Geometric Design of the Stradivari Model G Violin, Part One: Mold and Template

by Robert J. Spear

Originally published in American Lutherie #93, 2008

see also,
Geometric Design of the Stradivari Model G Violin, Part Two: f-holes by Robert J. Spear
Geometric Design of the Stradivari Model G Violin, Part Three: The Scroll by Robert J. Spear



I have little doubt that artists, artisans, and architects of the Renaissance and Baroque used some system of guidance for their drawings that was based on the knowledge of geometry and the use of straightedge and divider. I began my drawing adventure almost five years ago by following the guidelines for the geometric design of the Model G in Sacconi’s book and soon discovered errors. Even so, I was convinced that it would be worthwhile to use a classical Cremonese approach based on geometry because I wanted to see if I could integrate it with Hutchins and Schelleng’s scaling theories used for the New Violin Family. While the acoustical aspects of the exercise are not germane here, I worked to realize a design system that would essentially produce a second generation of octet instruments close to a classical Cremonese violin in the style of the Model G Stradivari. My goal was to impart a greater uniformity to the octet family’s models, but to keep this article within bounds I have confined my remarks to the violin.

There are those who question whether geometric design really played an important role in violin design and suggest that the model outline could be designed freehand. Others allow that some sort of geometrical or proportion scheme was used, but that it was not based on the golden section. A few ask why one can’t just get a good photo of a good model and enlarge or reduce it at the local copy center. You can (and I did at first), but because strange things start to happen in the larger and smaller instruments during the scaling process, straight scaling does not hold up. Still others, including Sacconi, stress that the eye was the final arbiter of any design, no matter how it was derived. I will attempt to address all of these points in this series of articles.