Friday, January 25, 2008

11. Looking/Memorizing Ahead

The concept of looking ahead as you play is an important one, and I'm working hard on developing this skill. The idea is this: most sight-readers are scanning and memorizing upcoming notes while their hands are on autopilot, playing the current measure with no conscious thought necessary.

In today's post I talk about some things I've learned on this topic.

There are a few people who claim that they do not look ahead at all, but most good sight-readers feel that it's an important component of their skill.

I used to think that sight-readers looked several measures ahead, but it seems that most only scan ahead about one measure at most.

Here's an exercise that my piano teacher did with me years ago: She let me study a measure as long as I wanted, but when I gave a nod to indicate that I'd memorized it, she'd cover it up. I wasn't allowed to play it until it was covered. This would continue with each measure of the song. The idea was that I'd play one measure while memorizing the next -- just as I should be doing when sight-reading.

I hated that exercise! Yesterday I tried doing it myself, forcing myself not to look at a measure while playing it, and I still hate it. I think I know why I don't like it. Take a look at this line from a song I used for this exercise:

This is a very simple piece, but there's a lot going on in some of those measures! Take the first measure, for example, how are you going to memorize that quickly? It's much easier just to play it. Can you do this in two seconds (the time allotted to one measure at 120 BPM)?

"OK, let's see. First beat, there's a B in the bass with a minor third starting on D in the right hand, then both hands play a G, followed by a D in the bass, with a B and an F in the right hand, BTW all quarter notes so far, and now the last beat has another G in the bass and treble, but the right hand G is a dotted eighth followed by a sixteenth."

Or perhaps, in a more sophisticated way:

"OK, the bass arpeggiates a half-diminished B minor with 1-5-3-5 quarter notes, while the right hand first does the 3-5 of the chord, up an octave, plus the sixth, then a 1-5, followed by a pair of G eighth notes that are swung."

Even those long descriptions don't adequately describe the measure.

Presumably I would be memorizing that measure in a more non-verbal way, perhaps even memorizing the way my hands would feel as it's played. Maybe I'll be able to do that in the future, but for now, this is just too much stuff for me to memorize; it's more information than humans can normally put in their short term memory.

So yesterday I had this revelation:

I don't have to look/memorize ahead a whole measure at a time.

The distance ahead that I scan can depend on the difficulty or simplicity of the measures. When there's a lot of information in the notes, I might read only the next beat as I play the current one. When I come to a measure that's simple, I can take the time to look further ahead.

Now that I'm no longer rigidly trying to bite off full measures at a time, I've had more success looking ahead. Yes there are still times when the current notes are difficult enough for me that all looking ahead gets canceled until further notice, but for very simple music I can experience the concept of looking/memorizing ahead first hand, and it feels good.


Anonymous said...

Hi, great blog!
As an aside to this section it follows then that if you memorise the difficult sections that deplete your read-ahead buffer you should be able to play the piece "at sight" faster. And thus your "sight reading" repertoire can be greatly expanded and played at higher speed by memorising these difficult sections.

Richard-T said...

This is a truly excellent self-analysis of how you're generally going about your task - thoroughly astute and highly informative. And the conclusions you've reached about a)not needing to read ahead/ memorize more than a small number of notes, and b) needing to find a less verbal way of mentally decoding and re-encoding the notes you're scanning in order to memorize them effortlessly are 100 per cent "nail-on-the-head" - as a sight-reading coach I couldn't overstate their fundamental importance. Surprisingly few students come to pinpoint sight-reading's dependence upon short-term memorization and consequently take no steps to research the subject. So it's great to see your link to Miller's classic paper. Larry Squire's more up-to-date work in the field is also well worth checking out.

On the basis of this post's content, may I offer a couple of pertinent tips? The first one is I think the single most important clue to achieving successful results in all aspects of piano-playing ever stated by a top-professional: it's Artur Schnabel's "One should never make any music, not even sound one musical note, without a musical intention preceding it." That sets the agenda for whatever you're practising for - fluency in playing from memory, perfecting the accuracy of your key-accessing or key-striking movements, or increasing the effectiveness of your sight-reading. The logic of Schnabel's words is easily spelled out: Sounding notes requires making playing-movements; all successful playing-movements are goal-orientated (ie, configured to achieve a particular intention); without a preceding intention, the brain cannot assuredly access or process the information it needs for configuring even a single playing-movement, let alone successions of them.

I think Schnabel's advice provides a really solid foundation for thinking about the fundamental problems of sight-reading, for figuring out and exploring the mental operations required for it, and structuring these operations into a seamless, generally effective sight-reading mindset. For example, if musical intentions are essential for configuring your movements, it follows that your primary concern in sight reading must be to define the musical intention of each movement as clearly as possible in advance of its execution, and - because of the very short time-span available for that - define it in terms that are directly transferrable to, and directly translatable by, the motor areas of the brain that configure your movements.

The latter of these inferences seems to me to link up pretty directly with your two aforementioned conclusions. My own tips below concern developing expertise in managing the short amount of time you have to internalize (i.e input and mentally register) the notation you're about to scan. Achieving fluency (i.e.,continuity of execution) presupposes having in mind the intention to make a minimum of two successive movements prior to executing the first and even - this is crucial - prior to scanning them. The intention to make a certain number of movements provides a set structure for inputting and registering in short-term memory the musical information relating to that number of movements upon actually scanning the notation ahead. As soon as you have that information in short-term memory, think no more about the movements per se - simply allow your motor system to execute them in accordance with your formed musical intentions for each, focusing especially on ensuring that the second of your musical intentions is felt as a seamless follow-through to the first (assuming you're scanning just two notes at a time).