It was a rule about taste that got a name about hell, and the name stuck because it was better.

The difference between the quietest and loudest parts of a record.

Most music made before 1990 has 12 to 18dB of it. Most music from the peak of the loudness war has 2 to 6dB. Your ears can tell the difference even if you have never thought about it, which is why older records feel different and why some people insist they ‘feel’ better.

Dynamic range is what makes a loud moment loud. Without a floor there is no ceiling worth reaching. A chorus can only land if the verse gave it somewhere to arrive from.

It is also the thing you cannot get back. Volume is a control you have. Loudness is baked into the recording. Turn a crushed record up and it gets louder. It does not get better.

Vinyl had physical limits and those limits were doing everyone a favor - though, maybe we didn’t appreciate it at first.

Loudness on a record is dictated by groove spacing. A louder cut means wider grooves, which means less audio per side. Look at the spacing on a twelve-inch club pressing sometime, and you can see it. The format said no on your behalf.

CDs had no such constraint. Digital gave engineers a hard ceiling and nothing stopping them running right up to it. So between roughly 1995 and 2010 they found out exactly how far they could push before it broke.

The logic was competitive and on its own terms it made sense. On a radio playlist or a shuffled CD changer, the louder track grabbed attention. Nobody wanted to be the quiet one.

DEATH MAGNETIC in 2008 is the case study everybody uses, because it is the clearest one.

The record arrived brick-walled in the mix, before it even reached the mastering engineer, who subsequently distanced himself from it in public. 16,000 fans signed a petition asking for a remaster. The band was not especially bothered.

Then something genuinely cool happened. Players ripped the audio stems out of GUITAR HERO III, which used a different, unsquashed set of mixes, and discovered the game version sounded like the record everybody thought they had bought.

The good version existed. But only because fans made it so.

How to hear it: play something from 1985 and something from 2008 back to back at matched volume. The older record will have quiet bits in it. That is not a fault, that is the floor doing its job.

→ Crushed Headroom

A hold of unspecified length. The performer decides when it ends.

The notation symbol, a dot with a curved line over the top, that tells you to hold a note or a rest past its written duration. No fixed length attached. Sometimes nobody decides and the note simply stays until it is finished.

It sits above a note, a chord, or a rest. Sometimes it sits above a bar line, which means hold the silence rather than the sound.

Live, singers who can hold a note longer usually do, because when it is live they can perform the track as it was probably meant to be heard.

The interesting part is not the length. It is the handover.

Almost every other marking in notation is an instruction with a defined value. Tempo, dynamics, articulation. A fermata deliberately withholds that, which means the emotional weight of the moment gets decided in the room, by the person performing, in front of you.

That is why it feels different. And usually two performances aren’t the same.

When the music stops doing the work your brain takes over. It fills the gap, it holds the tension, and it keeps time even though there is nothing to keep time against.

Loudness is relative. The same bass note sounds quiet after a wall of sound and enormous after a bar of nothing. Contrast does the heavy lifting, which is why silence only works if the record has dynamic range left to spend. A master squashed flat against the ceiling has no floor to drop to, so there is nowhere for the quiet to go.

That is the connection most people miss between the loudness war and why so much modern music feels relentless. It is not just that it is loud. It is that you can’t turn it down without it still being a big fat flat line of LOUD. Nightmare.

A Nature Neuroscience study on music and dopamine found the chemical releasing at two separate moments. Once in the nucleus accumbens at the peak, when the sound actually arrives. And earlier, in the caudate, during the anticipation, before the payoff has sounded at all.

The reward fires in the silence too.

Which means when a producer holds a gap for an extra beat they are not just building tension in some abstract artistic sense. They are extending the part of the experience your brain is already enjoying.

How to hear it: IN THE AIR TONIGHT sits behind minutes of near-nothing before the fill lands. Time how long you spend waiting against how long the fill actually lasts. The ratio is absurd and it works completely.

→ Fermata

Let the sound bloom, then cut the tail before it can decay.

Reverb makes a drum sound like it is in a room. A gate slams the door on that room mid-sentence. You get the size without the mess.

It is the trick behind that IN THE AIR TONIGHT fill, and behind every eighties record you can identify from two seconds of drums.

The real point is control over the air at the edges of a sound. Whether a chord rings into nothing or gets choked. Whether room tone stays in or gets edited out. That is a production decision, not an accident.

→ Fermata

The gap between the loudest peak in a record and the point where it starts to clip.

Think of the space between the top of your head and the ceiling. When there is enough of it, the loud parts actually sound loud. When there is not, everything sits at the same volume and there is nowhere left to go.

Our brains love contrast. We are built for highs and lows. We are considerably worse at processing a solid wall of sound, which is why a crushed record gets tiring rather than exciting.

A well-mastered record keeps 3 to 6dB of headroom before final limiting. That gap is where transients live.

A transient is the initial spike of a sound before the note itself arrives. The crack of a snare, the pluck of a string, the click at the front of a kick. Your brain reads the transient first and uses it to work out what instrument it is even listening to.

Crush the headroom and you crush the transients along with it. The snare stops sounding like a snare and starts sounding like a shape where a snare used to be.

Nobody writes petitions about transients. They just say the record sounds like plastic and they do not know why.

The second casualty is stamina. A crushed record does not get exciting, it gets tiring, and you turn it off earlier than you meant to.

How to hear it: OK COMPUTER. KID A. YANKEE HOTEL FOXTROT. THIRD. LATERALUS. IOWA. CRACK THE SKYE. All of them came out during the war. All of them kept their range. All of them sound enormous. Nobody has ever petitioned for a remaster of any of them.

→ Crushed Headroom

Loudness Units relative to Full Scale. How platforms measure perceived loudness.

Not peak level. Perceived level, weighted roughly the way human hearing works, averaged across a whole track.

Peak level tells you how tall the highest spike is. LUFS tells you how loud the thing actually feels. Those are very different numbers and only one of them matters to a listener.

Every major platform normalizes to a target. Spotify sits at -14. Apple Music at -16. Amazon at -14. Whatever you deliver, the platform adjusts it toward that number.

Which means the loudness arms race is over and nobody told half the industry. Master at -6 LUFS and Spotify turns you down by 8dB. You do not end up louder than anybody else. You end up flatter than everybody else at the same volume.

Loudness is a setting you can undo. Dynamic range, once it has been compressed out of a master, is gone. Bands gave up the breath in their records for nothing.

→ Crushed Headroom

Multiple pitches on one syllable.

One word, several notes, no break. When Whitney Houston stretches the word "I" across five seconds and about four different pitches at the start of I WILL ALWAYS LOVE YOU, that is melisma.

Most singing is syllabic. One syllable, one note. You say a word, it sits on a pitch, you move on. Melisma breaks that. The syllable stays open and the pitch moves underneath it.

Gospel built it into what we recognize today. Whitney perfected it. A generation of talent show contestants misused it for twenty years and nearly buried it, which is how you know it was powerful enough to survive.

The lineage is a straight line and it starts in church.

Mahalia Jackson shaped most of what we hear today. Her phrasing is the template. Aretha Franklin carried it into secular music, and she grew up the daughter of Reverend C.L. Franklin at New Bethel Baptist in Detroit, which is not incidental. Whitney Houston learned it from her mother Cissy, who sang gospel her entire life. Beyoncé knows exactly where it came from and exactly what she is doing when she uses it.

Every deep, rich, extended note you have ever felt in your chest pulls from way back in the pews.

Melisma is pitch movement on one syllable. Sustain is holding one pitch for a long time. They frequently happen together, which is why they get confused, but they are separate skills.

I WILL ALWAYS LOVE YOU does both, in sequence, which is why the performance gets studied. The opening is melisma: multiple pitches, one word, stacked and released. The final chorus is sustain: chest voice into mixed voice into head voice, one breath, no gap, no scramble, around twelve seconds.

Bill Withers holds a high E for eighteen seconds on LOVELY DAY. It sounds effortless, which is the trick. You do not notice the athleticism because he refuses to perform it as athleticism.

Chester Bennington holds a non-melodic scream for seventeen seconds on GIVEN UP, dropping octaves as it extends and losing none of the force.

Al Green on LET'S STAY TOGETHER is the quietest version of the same idea. The falsetto arrives, everything else drops back, and the note just floats. He barely pushes it and it stays anyway. A long note does not have to be loud to be obscene.

Two things are happening. The first is resonant frequency. The second is breath.

When you are properly tuned into a performance you hold your breath alongside the singer. Nobody asks you to. Your body just does it, and by the time they release the note you have been holding tension for the same duration they have.

That is the whole reason it works. You are not observing the note. You are participating in it.

How to hear it: sing one note and time yourself. Most people run out somewhere around six seconds. Bill Withers did three times that, on a high E, and made it sound like he was not trying. The gap between what you just managed and what is on the record is the point of the exercise.

→ An Obscene Sustain

A note your brain builds out of the harmonics sitting above it. It was never in the room.

Every note is a fundamental frequency plus a stack of harmonics, at integer multiples of that fundamental. A bass note at 100Hz carries harmonics at 200, 300, 400 and 500Hz.

Your phone speaker cannot reproduce the 100Hz. It can reproduce everything above it. Your brain receives 200, 300, 400 and 500, recognizes the pattern, works out what the missing bottom note must be, and generates it internally.

You hear a note that was never in the room.

Dutch physicist J.F. Schouten described the phenomenon properly in 1940. Licklider expanded on it in 1954.

But the story worth telling is Bell Labs in the mid fifties. Miller and Licklider were given a straightforward brief: define the minimum acceptable frequency response for a telephone. Not perfect. Just good enough. Bell wanted to know which corners they could cut and still have a phone call that worked.

They set the upper limit at 3,400Hz. Fine. Then they looked at the bottom end and found they could start as high as 340Hz.

Here is why that is strange. A man's voice has a fundamental of roughly 125 to 150Hz. A woman's sits about an octave above. So a phone system starting at 340Hz was cutting the fundamental frequency of the human voice out completely.

And listeners still identified male voices perfectly. Still heard the pitch. Still heard the depth.

The brain was not receiving the fundamental. It was calculating it.

Play the harmonics into separate ears, splitting them so neither ear receives the full stack, and your brain will still combine them and produce the missing fundamental.

That is the detail that proves it is a construction rather than a physical artefact of the speaker or the ear canal. Your auditory system is doing pattern completion on incomplete data, the same way your visual system fills in your blind spot without telling you.

Most phone speakers cut out below roughly 80Hz. Laptops are worse. Producers mixing for streaming know exactly where their audience is listening, so they are not only mixing for monitors. They are mixing for your brain, placing harmonics in the range your speaker can reproduce, knowing you will manufacture the note underneath and hand it back to yourself as real.

This is also why some records feel physically present on small speakers while others sound thin. It is not the speaker, it is the mix. The difference between a mix that travels and one that does not is partly this. The ones that travel are built for your hallucination.

How to hear it: play SWIMMING POOLS on your worst speaker. Cut everything below 100Hz on an EQ. If it barely changes, that is not you doing it wrong. That is what you’re looking for. The speaker was never producing it. Then put your hand flat on the speaker while it plays. You will feel almost nothing while your brain continues insisting the bass is there.

→ Involuntary Hallucination

The negative space between the kick and the snare, measured in milliseconds.

Where a rapper, a singer, an organ player, or anyone with a melodic line decides to land their notes in relation to the drums underneath them.

Drum machines are perfect. Humans are not. Quantize exists to drag every note back onto the grid automatically. Pocket is what happens when you understand space and timing on a whole new level.

Get it wrong, and the song feels stiff. Karaoke energy - ‘cause we’re a beat behind and we know it. Get it right and the whole thing breathes. Get it disgusting, and you find yourself rewinding seven times trying to work out what just happened to your chest.

At 90 BPM there are 666 milliseconds between beats and roughly 166 milliseconds between sixteenth notes. That is your working space.

On the grid is quantized and stiff. Behind by 10 to 30ms is laid back, which is where hip-hop, R&B and neo-soul live. Ahead by 10 to 30ms is urgent, which is punk, drum and bass and drill. Past 30ms in either direction, you have stopped making a decision.

The difference between the two is consistency. A musician in the pocket is late by the same amount every single time. Somebody who is just late is late by a different amount every bar, and your body can tell instantly even if your brain cannot explain why.

J Dilla turned quantize off entirely on his MPC3000 and played the drums in by hand. That much is fairly well known. The part that matters technically is what he did with the two elements.

His snares often sit 30 to 50 milliseconds behind the grid. His kicks often sit ahead of it. Two hands of the same human pulling the grid in opposite directions at the same time, and your ears have to negotiate it live.

Anyone can drag a snare. Almost nobody can drag a snare, push a kick, and keep a vocal pocketed against both.

D'Angelo holds that feel for 78 minutes straight on VOODOO. Every element is placed with intent, and the whole thing sits so far behind the beat that first-time listeners frequently think something is wrong with the playback. Nothing is wrong with the playback. Most pocketed record ever made, and we will die on this hill.

When timing is consistently placed behind the beat your body reads it as relaxed, because that is how relaxed movement works in the real world. Pushed timing reads as urgent for the same reason. You are not analyzing milliseconds. You are recognizing a physical behavior you have seen ten thousand times in people walking, dancing and talking.

How to hear it: strip the drums from a Pusha T verse. The internal rhythm is still there, fully intact. Do the same to a verse from someone who was simply behind the beat by accident and you will hear somebody drifting, speeding up, correcting, drifting again. One of them made a decision. The other one had an accident that got left in.

→ The Disgusting Flow Pocket

The study of how you perceive sound, as opposed to what the sound is actually doing.

The gap between the physics and the experience. A speaker moves air in a measurable way. What arrives in your head is a construction, and the two are not the same thing.

Almost everything else on this page is a psychoacoustic effect. The bass you hear that is not there. The silence that feels loud. The note that sits in your sternum. The color that arrives with a chord.

It is also why arguments about whether something sounds good are so hard to settle. You are both hearing a version your brain built, and the construction process is not identical.

The pitch that activates a physical object. Including you.

Everybody has one. A sustained note at the right pitch activates your chest cavity, your skull, and the room you are sitting in. Some notes land in your sternum. Some sit behind your eyes.

It is why one person's favorite moment in a song does nothing for the next person. What a Hozier yell does to some listeners, it will not do to others. We all prefer certain pitches, and part of the joy is finding the one that hits you exactly right.

How to hear it: hum slowly, from your lowest note upward. Somewhere in there, your face will buzz a little. That one is yours - cool, huh!

→ An Obscene Sustain

One sound automatically turns another one down.

A compressor reduces volume when a signal gets too loud. Normally it listens to the sound it is controlling. With a sidechain it listens to something else.

Feed the kick into the sidechain input, put the compressor on the bass, and every kick hit ducks the bass for a few milliseconds. Kick stops, bass comes back. Breathing.

Sidechain compression was not invented for music. It came from broadcast, where it is called ducking. The DJ starts talking, the music automatically drops underneath the voice, the DJ stops, the music comes back up. No engineer riding a fader. Purely functional, and a genuinely elegant solution to a problem that had nothing to do with groove.

Mixing engineers borrowed it because the kick and the bass live in the same low frequencies and fight for the same space. Sidechain the bass to the kick and the kick gets a clean pocket to punch through. Done properly you cannot hear the technique at all. You just hear a low end that is defined instead of soupy.

For decades that was the whole job. Invisible plumbing.

French house producers worked out that if you stopped being polite about the gain reduction, the ducking became a rhythm in its own right. Daft Punk built grooves out of it. By the time CALL ON ME landed in 2004 the pumping wasthe hook. There is not much else going on in that record. The compressor had stopped being a tool and become an instrument.

Then big-room EDM took it and hammered it until it turned into self-parody. Same story as the loudness war: take something built for control, max it out for impact, and lose the thing it was protecting in the first place.

deadmau5 sidechains to a ghost kick. That is a kick drum which is muted, completely silent, and exists purely to trigger the compressor. The track breathes around a drum that nobody can hear.

Kevin Parker runs entire Tame Impala records through compression as a texture rather than a groove tool. LET IT HAPPEN heaves and swells with no dance floor anywhere in sight.

Finneas ducks whole arrangements a few decibels under Billie Eilish's vocal, so quietly that you would never name it. You would just say her voice feels close.

That is the tell of someone who understands the technique. They are not using it to make a track pump. They are using it to control where your attention goes.

Threshold is the level at which the compressor starts working. Ratio is how hard it squeezes once it does. Attack is how fast it clamps down.

Release is where the entire feeling lives, and it is the one most people get wrong. At 128 BPM there are roughly 469ms between beats. Set the release around 250 to 350ms and the ducked sound swells back up just before the next kick arrives to push it down again. That swell is the exhale. Too short and it sounds jumpy. Too long and the bass never comes back at all.

Gain reduction of 2 to 3dB is invisible glue. 6dB is noticeable. 10dB and above is CALL ON ME territory and everybody can hear it.

How to hear it: the wub is the pump on Muse. Nine Inch Nails is kick ducks. Billie Eilish on OXYTOCIN is the obvious one. If you want a challenge, try YOU SHOULD SEE ME IN A CROWN.

→ Sidechain Breathing

Someone who hears the individual harmonics rather than the note underneath them.

Researchers split people into two camps. Spectral listeners focus on the component frequencies of a sound. Fundamental listeners hear the relationship between all those harmonics and perceive the note beneath them more readily.

Most people sit somewhere on a spectrum between the two, which means the missing fundamental is stronger for some people than others. A natural fundamental listener will hear the bass on a phone speaker with real conviction. Somebody who leans spectral will find it thinner.

No one is better. You are just not hearing the same thing, which is worth remembering the next time somebody insists your speakers sound thin.

→ Involuntary Hallucination

20Hz to 80Hz. Less something you hear than something you feel.

The physical pressure in a room. The pump in your chest at a gig. The reason a club sounds better than your kitchen.

On a small speaker it is almost entirely absent. What you get instead is your brain generating it from the harmonics above, which is the missing fundamental doing its work.

Producers mixing for streaming build for exactly this. Place the harmonics deliberately and the effect fires on a phone. Ignore it, mix only for the studio, and wonder why the track sounds thin everywhere else.

How to hear it: put your hand flat on the speaker. You will feel almost nothing below a certain point while your brain insists the bass is there. Physical vibration and perceived bass are not the same thing, and your hand is more honest than your ears.

→ Involuntary Hallucination

Take things away before a section instead of adding to it.

The ear reads the return as bigger even though nothing actually got louder. It is the neatest trick in production and it costs nothing.

Anyone can leave a gap at the start or the end. The skill is cutting one in the middle and knowing exactly how long to hold it. Longer journey, harder landing, right up to the point where the audience thinks the song has finished and claps.

Sidechain compression is this same idea, automated, running sixty times a minute.

→ Fermata

One sense involuntarily triggering another.

Not imagination. Not metaphor. Automatic, consistent, and usually fixed for life.

Grapheme-color synesthesia gives letters and numbers a color. Chromaesthesia does it with sound. There are dozens of documented variants, including people who taste words and people who feel shapes on their skin when they hear music.

A significant number of musicians report some version of it, which raises a question nobody has fully answered. Does synesthesia pull people toward music, or does a lifetime of deep listening rewire the way the brain handles sound? Both are still on the table.

The threshold is the interesting part. Sound triggers visual cortex activity in synesthetes, but mild activation of the same regions happens in everyone.

→ Chromaesthesia

The range where a melody actually sits, as distinct from the key or the full range of the piece.

A song can be written in a comfortable key and still be exhausting to sing, because tessitura is not about the highest note. It is about where the line spends most of its time.

Sit a melody slightly too high for the voice carrying it and everything tightens. The singer is not failing. The arrangement is asking for something the register cannot give comfortably, and you hear effort where there should be ease.

It is the reason some songs feel draining to listen to when nothing is technically wrong with them.

How to hear it: find a live version pitched down a semitone or two from the record. Same singer, same song, suddenly relaxed.

The initial spike of a sound, before the note itself arrives.

The crack of a snare. The pluck of a string. The click at the front of a kick drum. Your brain reads the transient first and uses it to identify what the instrument even is.

Which is why crushing headroom does so much damage. Squash the peaks and the transients go with them. The snare stops sounding like a snare and starts sounding like a shape where a snare used to be.

It is the most-lost and least-discussed casualty of the loudness war. Nobody writes petitions about transients. They just say the record sounds like plastic.

→ Crushed Headroom

Three whole tones. The most unstable interval in Western music.

An augmented fourth, or a diminished fifth depending on which way you are writing it. Either way it sits exactly halfway across the octave, which is the problem. It does not point anywhere. It wants to resolve and it will not tell you where to.

Medieval theorists called it diabolus in musica and kept it out of sacred composition. Not superstition. It simply refused to settle, and settling was the job.

Which is exactly why it never went away. Jimi Hendrix built PURPLE HAZE on it. Black Sabbath opened their entire career with it. West Side Story puts it on the first two syllables of MARIA. It is in the Simpsons theme.

How to hear it: play any note, then the note six semitones up. Hold both. The thing you want to happen next is resolution, and the composer decides whether you get it.

Another name for the missing fundamental. Also called residue pitch.

Not quite interchangeable, depending on who you ask and which decade they trained in. Broadly, it is the pitch you perceive when the frequency responsible for it is not physically present.

See missing fundamental for the version with the Bell Labs story in it.

→ Involuntary Hallucination

Twenty-two terms. You now have language for things you have been noticing for years without a way to describe them, which means the next time a record does something to you, you can be specific about what.

The missing fundamental is the clearest available proof that hearing is not passive reception. Your ears collect. Your brain decides. What arrives in your head is not the sound that left the speaker; it is a reconstruction built out of pattern, expectation. We just fill up gaps.

Which means every argument you have ever had about whether something sounds good was, technically, two people comparing two different constructions and calling them the same record - fun until you realise no one can really be right… (apart from us).

Every one of these started as a newsletter. We send one every Thursday. It is free, it is written by a person, and it is for people who actually listen.

For those who like deep cuts. Subscribe; it’s free.