From the Music StackExchange:

Below I document my thought process, which is obviously wrong, else one of the billions of people would have commercialised it.

Assumptions:

  • A good piano player (let’s substitute piano for instrumental hopefully without loss of generality) can get a note-book as input and produce a mostly accurate audio rendering of it (output).
  • The major product that a piano player manufactures is an audio rendering of the note-book. Show and autographs come second.

Observations:

  • For the late 20 years, there have existed multiple computer programs that perform this. They even allow the user to chose the type of piano.

Limitation of scope:

  • This question excludes live performances as there the artist dancing is a major factor. It considers only studio recordings.
  • It also excludes composers and other higher order artists. I believe playing an instrument is the way to learn to compose for it.

Questions:

  • Why are instrumental performances still a thing? Why spend 10 years to be good at piano, when a computer can spend 0 and be perfect?
    • Antithesis1: The beauty of music is in the small mistakes an artist makes. My response: noise can and has been statistically modeled since forever.
    • Antithesis2: The years of effort the artist has put into learning the instrument makes the music all that more meaningful. My response: IF we only care about the output audio file or acoustic vibrations, performance is not important.

This question has plagued me for years but not being an artist myself there’s zero chance to answer.

In terms of the Music SE, I don’t like this question. It’s all opinion. Opinions can change, and two people can have diametrically-opposed opinons. It doesn’t help.

But…

  • “Perfect” Isn’t Meaningful Here

Here’s an extreme case: In 1955, pianist Glenn Gould recorded the Goldberg Variations. He did so again, in 1981.

If I was to take the work as written, run it through Sebelius or something else to get it to MIDI, then find a piano that accepts MIDI (they do exist; I’ve seen them on YouTube but can’t find it right now) and get an audio recording. It would differ from both the 1955 and the 1981 Gould recordings. It would differ from all the thousands of recordings from different pianists, all of which are “accurate audio renderings” of the text.

Variations in tempo, in dynamics, and a series of other factors can determine which move you more. Some are outside of music: Glenn Gould is a name I’ve heard, a name given reverence by other musicians, so I’m primed to prefer his versions.

  • Live Music

This question excludes live performances as there the artist dancing is a major factor. It considers only studio recordings.

I would suggest there are two kinds of live music, and the kind of music he suggests would work incredibly well with the second.

  • Live Music: Jazz and jam-band music are the key here. The band members listen to each other and react to each other, as well as to the audience. The music can take left turns and each show varies from the last, because the players come in with different feelings and the audience is different.

  • Live Shows: Here, the point is the show, the spectacle on the stage. From Opera to Pop Divas, there’s a large degree of regimentation because if a player feels a Gmin7 instead of a G5 power chord, the show won’t build the way it built last night and the night before. There are click tracks which connect the lighting and the effects and sometimes the vocal tracks, because let’s face it, singing on-key in the middle of a long cardio workout with props and choreography can be hard. You need to belt that high note now; you need to breathe in to provide oxygen to your body now, and your body won’t wait. In this environment, having perfect generated accompaniment, as opposed to a player expected to play the same bit every night makes a lot of sense.

So, in one case he’s right for the wrong reason, and in the other, he’s wrong because he’s right.

  • We Kinda Do It Already

The drum machine has existed as a commercial product since 1959, and started showing up in recordings in the 1970s. I think the first album was Sly Stone’s There’s A Riot Goin’ On. In the 1980s, it became a more common thing.

Drummers first responsibility is to keep time, but without a metronome or click track, musicians and bands tend to speed up through a performance. Thus the aforementioned click track, which provides a consistent tempo. If, instead, we program a drum machine to give a specific beat at a specific tempo. Now there are genres where that consistent tempo is a bedrock requirement.

But speeding up is musically valid. Playing louder or softer, playing simpler or more complex; this can all communicate meaning. There are musicians whose work leaned toward drum machines in the 80s and 90s, and that change caused me to check out from them for a decade or more.

  • What Is Perfect, Anyway?

It is a mistake to believe that a songwriter knows the form a song needs in the beginning. We can go to Leonard Cohen’s “Hallelujah” as an extreme case, where a cover of a cover is the canonical version, but I recently heard Mike Campbell’s story about Tom Petty’s “You Wreck Me”, where the line was “You rock me”, but Mike and others saying “that’s generic; you can do better”, and that Petty didn’t want it on the set list because he wasn’t happy with it, until a few shows into the tour when audience reaction proved it was a winner.

The drums to Cream’s “White Room” are a key to the song’s popularity, and Ginger Baker felt it was significant enough that he should’ve gained a songwriter credit. I’m not a musical copyright lawyer, but his case is compelling to me.

In all these cases, the creation of the song wasn’t complete until it was performed and interpreted by other musicians with the freedom to determine what the best version of the piece is, and their performance, not a compiled and generated original, defines what the song should be.

  • Let There Be Songs That Fill The Air

If my words did glow with the gold of sunshine
And my tunes were played on the harp unstrung
Would you still hear my voice come through the music?
Would you hold it near as it were your own?

It’s a hand-me-down, the thoughts are broken
Perhaps they’re better left unsung
I don’t know, don’t really care
Let there be songs to fill the air

This is the beginning of “Ripple” by the Grateful Dead, off their American Beauty album. To me, this gets to the heart of this question, and to the point of music. There could be a perfect version of the song, with glowing words and perfect music, but it’s good only if the songwriter’s soul is felt, and the audience embraces it. It comes through the songwriter and isn’t perfect. Because it isn’t perfect, maybe it shouldn’t be played, but it is good that songs are played, being passed along from person to person.

I play to jam tracks and the like, but I like it best when I play with other people, because, like I said above, I feel music is a conversation. I am more attuned to the communication between musicians, less so musicians with audience, but that’s my deal. Whether we’re talking the folk process or the chain of influence, it’s learning from other people’s music, adapting it and passing it on.

  • Coda

In short, yes we do it in some cases. It works in those cases, but it isn’t the common case, because that’s not what we believe music is and music is for.

If you have any questions or comments, I would be glad to hear it. Ask me on Twitter or make an issue on my blog repo.