Un-Natural.
Playing the violin in the conventional manner requires some rather un-natural contortions. The left elbow is held more or less directly below the violin neck. This requires some contortion when the instrument is played in the third and higher positions, with the left hand near the body. In first position, with the left hand near the scroll, more contortion is needed to keep the elbow directly below the violin neck. This problem is more extreme for viola players as they extend further to play a larger instrument. Viola is notoriously physically harder to play, and viola players suffer more injuries.
It occurred to me that some of this problem could be relieved by twisting the viola neck so that the elbow wouldn’t have to reach so far to achieve the conventional position relative to the left hand. At first the idea of a twisted neck seemed absurd. As a restorer, I’ve had to straighten twisted necks. Twist has always been a boogyman to be slain, but could it actually be a benefit?
Visualizing
I found the whole thing difficult to wrap my head around. Wouldn’t a twisted neck require twisted strings to go with it? These might be hard to find. With the simple experiment of holding a loop of string in my hands like a cats cradle and rotating it, I convinced myself that it would be possible to string a twisted neck using normal straight strings. At this point I was hooked on the idea and couldn’t leave it alone.
The way that I imagined this working was to have a normal bridge on the viola, and then to add an even twist along the length of the neck and fingerboard, ending at the nut, where there would be a normal head. To visualize this, imagine holding the viola on your shoulder, ready to play. This viola has a soft neck. Grip it firmly in first position and rotate clockwise, as though turning a door knob. After that the neck becomes rigid and, voilà, you have a twisted-neck viola!
This was a fun geometric problem and, if it worked, it could possibly bring some relief to suffering players. I imagined that it might even extend the career of a mature player, and I thought that this was worth investigating.
Reality Check
Experience has told me that if one of my great, revolutionary ideas is any good, then someone else has already thought of it and explored it. To save time and disappointment, I went on the internet where I found that people, such as Torzal Guitars, have been making guitars with twisted necks for some years and for the same reasons that I had in mind: to address the physical contortion necessary to play. The problem is worst with electric guitars which people like to play with a cool, low-slung hold. Again, the problem gets worse with larger instruments, and the most popular application for twisting the neck is in bass guitars.
Building
How to make the neck and fingerboard? The twist needed to develop at an even rate, and the mating surfaces of the neck and fingerboard had to spiral in exactly opposite directions. I thought of several schemes based on laminations, molds and chalk fitting, and then realized that this would be a relatively simple task using a CNC machine. A normal neck and fingerboard would be drawn on the computer and these could be twisted at a keystroke.
I went to my friend Mike Center, who has a small CNC machine that he uses in making guitars. Of course it wasn’t quite as simple as I thought, but after a good number of hours and many trial runs, Mike presented me with a fingerboard with a 20° twist along its length, and a neck block with the mating twist in it.
I took the fingerboard, neck and a viola body with me to the 2023 Oberlin Volinmakers Workshop where I assembled it.
Tone
One of the big unknowns was how the twist would affect the the sound. I had realized that the twist would alter the angles that the strings bend or “break” over the bridge, significantly increasing the break angle on the A-string. Break angle can have a big effect on the tone of a fiddle so this was a concern. Sam Zygmuntowicz suggested that I could relieve some of this by using a split-height saddle, which I did. In the end the viola sounded normal!
Playability
The reaction to the finished instrument has been interesting. Some people seeing the neck, couldn’t play it at first. Others had less trouble, and some, not noticing the unusual neck, played for some minutes without noticing.
A fatal flaw
Though both the tone and playability of the viola were good, there was a problem: the C-string buzzed on the fingerboard on all the notes up to E in first position. The rest of the fingerboard was buzz free. At first I thought that this could be fixed by adding more “scoop” to the board. This is the standard approach to fingerboard buzzes so I started radically hollowing out the board in the first position where the buzz was occurring. That didn’t help, so I started to analyze what was going on. The problem, I realized, was that the string mostly vibrates from side to side on a plane that is aligned with the direction of bowing, and it was catching on the most elevated section of the fingerboard. The most obvious solution to this would be to de-elevate that section of the fingerboard by untwisting the fingerboard, but that would defeat the original aim of the exercise.
A second solution
It occurred to me that if the string was placed right on the very extreme edge of the fingerboard, then it wouldn’t be able to touch the board as it swung out towards the bass. This would require a radical realignment of the pegbox relative to the fingerboard. Or, I could mimic the same effect by adding a “Romberg” to the fingerboard, where the string would sit on a crest and the board to the bass side of the string would be out of the way. I tried it by adding a cello style flat section, or Romberg, to the board. I tried this and it didn’t work. The buzz remained in the first position on the C-string.
End of the Road
At this point I felt that the experiment was at an end. My plan had been to find a serious viola player who was suffering from shoulder and neck problems and loan them the viola for six to twelve months and see if the twisted neck would in fact help with their problems. However I know that most players would rather sacrifice their bodies than have to play an instrument with permanent string buzzes, so I didn’t pursue that avenue.
The viola hung in the shop for a year, it was fun showing it to players and messing with their minds, but it just wasn’t a viable instrument. So when I needed a viola to send to Tom Metzler’s Contemporary Violin and Bow Makers Exhibition in LA, I removed the twisted neck, fitted a conventional one, and sent it off.
Unexpected Twist in the Tale
A few days later, i started writing this blog post to keep other people from wasting their time down this particular rabbit hole. While making the diagrams to explain the concept of the direction of string vibrations and the use of the Romberg, I realized that I had not exhausted all the possible solutions. I had placed the C-string so that it sat on the top of the crest of the Romberg. If I had placed it in the conventional manner, with the C-string over the flat section, then I think there is a good chance that this would have solved the problem.
Sections of the fingerboard and strings at: A. the wide end of the fingerboard, B. The nut end of the fingerboard.
The green arrow represents the direction of bowing and string vibration. The pink smudge represents the extent of travel of the vibrating string . Up near the nut the side to side vibration is not as wide, but, due to the twist in the neck, the angle of bowing, relative to the fingerboard is much steeper, causing the vibrating string to touch the fingerboard, and buzz.
Enlarged section of the fingerboard and strings at the nut, showing: C. the way that I cut the Romberg, leaving the string on top of the crest. Apparently this was not enough to clear the fingerboard from the vibrating path of the G-string. D. shows how I should have cut it, in the conventional manner, with the C-string over the flat of the Romberg.
Had I not removed the twisted neck I could have adjusted the Romberg and tested the theory in about 20 minutes. As it is, I still have the twisted neck and It could be fitted to another body but this would take a couple of days work to put it into the body and another couple to take it out again if the experiment failed. After kicking myself for the near miss in timing, I decided to write up the project anyway and will report back when I get around trying the neck in another body.
