Dunno if this has been asked before (can’t imagine that it hasn’t been, but I don’t recall seeing it or the answers), but: is there a particular reason or theory behind this? Is it a practical measure to play best in tune in all keys (shades of equal temperament, gasp), or, in the case of modern makers who still do this, is it basically just inertia (it was good enuf for ol’ Rudall, so who are we to question), or is the truth something else?
I’ve always assumed that a big reason for this was to keep the third octave of the flute in tune. These old flutes were built to play over a wider range than we modern ITM players require of them. Modern makers, whose flutes are used largely although not exclusively for ITM, are able to make a two-octave compromise to get the flute as in tune with itself as possible, whereas the old makers had to extend that compromise over three or more octaves.
Another factor that may play into it is that many of the old flutes were built to play in pitches other than the current concert pitch of A440. I know my Rudall was, and I mess up the internal tuning of it a bit when I pull it out to play at 440. But you gotta do what you gotta do. I don’t get yelled at when I have one or two notes out of tune with the fiddle, but when it’s all of the notes I sure do!
I’ve played the Siciliano from the E-major Sonata, which itself is actually in C-sharp minor for the most part.
It sounds pretty decent on the Hamilton.
On my old German 8-key, which has its flat D and F-sharp and C-sharp and all the rest, this Sonata just sounds wonderful, much better than on the Hamilton. The intervals just sound truer in this key than on the Hamilton, even though by the tuner the Hamilton is much closer to “correct.”
So yes, from experimentation, I would say one of the reasons the old flutes were tuned like that was to make playing further afield in different key signatures sound as good as possible.
Chris Norman once said to me, “Don’t blame the flute for your inability to play it in tune.” I’ve gotten pretty good at playing different flutes that demand different adjustments to play in tune.
Another thing to consider is that many of these flutes aren’t designed to be played in equal temperament. To my ears a lot of these “bad” notes actually sound in tune, not that I can’t play them in the same intonation as a record I’m playing along with.
That’s not to say that there are flutes out there that need some keys to be vented to play more easily in tune, or that some makers make keyless flutes that might be very closely based on those that needed keys vented to play in better tune. But the adjustments necessary to correct those tuning “features” aren’t huge.
FWIW, it seems to me the keyless conical flute (and the boehm flute as well) is, to a certain degree, flawed in terms of being able to play spot-on in tune. I think all flute players play out of tune at times as I believe I’ve heard the greats from ITM and classical do so.
But there are times, when the flute is warmed up, I’m warmed up, energized, focused, loose, the stars align, the mojo is working, and the music coming out is very very in tune. I suppose it has to do with minute instantaneous adjustments of embouchure, air control, finger coordination, mental focus, muscle memory–in other words, flute playing and good intonation are a spontaneous phenomenon arising out of a dynamic living system. Certainly a good flute can make it easier, but it’s only part of the formula.
I think that there are two main reasons: 1) they were made to a different pitch level (430 or even lower depending on age of the flute, and 2) most ITM players don’t use the proper fingering that the flutes were designed for (opening the f key when playing f#, for instance). Even when you play at a lower pitch and use the historical fingerings one still has to lip notes in tune; you play in tune not the instrument. By the by, I’m not an ITM player, I’m much more a traverso player where you have to really have a flexible embouchure as alot of notes are out without adjustment - its the nature of that beast (and the cross fingerings can be very different in timbre and response).
When I was playing Bassoon back in High school I found that Bassoons are made in a way that certain notes are out of tune and this is done on purpose. I later found out that this is pretty much the case with all of the orchestral winds.
I never really found out why, but I had always assumed it was to make compromises between certain notes as if one note was perfectly in tune it would throw another note way out of tune, so the solution was to make both notes slightly out of tune. Then it was up to the player to correct it. Now, I am no expert and do not know the first thing about physics, but this seems like a likely cause to me and it does make sense. Orchestral instruments are made to play well in all keys not just a few keys like we use in ITM. This probably has something to do with it.
We are still struggling with the laws of physics and the stretch limitations of our puny hands. And this has been well known since Boehm’s time. Remember Carte’s helpful image:
He’s taken an 8-key flute and put all the keyholes in line with the finger holes, to illustrate that, while ideally all the holes should be laid out in a logarithmic scale, like a guitar fretboard, in an 8-key they end up as several clusters, limited to what we can reach.
Ideally the hole that determines F#, hole 5, should be well up the flute and smaller, and hole 6 (E) should be further down and bigger. But who around here is going to vote for another 3/4" between fingers R2 and R3?
Similarly c# (hole 1) should be considerably further up the flute, to get it up to pitch, but that would make the c natural cross fingering even sharper!
My feeling is, to answer the question we started with, we need to separate out two important issues - the tuning of the notes we do have control over, and the problem notes we don’t have control over. Ignore F#, C# and the crossfingered Cnat, and look at the scale of the other notes. Before Siccama that scale was suitable for playing around 430 Hz, and it’s probably no coincidence that domestic pianos were tuned about there. Siccama used a range of scale lengths, and Pratten, deriving from Siccama’s work, settled on a scale length suitable for somewhere around 445 Hz (possibly Society of Arts Pitch?).
So we find Rudall style originals tend to be a bit flat at the bottom and a bit sharp at the top when tuned to modern pitch, but Pratten’s are pretty good. Modern flutes might be the same, or might have been tweaked to make playing less difficult. But while modern makers are free to tweak their scale lengths and get the controllable notes to their liking, they are still limited by stretch in terms of the “problem notes”.
Thanks, Terry; this helps. My question, then, is this: it would seem to me that a solution for the F#, at least, could be either enlarging the hole (but that might be counterindicated; on a simple-system flute a finger can only seal so much), or in the case of wooden flutes (where we have the luxury of chimney depth equalling wall thickness, ergo material to spare), undercutting, yes?
As to why a crossfingered Cnat (oxx ooo) must needs be sharp, I expect it has less to do with the holes themselves than the physics of the tone generated by the crossfingering. Obviously one can’t make the C# hole smaller, as one would have a too-flat C#. That’s no good. At least we have multiple options when it comes to nuancing our Cnats!
Yes, the practical limit to enlarging hole 5 seems to be about 11mm, and that puts it too large for some thin-fingered people. Undercutting certainly helps, but again has limitations.
And yes, everything we can do to increase the flattening effect of the cross fingering assists in distancing c# from c nat. oxx xox is good in this regard.
The best solution for those who really would like both c and c# to be accurate would appear to be the c thumb hole. Boehm was right! (even though he did it as a key so he could get better Bb/c functionality)
A couple people above have suggested that flutes used to be pitched lower than 440, but at least in England the opposite was true: “Old Philharmonic Pitch” was A=452.
Anyhow what nobody has hit upon is that what scale is “in tune” is situational, not fixed.
If you want to play the flute in tune with the pipes you’ll want to be in Just Intonation like the pipes are. The Fs will be 14 cents flat, for example.
If you want to play the flute in tune with guitars, keyboards, etc you’ll want a flute tuned to Equal Temperament.
The Just Intonation scale in D mixolydian is:
D 0
E +4
F# -14
G -2
A +2
B -16
C -31
(numbers reflect the deviation from Equal Temperament in cents)
That was certainly the official pitch of the Philo movement. But take an original flute from before 1847, tune it to A452 and see if you enjoy it. It’s unbelievably ghastly. They are hard enough to play in tune at 440 - they are perfectly impossible at 452.
Around 1847, Siccama released his 10-key, and did something quite unusual. He varied the scale length of the body, presumably in response to consumer request. It wasn’t a serial thing, a sign of development; he clearly jumped back and forth. Look at the C#-D# lengths on my Extant Siccama Flutes page.
Subsequently, in 1852, Pratten released his Perfected, based on the Siccama, and choosing the shortest of scale lengths Siccama had used. Pratten was the leading professional player of the time, so we can imagine he was aiming his flute at his professional colleagues. But even he didn’t go straight for 452. By my reading, he went for 445. So his flutes work well at 440. Bearable at 452, but not ideal.
I have only ever found one conical 8-key flute I think was specifically tuned for 452, and that was by Mahillon, a Belgian company. Weird, huh - Continental pitch was 435. Presumably made for export. And yet we find nothing but high pitch Boehm’s, Carte 1851 and 1867 models, Rockstro Models, Radcliffs etc.
I can only conclude (at this time) that the 8-key remained an amateurs instrument, set at the old domestic pitch, while the new-fangled cylinders satisfied the needs of the professional giggers at 452. I think Siccama and Pratten went as far as a compromise instrument around 445.
