Δαφνη Ευτυχια Αρθουρος

"Why is it that so few dog and pony shows actually include either dogs or ponies? I'd like them so much better if they did." -- sjo, 2005-03-23

A tangential answer to questions I asked in an early-music newsgroup about notation for viola da gamba[1] mentioned period musicians being expected to be able to read in "all nine clefs". This startled me, because I only knew five clefs (six if you count percussion) and really think of those as three (four if you ... yadda yadda), one of which moves. I wondered what the four I'd missed were. (As I commented elsewhere, "Ah, the entertainment value of the lack of a formal music education.")

The answer turns out to be that just as C-clef on this line is "tenor clef" and on that line is "alto clef", etc., if the G-clef is on the bottom line it's called "French violin clef", and there's a name for F-clef on the top line and a couple more names for different placements of a C-clef. This implies that the answer to a question I'd wondered about in the past, "Can I move the G- and F-clefs in current modern notation, or is that only kosher for 16th Century scores[2]?", seems to be, "Yes, they move a little (to specific places) today, technically, but I'll still confuse most people if I do despite its being officially correct."

(Why would I want to move the clef? For the same reason my ancestors did and C-clef instruments still do: I hate ledger lines. Of course, being a guitarist, I'm going to get ledger lines someplace no matter where I put whatever clef and what '8va' mark I throw in, but boy was I wishing I could move the clef on those alto recorder parts last night that kept going higher than the soprano parts!)

Creeping gradually to the point of this entry ... I know there's a big overlap between mathemagticians and musicians, but I feel extra math-weenie-ish -- or is it more of a programmer thing? -- for still not thinking of them as nine distinct clefs, but as three clefs / clef-symbols, which can be placed on any line. It seems more ... well, more mathematical that way. More logical, more pattern-based, less abitrary-memorization-based.

That is, I've got three symbols and a simple rule for interpreting them: the shape tells me what note the clef identifies the location of, and the position specifies that location. Three symbols and a rule -- not much to remember! And it's not like moving the clefs gets in the way of sight-reading from Arbeau, right? (Uh, right? Or is it just that only the folks who "got used to it pretty quickly" have bothered to mention it much?) This is exactly what happens with the "remember the different names" approach, just without bothering to remember the names. It's still what the clefs mean whichever way you learn them.

My approach here does not strike me as especially clever, it's just that I find myself wondering why, if so many musicians are mathematically inclined, they haven't taken the lazy-mathematician approach.

Of course the fly in this ointment is that I still have to keep track of "these singers will get confused unless I put the clef here, those singers will get confused unless I put the clef there, this instrument wants it on this line," etc. (Note that it still saves me the step -- except in vocal parts, where it's obvious -- of remembering which instrument takes what clef-by-name.) But the solution to that (*ahem*) is simply to go back to expecting folks to be able to deal with the clefs moving (this is more common in Europe than the US?) ... and for the composer/arranger to choose sensibly to minimize ledger lines. :-)

[1] The direct answers were that bass gamba is written at pitch in either F-clef or C-clef (but gambists can be expected to read G-clef and play it an octave down, as well); and that double-bass has been written an octave higher than it's played for as long as folks have been writing for it (I wasn't sure if that was a later adaptation or went all the way back).

[2] I don't know whether music scholars count Arbeau's notation in Orchesographie as "modern" or "almost modern" notation, but I think of it as modern notation because the symbols and their meanings are congruent to the symbols currently in use (change diamond note heads to round and cut the time values in half and you're pretty much there), it's a five-line staff, and he uses bar lines -- in other words, I didn't learn a new notation system to read it; I just got used to "looks funny", just as Shakespeare and the King James Bible are Modern English. (Though if I understand correctly, bar lines weren't in common usage -- Arbeau threw them in to make it easier to line up the dance instructions to the music ... have I got that right?)

Driving home from rehearsal thinking about all this, it struck me a) how dense the information content of musical notation is, b) how most of it is very simple with most of the complexity being from the juxtaposition of so many simple encodings of different things onto the same space, and c) how redundancy helps legibility (e.g. optionally using horizontal space to indicate time to make the note-shape encoding of time easier to read).

I think of "reading music" in the sense of "deciphering the information represented on the page" as being easy, though I'll concede that sight-reading quickly and accurately is a harder skill. This fits with my perception of it as several very simple systems overlaid; no one step is (usually) tricky or complex to interpret. It also fits with the just-mentioned difference between "I can make sense of this" and "I can interpret and execute this on the fly": the layering of so many simple encodings means that simultaneously decoding all of them -- while playing and ideally also interpreting and nuancing -- adds up to a more difficult task.

(FWIW, my sight-reading is nowhere near as good as I'd like it to be. Maybe if I regularly got around to practicing sight-reading, I could fix that. Not that I ever expect to catch up to some of my friends in that department.)

There's a phenomenon I haven't gotten to make sense to me yet despite my observations so far: musicians who can read music but not write it. (I think I've babbled about this here before.) I've had several people who read music tell me they can't write it, and some have been surprised and impressed that I can do so. This confuses me: since I can read music, it makes sense to me that anything I can play, I can write down (with the caveat that unusual rhythms can take me a (possibly long) while to figure out how to notate). The idea of not being able to reverse the reading process confuses me, since all the information about the encoding system already has to be present, though I'm sure that I'll eventually see what the obstacle is. I won't be surpried if it turns out to be entirely a matter of self-confidence and a willingness to choose from multiple correct options ... but neither will I be surprised if it turns out to be some cognitive leap I didn't notice having taken. Until I figure it out, I'm perplexed by the phenomenon.

The folks I've mentioned this to who can write music haven't seemed to find the writing ability special, so it's probably not just me who doesn't get this.

And then there's the funny-feeling-in-the-head that comes from taking a tune you know on one instrument, say a guitar, but not really "by ear" so much as "in the fingers", and trying to play it on another instrument (say, a recorder) without sheet music, by remembering how the guitar finger motions feel and translating those rememered sensations into recorder fingerings on the fly. When it doesn't work, it becomes obvious how messed-up trying to do that is. When it does work, it makes one's hands and the inside of one's skull feel really, really strange.