The songs I almost didn't release

When people hear the figure of 273 singles, they tend to think I release everything I make. That I write a lyric in the morning and by the afternoon it's on Spotify. It's an understandable idea and it's false. Behind every released song there are others that fell by the wayside. And there's an in-between category, the one that interests me most today: the ones that very nearly stayed out and in the end made it.

Those songs have something in common. For a few days, sometimes weeks, they sat in a folder I call, without much imagination, «ya veremos», "we'll see". It's my label's waiting room. MIKE SAIN'T LOUIS is a one-person record company, so in that room there's nobody but me, watching the clock and dithering.

Too silly

The first reason I almost don't release a song is that it strikes me as too silly. I write with humour, and humour has a very fine edge. Some lyrics make me laugh like mad while I'm writing them and the next day look like an after-dinner quip stretched out to three minutes.

The trouble is I can't always tell the silliness that works from the silliness that doesn't. A title like «La dieta del cucurucho» (the ice-cream-cone diet) can be read as a light joke or as something with more underneath, because behind any diet there's a conversation about the body and about what we've been told we should look like. That double reading is what I'm after. But when I doubt, I really doubt, and the «ya veremos» folder fills up with songs I can't tell are light or hollow.

What I've learnt is that this doubt almost never gets settled by thinking. It gets settled by listening to the song somewhere else, at another time, in another mood. If I laugh again, it goes out. If it embarrasses me, I look at it twice as well, because sometimes embarrassment is a sign I've touched something I didn't mean to.

Too much mine

The second reason is the opposite. Some songs I almost don't release because they feel too personal. Not because they reveal anything scandalous, but because when I reread them I recognise myself in them too much. I see myself naked in a way I don't fancy showing.

With these, the trick doesn't work. I can listen to them anywhere at any hour and the feeling is the same: a touch of vertigo. What helps is thinking about who might hear it. Not the general public, which is a very abstract idea, but one particular person going through the same thing. If I think of that person, the song stops being mine and becomes a little bit theirs. And then it's much easier to let it go.

I write about gay love, exes, pride and self-esteem. You can't write about that without leaving something of yourself on the page. I knew that from the start. What I didn't know is that modesty would turn up so late, with the song already finished, at the very last step.

Too similar

The third reason is more practical and more boring. Sometimes a song sounds a lot like one I've already released. The same subject told from almost the same angle, or a production that's too reminiscent of one from a few months back.

With a catalogue this size, it happens more than I'd like to admit. I produce the music with artificial intelligence tools, and although I direct them, discard and choose, some sounds keep coming back. Just as some obsessions keep coming back in my lyrics. When I see two songs too close together, I usually hold back the second. I don't always bin it. Sometimes it just needs air, time to pass so it sounds different next to everything else.

Here I have a contradiction I can't resolve. On the one hand it seems honest not to repeat myself. On the other, the artists I love most repeat themselves endlessly, and that's part of what I love about them. A style of your own is also a way of insisting on the same thing.

Why they came out in the end

Almost every song in the «ya veremos» folder ended up released. That says something about me, and I'm not sure it's good. Maybe I'm too lenient with my own work. Or maybe doubt is part of the process rather than an alarm bell.

My current theory, which will change, is this: if a song provokes something in me, even discomfort, it deserves a chance. The ones that really shouldn't come out are the ones that provoke nothing. The ones I listen to and think "fine, competent". Those stay out without any drama. The others, the ones that make me doubt, usually have something alive in them.

There's another, more prosaic reason. If I held every song back until I was completely sure, I'd never have released a single one. Total certainty never comes. What comes is a day when the doubt weighs less than the wish for someone to hear it, and on that day I press the button.

These days I let songs spend a while in the waiting room, but with a release date. I don't leave them there indefinitely, because I've found that a song forgotten in a folder doesn't improve. It goes cold. And a cold song is much harder for me to stand behind than one that still burns a little.

I've also learnt to listen to my doubts without always obeying them. A doubt tells me to look at something more carefully. It doesn't tell me to throw it away. Confusing those two things would have cost me half the catalogue.

All the ones that came out, including the ones that almost didn't, are there with their lyrics at mikesaintlouis.com.