It's a given: the process of working with musicians to record their songs is unavoidably an elastic, several-steps-forward-and-back process. Never entirely straightforward, always a little "meandering" in a way. Still, you as the engineer have a sacred duty to keep ontop of things and steer the ship into some kind of port-- that's the heart of your job. And yet... and yet... I learned last night AGAIN how important it is to let the talent "have their way" even though they're pursuing some idiotic notion doomed to fail.
This was a woman playing her piano tune. She did a few takes until she had a nicely paced performance, this was a ballad for Juliet to sing (of Romeo and Juliet) and each take got better until the last one was sublime. She went to add the vocals, and realized it was a squeak too high for her range... sat back down at the keyboard, taught herself the song in a lower key, ran through it until she hit the same sublimity... finally a good take. We've gone through, I dunno, twenty passes by this time?
Adding the vocals, it developed there was a rubato moment going into the second verse... you could call it a pause for effect... or you could call it a mistake. Once she'd finished the lead vocal and harmonies, and we were mixing it, she decided it was "mistake" territory. I advised her that the "repair" would involve snipping out the extra beat--exactly where, and how it would effect things, aye, there is the rub--but she said, "No, let me just play another piano track to cover it up."
So here-- I knew the issue was not a "cover up" issue--the issue was an "extra beat" issue, and no amount of covering up would cure the underlying problem. I expressed this view, but when she insisted on her way, I patched in an extra stereo track and let her try her approach. After fifteen minutes of this, she realized I was right. "You are right!" she exclaimed. And it struck me that this little "waste of time" (which is, technically, not ever wasted at all but money in the bank) was really good all around-- I had shown how flexible I was prepared to be, even knowing full well I was right and she was wrong.
When your customer is wrong... hmmm... that's not supposed to be possible, but this is a business where it happens alot. There's more than a little irony in it. Why isn't the person singing the song the best judge of how the song "should sound," isn't that sort of obvious or something? It might be obvious if the only person ever to hear it was going to be the singer. And, uh... that does happen... but lately I've been thinking of myself as more like a lawyer than anything when I record someone, I am their interface to the world at large, I understand their ambitions and frame of reference, and I am there to make their case. "I advise you to dip out a fairly broad swath, oh maybe 1.5 dB down, from say 400 Hz to 1500 Hz, this will allow the attack of the strumming guitar to get a little more prominent play, I think that will enhance your credibility." Essentially, this is what I'm thinking when I mix something-- what is the effect being created, here?
Sometimes it's a question of the musician losing perspective. I've done tons of furtive little edits/corrections after being directed in late-night sessions to skew some minor element into offensive proportions, all to make some kind of damn point or something. Hiking a fill to ear-blasting levels, lest it be missed-- that sort of thing, things that on sober reflection were stupid.
On the other hand, you've got to find the true "aurora borealis" of the song at hand--and only the musician knows what that is, the animating force that gives it life, that makes it move and stand still. I fell into doing a series of polka albums for a teenage band--three horns, keyboards, walloping drum/bass section, accordion of course--recording everything as crisply clearly as I could... still, I looked to the band leader for cues as to how it all should fit together. What was the emphasis, here? Stabbing, shimmery horns? The pulsing, essentially disco beat? How much leeway with the vocals? The vision thing.
This whole business has a vast gray area where "personal taste" meet "industry standards" and really, in the end, there aren't any "standards." After the fact you can recognize something that turned out especially successfully, but in the thick of things, there's no way to know what will work-- what the "best" thing is, the best mic or pre or warm-up protocol or key to sing in or time of day to get started or this or that mix decision. It's all nebulous. Everything is guess-work. I guess that fits, because the final product is also fairly abstract: the emotional power of someone singing a song, and then the whole question of what destiny this particular recording has in its financier's eyes.
So-- engineers are machinery operators, first and foremost but hopefully not entirely. Our machinery is predictable, reliable, and altogether creates its own little world. You cannot imagine a recording engineer without a pile of equipment and buttons to push and unpush. In your mind it's a scaffolding of how the recording process works and interlocks, how certain decisions dictate others and what you can affect how, where, and where not.
The core of the problem is like so: music is an "emotional" content-based product. The only reason anyone voluntarily (or involuntarily!) brings music into their life is to consume its mood-enhancing properties. It's blaring out of the hidden sound system at my neighborhood Stewart's convenience store to buoy the mood of the employees and the customers, that's the only reason it's there. It doesn't keep the walls clean or anything.
But the machines that capture and condition music are not emotional at all, they are purely scientific, some capable of the most exquisite control of the most peculiar parameters. You need the ability to gauge the entire process, to bring to bear a sense of how machines and moods can be matched-- you really need the ability to see how it's going to look from the outside. I've had, and I'm sure you've had, the experience of someone handing you a "great" CD-R of stuff they were "really proud of" and it sounds awful. The ideas aren't awful, and the songs aren't awful, and the passion in the singing is downright joyous, but the SOUND IS AWFUL. And it's not that these people are fools, exactly, it's that they've let their judgment be clouded by their emotions. They don't really hear the sound... they just hear the goodness, such as it is.
I've duplicated many more than a few dozen CDs, 10 minute programs, for a guy who played guitar into a Radio Shack alarm clock with an onboard cassette deck. Actually, he recorded onto a boombox, and then did a little "open air" mixing onto the Radio Shack gizmo, adding leads. It sounds like a scuffed and scratchy blues record from the twenties, overwhelmed by static and hissiness, the tones warbling due to the wavering playback speeds. He's convinced it captures his genius... I shudder to think of someone cueing it up, because the artwork we came up with gives no clue. Elegant, color, glossy paper, bold but sedated typefaces. He's leaning out of the doorway of Sun Records in Memphis, jaunty smile. To give him the benefit of the doubt... maybe the performances are moving in their own way... still... in this day and age to settle for anything less than 24 bit silence... seems to me perfectly absurd. Clash of cultures, maybe?