Friday, September 29, 2006

Solfège battles

Today my counterpoint class got waylaid by a discussion of movable-Do solfège versus fixed-Do. One of my students has absolute pitch, and is completely convinced that fixed-Do is the way to go. She also studied in France for a year, so she may be influenced by the Conservatoire system. As I've written before, fixed-Do solfège relies upon a strong pitch memory, and this memory requires constant reinforcement. My student tried to argue that fixed-Do is an interval system, but it is only a generic interval system. The C - E major third is sung "Do Mi," as is the C# - E minor third and the C# - Eb diminished third. So at best, singing "Do Mi" signifies a third, but cannot specify which kind of third it is.

I teach my students that solfège is meant to remind one of aural images. Movable-Do solfège reminds the singer of tonal functions. Tonic is Do - Mi (or Me) - Sol, Dominant is Sol - Ti - Re. This holds constant regardless of the key, so students can rely upon the aural image of these functions to aid in singing. Fixed-Do solfège reminds the singer of pitch classes. Tonic functions switch with each new key, but pitch names remain constant.

I actually use both systems in my own teaching. My students learn Movable-Do syllables, and also sing on letter names without accidentals. The latter system is exactly like Fixed-Do in that a C# and a C are sung exactly the same way, but without translating from the English name to the Italian name. I emphasize the Movable-Do more than the letter names, but being aware of specific pitches while singing is of practical benefit to instrumentalists. The letter system also allows the practice of clef reading and transposition. I actually got this idea from Bill Marvin, current director of aural skills at Eastman. Bill uses movable numbers (One, Two, Three, Four, Five, Six, Sev) to indicate the scaled degrees and uses the Fixed-Do syllables, so switched from my practice. As Eastman has a large number of students from countries that use solfège syllables as note names, this makes some sense. However, it is rather difficult to sing numbers with any speed, given the ending consonants of the last four numbers. Letter names also have that problem (try singing "eff gee" six times very quickly), but as I emphasize the movable system more than the fixed system and the letter system has only one note that ends with a consonant, it still ends up winning the contest.

To me the most important thing is that my students can sightread well, can look at a score and play the music in their head, and can listen to music and recreate it on a staff or on their instrument. For the 95% of students who do not have absolute pitch, movable-Do will help in these tasks; fixed-Do will not.


Becca said...

What's your take on minor sight-singing methods? Do- or la-based minor? And why? (I've used la-based minor in past, and I'm curious about do-based minor.)

Anonymous said...

I begger to differ on one point, Scott. My experience is that for instrumentalists, moveable do is worse than useless when it comes to real-life sightreading and learning pieces and playing them on their instruments, at least on string instruments, the only area in which I have many years of actual teaching experience. I grant moveable do is great for geting a sense of how the music is put together, and with the different vowel sounds it does create more awareness than singing numbers for the scale degrees. So I grant its usefulness in the development of general musicianship, awareness, and understanding.

I'm glad you use also use a fixed system with the letter names. I have my cello students sing the pieces they are learning for me with the letter names. To play and sight read well, has to be a stable and instantaneous linkage in the brain of the player between a note on the page, a pitch, a SINGLE letter name, the spots on the fingerboard where that pitch can be played, and the myriad of ways to actually play it.

For the purpose of actually playing music on an instrument, looking at a G, a cellist must hear the pitch in her head, know what the note is, and select from what be as many as 20 different ways to finger that note (i.e. if the note is the a half-step higher than middle C, it can be played in that octave on each of the four strings with any of the five fingers).

If you look at a the top space in the bass clef and that note is do or re or mi or sol or fa or ti, it doesn't help much when it comes to knowing how to finger it. And on an instrument like the cello which uses three clefs, a fixed system because all the more important.

I was trained in fixed do and often use it myself because as you point out, the letter names do not all roll trippingly off the tongue. And the reasons I outlined above are, as I've understood it, much of the underlying reasoning for the French fixed-do solfege system.

I don't know any instrumental teacher who encourages some form of singing with note names as part of the learning process of instrumental pieces who advocates a non-fixed system. I have heard more instrumentalists than I can count, though, give students the message I'm summarizing here.

As long as the kids get fluent singing with the letter names, I can live with y'all teaching this moveable do stuff instead of using numbers. Especially since the value for instrumentalists is singing the note name internally while playing--where it doesn't matter if the consonants are problematic to physicially enunciate or not.

And I imagine that doing moveable do in ear training classes and also singing with fixed letter names is better in some ways than fixed do and numbers (again from the prespective of preparing to play a piece on a string instrument) because it's also very helpful to sing the piece singing the fingerings. That way we don't have two competing number systems!

The biggest frustration for me at DePauw is that since I think in fixed do, it's hard for me to do extensive demonstrating for my students of singing pieces using letter names; I tend to short circuit and start mixing them together. So I sing fixed, they sing letter names, and we are surviving quite nicely.

Nevertheless, I do like to regale them with my anecodtoe about the legendary and fearsome Renee Longy storming into my eartraining class at Juilliard, evidently having just left a meeting of the theory and ear-training faculty. "I do NOT understand this moveable do! Why do they want to do all the pieces in do? Some pieces are in re, some are in fa. Music would be so boring if everything was in do!" Mme. Longy was a more than a bit elderly at the time; we all found it (very silently) amusing that she had missed the point.

Scott Spiegelberg said...

Becca, I teach Do-based minor for the same reasons that I teach movable-Do. Do-based minor keeps all of the tonal functions fixed with the same solfege syllables, with only chromatic alterations. Thus the dominant chord is still Sol-Ti-Re, as opposed to Mi-Si-Ti in La-based minor. La-based minor is easier for switching between relative major and minor keys, but as I don't require students to shift Do unless they want to (though I do encourage it in modulations), students can still end up singing a relative minor passage in La-based solfege if they desire.

Eric, as a trumpeter I use movable-Do all the time for sightreading. With transpositions up the wazoo and trumpets in five different keys, I can't take note names at face value. A written G4 could be an F (Bb trumpet), a G (C trumpet), an A (D trumpet), a Bb (Eb trumpet), an E5 (A piccolo) or an F5 (Bb piccolo). It could also be an A as written for a D trumpet, which I play as a C on my A piccolo (a common occurence in Baroque music). So it is much easier to know what key I am playing in, and then let tonal functions translate what notes I should play. That's also how I do clef reading. My trumpet teachers often would sing in movable-Do solfege, as did my horn teacher.

I'm curious how you talk to your students about tuning. Adjusting pitch for thirds of major or minor triads requires knowing that the notes are thirds. And a Mi will be adjusted differently than a Ti, since the Ti needs to stay a "true" half-step away from Do. At Eastman we had a three-hour masterclass every year on tuning issues, with scale degrees used to communicate these tuning issues for various orchestral excerpts.

Anonymous said...

Well, I'm still trying to figure out how to play in tune myself. Maybe the problem is that I think in fixed-Do!

My experience has been that string players who discuss the nuances of tuning--what Casals called "expressive intonation", for example, and the differences from equal-temprement that my HIP friends confuse me with--do so just discussing the interval qualities and scale degrees by number or by mediant, submediant, subdominant, etc.

As I was writing my diatribe above, I was thinking that on transposing instruments fixed-DO may not be any advantage, at least not in the same way as it is on string instruments.

dan said...

One reason for using fixed do is that it is impossible to sight sing anything that modulates at all.
A moveable do teacher once tried to explain to me how it was possible, but it involved doing a theoretical analysis (which precludes sight singing)and still didn't account for the fact that during a modulation, certain notes will have two separate funtions (It's hard to sing two syllables at once). Using numbers to learn scale degrees comes in handy when later discussing roman numerals for chords and figured bass.

paul bailey said...


you are right that moveable-do falls apart whenever you modulate out of a relative key. i have no problem to let students sing fixed do on the harder more chromatic individual tests as long as they work sing moveable-do in class. as for the other battles over la based or do based minor, i think our students should be know both (the cole melodia book is a great example). if you are ever going to do any teaching in secondary school you never know what you which system you will find. any arguments about which system are better are usually trumped by seniority. at the schools that i have taught we may have not have agreed which was the best system, but for the students benefit we put our differences aside to agreed to teach only one.

ThumMeister said...

Two comments.

First, thre's the problem of "level of abstraction." Pitch is an abstraction of frequency; interval is an abstraction of the relationship among pitches; and temperament (comma sequence) is an abstraction of the relationship among intervals. Music theory happens at the interval and temperament levels of abstraction, but music notation and instrument interfaces are standardized at the level of pitch.

Much of the difficulty of teaching music theory, I suspect, arises from its being at a different level of abstraction that the "visible, tangible" notation and instrument interfaces.

If the display and control of musical information were at the same level of abstraction as music theory, would students gain the knowledge and skills of music making faster? I don't know.

What do you think? Please let me know at

ThumMeister said...

Second comment.

Another source of difficulty in teaching music theory, I suspect, is that music theory is abstract and intangible. Systems such as the neo-Riemann tonnetz make it more visual/tangible, but are divorced from performance.

The simple, consistent structure of music can be made both visual and tangible by using (a) a notation based on the combination of a chromatic staff and moveable Do, and (b) an isomorphic keyboard with electronic transposition to keep the notes of the current key associated with Do, Re, Mi, etc.

Such an isomorphic keyboard is composed of two overlapping tonnetz's, thus unifying performance and analysis. Such a notation does not need a parallel Roman Numeral system for analysis, as its use of Moveable Do (with a La-based minor) already presents information at that level of abstraction.

Displaying and controlling musical information in this way would appear to capture the structure of musical sound as the Periodic Table of the Elements captures the structure of the atom.

It combines a level of abstraction that is the same in every octave, key, and tuning with the concrete tangibility of a keyboard instrument. The keyboard note-patter fits very well on a standard computer keyboard, facilitating the deployment of online music lessons. It also fits the keyboard of the Thummer (, which, with its ability to control up to ten degrees of freedom, may have more expressive potential than any previous musical interface.

The downside of this system, of course, is that it is not suitable for most musical instruments. It should work great with the conputer keyboard, the forthcoming Thummer, and the human voice, however.

I would welcome your comments at

Anonymous said...

To me, fixed do is does not add anything additional to just using note names--it's just another way of doing it. C becomes do, D becomes re, etc. It's simply like saying something in one language versus another--neither is inherently better or worse, but you might as well use the one that is spoken most widely if you know both. It doesn't add any additional understanding. Why not just stick with the more widely known letters than bother with fixed-do? Moveable-do adds a lot for those not confused with pitch distinction. I've read that many people are born with "perfect" pitch, and many other brain capabiities, but that it is weeded out as being useless and in fact hindering language capabilities unless someone grows up learning a tonal language such as Vietnamese. Anyway, I digress.

Anonymous said...

I use the letter system for singing as it gives the benefits of both systems. I use numbers and roman numerals for chord progressions and figured bass. I do not have perfect pitch but have a finely tuned sense of relative pitch.