Why your pieces got better and your reading didn't

By Ashley Song
Reading time: 8 minutes

My teacher used to put a new score on the stand near the end of the lesson and say, just sight-read it.

I dreaded that more than anything else in the hour.

Not because I couldn't play. I had pieces — good ones. Pieces I'd worked on for months and could play properly, with pedal and phrasing and all of it. If you'd walked past the room you would have thought I knew exactly what I was doing.

But a page I had never seen before? I'd find the first chord, lose it, go back, and work through it one note at a time like I was fifteen again.

And the whole time: he's watching me do this.

I had been playing for twenty years.

For most of those twenty years I was certain I knew what the problem was. I hadn't practised enough. I wasn't consistent enough. Other people were putting in the hours and I wasn't, and that was the whole story.

It took me far too long to work out that it wasn't the story at all.

The moment I can play a passage without looking, I stop looking. I tell myself it's because memorising is good musicianship.

It isn't that. It's that the page stops being useful to me the second my hands know the way.

I'm not reading while I play. I'm reciting, with a piece of paper in front of me for reassurance.

If I stop in the middle, I go back to bar one. There's no other way in.

I thought this was normal. It is common, which isn't the same thing.

A page with no units on it has no entrances. I'm not playing a piece that has parts — I'm playing one long chain, and a chain can only be picked up at the front.

This is the one I've never said out loud to anyone.

When somebody suggests something simpler to read through, I find a reason not to.

Because an easy piece would take away my excuse. If I'm slow on something difficult, that's the piece's fault. If I'm slow on something easy, it's mine.

So I keep choosing music that's hard enough to explain why the first read went badly.

This is the one I explained away the longest, and it's the one that should have stopped me.

Twenty years of playing. And I still approach an unfamiliar page the way I did as a teenager: first note, second note, third note.

Everything else I do got faster with experience. This didn't. And twenty years has to be accumulating somewhere — if it isn't showing up on a new page, it isn't accumulating at all.

That thought is genuinely unpleasant. I avoided it for a long time.

Then came the afternoon that actually changed something.

I was stuck on a passage late in a piece. I finally worked out where in the score I even was, pieced it together — and then realised I had already learned it. Twice. Earlier in the same piece.

I'd been staring at that page for an hour. And once I'd finished learning it from scratch, again, I looked up and saw it: the same bars, the ones I had played effortlessly twenty minutes earlier, in the same piece, in the same ink.

I had paid for it three times.

That was the moment I stopped believing this was a practice problem.

I want to be fair to all of it, because none of it was stupid and some of it helped.

1. "Just sight-read more."

   The standard advice, and it isn't wrong exactly. But reading more pages the way I was already reading them made me faster at doing the thing that wasn't working. Repetition doesn't change how you're paying attention. It just scales it.

2. Note-name drills.

   They made me quicker at naming symbols. Naming symbols was never the bottleneck.

3. Slow, careful run-throughs.

   Genuinely good practice, and my playing got cleaner. It improved the accuracy of my decoding. It didn't change what I was decoding into.

4. A metronome.

   Fixed my timing. Left my reading precisely where it was.

5. Another method book.

   New pieces, same instruction underneath: here are the notes, work them out. I finished it and I was the same reader on the last page as on the first.

None of that was wasted. The pieces were real, the hours were real, and I can play things now that I couldn't play then.

I just never had anyone tell me what to look at.

Here's the part that reorganised everything for me once I saw it.

There are two different things happening when you learn a piece. One is getting that piece into your hands. The other is getting better at meeting music you've never met.

I had spent twenty years getting extremely good at the first one. Months on a single piece, polishing, memorising, refining. And every one of those months was an hour not spent reading something new.

My repertoire advanced. My reading stayed where it was.

Which means the two things people use to judge me — what I can play, and what I can read — had drifted so far apart that anyone putting a fresh page in front of me was, without knowing it, testing a completely different skill from the one I had been building.

That's why the lesson felt like an ambush. It wasn't measuring the thing I'd worked on.

When I finally described all of this to David, he didn't ask how much I practised. He asked what I saw when I looked at the page.

Then he explained it like this.

Notes are letters.

I climbed that rung as a child. I understood the logic. I was supposed to move on.

A fluent reader doesn't read the letters one by one. He takes in the whole word, recognises what it means, and moves to the next one, until he has a sentence.

For twenty years, that is not what I was doing. I was reading letter by letter, out loud, without ever thinking about what the words meant. Like a child who has just learned to read and can't tell you what the page said — because he was too busy with the letters.

And here's what reading by shape actually feels like, in one line:

Tihs snetcene is raedalbe.

You read that. You couldn't have read it letter by letter — the letters are in the wrong order. You recognised the shapes.

On a page of music, I was still doing it letter by letter.

That distinction mattered more to me than anything else I've written here.

For twenty years I had read my slow first attempts as information about what kind of musician I was. They weren't. They were information about one specific processing habit that nobody had ever asked me to change.

And the thing about reading it as a verdict is what it makes you do next. Go back and look at Sign 3 — the easy pieces I wouldn't touch. You don't walk towards the thing that proves you're not good enough. You arrange your practice so you never have to find out.

I'm not the only one who did that.

Read the second sentence again. That's not someone saying they feel better. That's someone walking towards the thing they used to avoid.

Not because my teachers were bad. Every one of them was competent and every one of them taught me things I still use.

But there's a skill nobody ever explained to me, and it's a strange one to put into words: deciding what counts as one thing.

Whether these eight notes are eight separate events, or one shape that happens to be written out. Whether this bar is new material or the same idea as bar four, moved.

Nobody ever taught me that, and I don't think it was withheld. I think it sits in a gap. Notation teaching gets you to the point where you can look at a symbol and produce the right sound at the right time — and then it moves you on to repertoire, because that's what lessons are for.

From the outside, someone who decodes accurately and someone who reads fluently look identical. They only come apart when a new page appears. By then everybody calls it talent.

The relationships that come back in a piece aren't hidden and they aren't a matter of taste. They're printed. Finding them is a procedure, and it has six steps: inspect, group, compare, predict, play, transfer.

He didn't give me exercises. He gave me a workbook — real scores, Bach and Chopin, with the recurring material marked, and those six passes to walk you through finding it yourself.

A first reading book. For someone who can already read.