Timing Question

I teach myself to play Celtic music mostly from recordings, typically tabbing it out to help me learn it.

I’ve started using proper tab-editing software (TablEdit) to do this, and now that I’m writing it out properly and have to set the correct duration for each note etc, I’ve noticed it’s hard to match the recording exactly.

For example, I am currently working on a recording of Lucy Campbell. It’s in D, 4/4 time, basically all 8th notes (two notes to a beat) plus a few triplets for effect.

But the two-8th-notes-per-beat are not *really* 8th notes because they last different amounts of time. I find that to match the recording as closely as possible the first note of each beat has roughly double the duration of the second… like a triplet with the middle note tied to the first.

I’ve noticed that existing tab just assumes equal duration so it sounds pretty horrible when midi’fied.

So what are the real notes here??

Ah ha. You have a revalation. Congratulations. You now know your comuter can’t play diddley. Welcom back to the real world

As Michael has pointed out, you are now discovering that Irish traditional music is essentially unnotatable because it is from an aural tradition. The same can be said of other folk music and jazz. BTW, you just try doing an exact notation in dots of what Tommy Peoples plays!
If you have software that reproduces on the page the dots _exactly_ as they are played on the track I can guarantee that it will be a hideous mass of black print and absolutely unreadable and unplayable. If you must go down this route I recommend that you set the software to go no further into detail than eighth notes (quavers in Britain). This will ensure that you’ll end up with something that looks a bit like a sheet of music you could buy in a shop. Even so, because the intonation of a live player on the fiddle, flute or whistle is never mathematically precise, the software may well interpret a flattish C# as a C-nat, for example.
Never forget that the dots aren’t the music itself; they only provide a skeletal outline of the tune, much as a road map will show you how to get from one town to another but can never show you the full details of the changing landscape and weather which you can only experience by actually travelling the journey. It is up to the musician, if he reads from the dots, to travel this journey and to interpret the dots in the light of his knowledge, experience and ability. In passing, I’d point out that classical musicians, who almost always play from the dots, have also to apply the same standards of understanding before the music can really be played and heard.
Any recording of music is just a snapshot of a particular performance, whether in public or in the studio (even if it has been cobbled together from several “takes”, or otherwise messed around with). The next time it is performed, even by the same musician in similar circumstances, it will be different, and so an _exact_ rendition into dots by software will be different.
The bottom line is to take the trouble to learn the music by ear - it really is the only way for an aural tradition. Listen, listen, and keep on listening. In my experience it helps to try to “see” the shape of the tune - its ups and downs - and this gives a good basis to fill in the details as I get more into the tune. The dots are only useful at a later stage to identify variant forms of the tune, or some detail that you’ve perhaps missed aurally, or for archiving, or as to serve as a brief reminder - if you already know the tune.
Trevor

Your computer can play the rhythm you’re talking about, but you can’t notate it in 4/4 very easily, you have to do it in 12/8 or 12/16, like this:

X: 1
T: Lucy Campbell’s
M: 12/8
L: 1/8
R: reel
K: Dmaj
A3 F2A A3 d2B|A3 F2A B2E ~E3|A3 F2A d2f e2c|d2B A2F A2D D2B|
A2D FED F2A d2B|ABA FA BE E2B|A2D F2A d2f e2c|d2B A2F A2D D3|| etc…

Can’t be bothered to do the other parts, but you get my drift.

Bar 6 was wrong.

This arises because music is, in fact, a representation of infinity, because any tune can be played an infinite number of ways.

As Duke Ellington once said “it don’t mean a thing if it ain’t got that swing,” (or in the case of this music, perhaps I should say “how it lifts”). Fortunately, as you get more used to the music, you will understand better how to translate the notes into a tune that sounds the way it should, and will be able to pick a tune up from the written page more easily. But I find it best if you learn the tunes primarily by listening, with the notes serving as an aid to memory during practice. Interestingly enough, I have a tunebook somewhere in my collection (can’t think of the name offhand) that attempts to notate the tunes verbatim, with swung notes broken down, cut and roll notes printed out, etc--and I never learned a tune from it--the blasted music was too busy and complex to wade through!

When I was heavily into teaching guitar, (cough, ahem) years ago, the prevailing tablature style was almost like standard notation, with stems and flags on the notes to handle the rhythm. I thought this was silly, since to make sense of it you had to go more than halfway into learning standard notation anyway, so I wrote my tabs with fret numbers placed where they fell in real-time. The bass notes were usually a steady 4/4 or 3/4, so it was pretty clear from those reference points where the in-between notes should go, but for emphasis, I also drew a mark under the staff at the location of the downbeats.

I had moderately good success with this in teaching tunes with straight (eighth note) rhytthm vs swing rhythm vs very Scottish pipey rhythm. Beyond that, it would have been hard to accurately convey fine distinctions, but I think maybe something like this is the best hope for accurately notating eye tee, uh, tee eye, uh, I mean that kind of music that we love so much but hate to name. It would be technically feasible for a computer to listen to a fiddle tune and spit out a real-time type tablature. Of course, most of us would still have to listen to the tune to get it right, but it would at least present the possibility of notating the rhythm exactly, for reference, and of having something like a MIDI interface to reproduce it accurately.

Would it be worth the effort to develop it? I’m thinking, probably not. Besides, it wouldn’t deal with pitch at all.

Yes.

I just posted some comments in the tune section for the Roaring Barmaid, which might help clear up some of the confusion about the B part.

As far as the original question, the rhythm-learning problem, if you tell me what recording of Lucy Campbell you are trying to notate, or can send me a sound clip, I can notate the rhythm for you in a finale 2003 file and send it to you as both sheet music and midi. I’m compiling a library of tunes with the rhythm transcribed exactly off recordings as part of a project on performance practice, so I could definitely do this one and then share it. That said, I don’t know how helpful the notation will be--the rhythm, as you noticed, is based on pairs of long and short notes, but the ratio of long to short can be very complicated--rather than 2:1, or 3:1, it’s more likely to be 20:13, which is able to be notated, but it looks pretty hairy. If you’re a visual learner with a somewhat mathematical/logical bent, it might be a help to you to see it laid out, but your ears are going to be your most important tool for learning the rhythm of Irish music. Most people probably use transcription tools just to be sure of all the pitches, and just ignore the rhythmic aspects of the notation. It takes a lot of listening, playing with others, and practice. Working with a teacher that is comfortable breaking this sort of thing down in a detailed way is very valuable, if you have access to one.

If you (or anyone) wants to see some of the things I’ve notated or hear about my project, I’d be glad to tell you about it or send you some finale files and midi files. It’s a work in progress and it’s not meant to be a learning tool, but if it’s of interest to anyone, drop me a line.

By the way, here’s the link for the Roaring Barmaid--forgot to add this before:

https://thesession.org/tunes/91

As Trev, Al and Bob point out, it’s pointless.

Actually, I think it’s worse than pointless. It’s missleading. I apreciate that you understand that “your ears are going to be the most important tool”, but I disagree that an “acurate” notation of the “swing” can be helpful. The main reason being that the swing is not consistent. It’s not consistent within one time through the tune, let alone from one time to the next it’s played. It depends on all kinds of factors, not least who you are playing with.

Try breaking the amount of swing down in the first two or four bars of Liz Carrol’s recording of Lost in the Loop as an example. An interesting exercise it may be, but it won’t tell you anything about diddley music at all.

“. . .the prevailing tablature style was almost like standard notation, with stems and flags on the notes to handle the rhythm. I thought this was silly.”

Stefan Grossman was also perspicacious enough to realize how silly this was for “folk” music, and always writes his tab with only the crudest indicators of timing and rhythm, produces recordings of the music notated with tab, and instructs people to play it like it sounds.

Had I been left with no learning reference but the printed works of Jerry Silverman and Happy Traum (who take the full notation approach to tab) God only knows what a mess I would have turned out (this is to mean no disrespect to Happy and Artie, who are both nice people and good teachers. I’ve never had the pleasure of meeting Jerry).

The idea that you can play even classical music properly from the written score is a myth. The only people who give the illusion of doing that are people who have listened to a good deal of classical music and already know what it’s supposed to sound like. Since most of these people have been listening to clasical music since before they can remember they suffer under the self delusion that being able to produce a good rendition of classical music from a score is “natural.”

This phenomenon isn’t restricted to to the classical genre. You’ll find much the same thing in Michael’s insistence that Irish music is so simple you don’t have to practice, it’s all in the ear. Anything you *already* know is simple and easy and it’s easy to loose sight of the fact that other’s don’t already know what you do.

I’m reticent to tackle Ostrich Feather’s post directly, she and I have already gone a few rounds on her work. 🙂

I believe what she is doing is of tremendous academic value and support her in it In fact I’ll want signed copies of her senior project/Masters/Doctoral thesis; and I’ll treasure them.

We know from history that things get lost. There are two John Hurt songs, for instance, that we know from the printed record were recorded, but no extant copies of the recordings have ever surfaced and John himself could not remember them when he was “discovered” in the early 60s. Those songs have been erased from history. Had someone notated them on paper we might still be able to reconstruct them. Certainly John could have remembered them if there had been at least tab available. Even recordings are not a sufficient record. We need the music availble in as many formats as possible in order to preserve it for future generations.

Ostrich Feathers herself, as a harper, is surely in a position to understand this better than most, as almost the entire corpus of Irish harp music was wiped out, unrecoverably, entirely because of the relience on the aural tradtion. For all practical purposes only those tunes that someone wrote out in notation survive (a handful got translated into the fiddle genre, but only a handful, since fiddle was used primarily to play a different style of music)

All that said, music is *sound*, not printed notation. The primary record is always the aural (whether live or recorded) and you play it like it sounds, not like how it’s written. I just don’t see any value in “over notating” to the current musician learning to play, except as a tool to be used in the rehablilition of the musical cripple who cannot play what they hear.

And even then only as a tool to help them learn to hear what they are hearing, so that they may learn to hear and play that.

Having good MIDI files that could be freely traded over the Internet could be a tremendous boon to some people, but there is great danger that the majority, unfamiliar with how it is *really* supposed to sound, would learn to play it like that, and most of the people who would benefit from such files already know how to translate the existing record into the way it’s supposed to sound, because. . .they already know what it’s supposed to sound like.

KFG

Some interesting points there. You say,“We need the music availble in as many formats as possible in order to preserve it for future generations.” I’m glad you concede that, “The primary record is always the aural”, but its the preservetion that bothers me.

I would rather say, “We need the primary record of the music to be always the aural, so that it may be grown by future generations, not preserved.”

Yes, historical renditions can be intersesting, but remember what Alfred Brendel said when asked why he played Mozart on a 20th century piano … “It sounds better.”

Yes, everything moves on. We even have tuneable bodhrans, these days. Do these sound any better? 🙂

Thanks to everyone for their contributions! Some great points and useful information. I am dying of flu at the minute so can’t follow this up properly, but sure hope to when I’ m better.

“. . .remember what Alfred Brendel said when asked why he played Mozart . . .”

Because he *can.*

KFG

Ah, but what if the notation takes into account the fact that it’s not consistent? 🙂

I’ve developed another method since I started doing this, and now have a way to get basically a snapshot of one person’s performance on a given day, where the relative length of each individual note is represented. This can of course be converted into a midi file as well. I never realized before that people like to convert things into midi for learning purposes--is it just easier than dealing with all the extra information you get in learning from a regular recorded performance? But in light of what Kevin (thanks for your support, by the way!) mentioned about midi files being helpful, maybe the “snapshot” method could produce clips that people could learn from with less danger of picking up bad habits.

I have a long way to go before I’m prepared to make an absolute decision on how useful or useless the notation could ever be. Since it’s part of my research, I have to keep an open mind for now. But it’s not how I teach my students. My guess is that it is much more efficient, not to mention fun, to learn firsthand from someone who knows how to do it, and anyone who has the opportunity to do that is extremely fortunate. Reuben was already working on this sort of thing on his own, and I thought that rather than having him reinvent the wheel, which he was working on anyway, I might as well offer to share my work, which he can do what he likes with, since he did specifically ask, “What are the real notes here?”

I wholeheartedly agree, by the way, that it is a living tradition and should be grown rather than preserved. I don’t see what I am doing as preservationist, though it could certainly serve that purpose should the need ever arise. The reason we need the primary record to always be aural is that if it was to occur that that aural record should disappear, the tradition would be dead, a historical footnote as lost and irrelevant as the old harp tradition. We need the aural record because we need this music not to die. But I don’t fear that Irish music is in any danger of being wiped out anytime soon. I do sometimes fear for the future of the harp in Irish music, because what is being so vibrantly preserved and grown in the aural traditions of other instruments is far less firmly entrenched in the young and peripheral modern harp tradition. That may be the most valuable thing I can hope to accomplish with my project, to ultimately initiate a dialogue about performance practice that will address, among other things, the divergent path that the harp tradition seems to be taking.

It’s not a case of fearing that the music would be wiped out. It’s a fear that it “grows” into something that is not as good as it was. And I feel that the ever increasing reliance on the many forms of recording music other than using only your memory is changing the music.

It’s happened befor, with other musics. The skill of Bach in improvising was lost. Scottish music is more regimented than Irish music because of it’s tradition of being notated.

Improvisation is still with us; in jazz (obviously), and in church organ playing, where it is a skill that is still fostered and expected of the organist. You could say that Bach’s improvising skills have been passed on and are still alive today.

Trevor

Thinking on from my previous post, perhaps the better session players should be encouraged to improvise a tune occasionally – a jig or reel perhaps. For all we know perhaps this already happens, but the player is too embarassed or modest to admit to it and mumbles something about having “heard it at a session in Clare last summer”, and no, he didn’t get the name of the tune either.

Trevor

Ha ha trev. I do it all the time. You go from one tune to another, but the next tune gets a bit lost and you end up flim flamming arround for 32 bars, then roll your eyes with a tut tut, “doesn’t anyone else know this one?” then you manage to retrieve it by going into a “propper” tune.