On the state of present-day violin making

An endeavour is made to outline some ways in which the modern state of violin making is, for the independent maker, not encouraging, and, from the creative point of view, not very rewarding.
The General Outline is largely the fruit of historical research, with a philosophical twist, as is the next section. The epistemological section questions several ways to gather knowledge of violin making in general, and of old violins in particular. And why, taking into account scientific methodology, the currently popular methods do not give tangible results. A tentative new approach, by directing research of old violins in a slightly different direction, is suggested.

General outline

Modern violin making is obviously in a very strange predicament: the best we can hope to make is a copy, in looks and sound, of an instrument made centuries ago. It has even come to the point that several well respected violin makers stopped working to order. The reason being that it makes the whole process rather stressy, because they don’t feel sure enough of the sound qualities of the resulting instrument. Imagine Amati telling Charles V that he was welcome to come around and see if there happened to be a few instruments that might suit his court orchestra.

The craft of violin making is, in the sense of creativity, obviously stagnant from the 19th century, or even earlier, on. The general string scene is nowadays dominated by the trade, offering either fantastically priced antiques, or mass produced instruments, priced so low that it obviously verges on the perverse. And that also from the 19th century on. Why, we would surely hesitate to call this progress, wouldn’t we? In any other craft the situation of modern violin making, wherein prospective customers try to avoid you, the living maker, instead trying to lay their hands on a product by one of your colleagues long dead (that in overwhelming majority of cases has nothing to recommend itself but age, and changes hands for an amount of money the maker would never even have dreamt of), would unhesitatingly be termed unhealthy.

That is absolutely not to say that modern violins cannot compete with old ones. By now we all know from several blind and even double blind testing sessions that the good modern violin easily competes with whichever revered old one. The problem seems to be not the top level of modern making, but the general level. We have to realize that what we consider still an all times high, is de facto the average quality level of the classical Cremona period. Compared to that, the output of modern makers generally is a lot more erratical and, to our perception, the average level lower.

Not surprisingly the modern way to remedy this, is to take massive recourse to modern technology: computer, CNC router, the sound analyzing laboratory etc. Introduction of technology some three decades ago however doesn’t seem to have brought us perceptibly further so far. So we resemble the physician who prescribes medicine and, if it doesn’t work, simply doubles the prescription. But in my view there is a good reason to suspect that modern technology is not the solution of, but part of the problem. For one, the old masters obviously didn’t have recourse to any of our modern apparatus. Apart from that they were real human beings, exact like we. I refuse to consider them as demigods. So, what worked for them, must inevitably work for us too.

The socio-economic point of view

The old ones worked for wealthy patrons, but after the final collapse of feudalism in the early 18th century, violin makers had to work for, as a rule not very wealthy, musicians. A change of economics that is hard to overestimate. As a consequence the classical organisation, production and transmittance of the craft had to disappear. When violin making gathered new momentum in the early 19th century, the scene had radically changed. The making of violins by independent violin makers was no more the economical stay of the trade. Essential trade secrets had not survived the transition, and without the well organized workshops of old, the transmission of the remaining ones was rather left to chance. To the recognition of this situation in the 19th century, the founding of the State School of violin Making in Mittenwald, 1858, clearly bears testimony.

A lot of energy was instead directed into making violins that any normal musician could afford. It was the era of the first industrial revolution and violin making also became more and more industrialized. So much so, that we have come to recognize the basic methods of 19th century violin making as the only way to do the job. The idea that there may have been in existence another approach, is considered weird nowadays. The main difference between inexpensive and well made instruments then and nowadays are the woodworking skills of the maker and the care bestowed. As to sound quality the two categories do not differ markedly. In both the quality differs from instrument to instrument, and kind of unpredictably so.

On the other hand a rapidly growing amount of energy is these days put into marketing classical Italian and in their wake other old violins, with Vuillaume as the first really important player in this field. He introduced the business model for the violin trade that has been successful to these days. Right now we see millions being spend on marketing “Cremona” as the ultimate brand name in violin making, thereby heavily capitalizing on the reputation of makers long dead.
And marketing, we all know if left to our own thoughts, is for a large part make believe.

The epistemological point of view

That is to say: in which ways can we gather and transmit knowledge of violin making.
Maybe we can agree on the two most important ways, that is to say quantitatively, to expand our knowledge of violin making: careful study of top level old instruments and/or scientific study of the acoustics of the violin. In the following I hope to make clear that, and why, both methods, no matter how attractive and viable they look at first sight, will finally be unable to generate the desired results.

Let’s start with the old instruments. And let me admit first that there is a huge amount of violin making that we can learn from studying the old masters. Violin making, mine not excepted, is unthinkable without the legacy of the old makers. But I honestly do think that we can also learn from them, that each violin was started as a fresh project, whereby not a single effort was spent on exactly copying an example. Yes, they built after existing moulds or other plans, but did so with a total disregard of strict uniformity, incidentally called “refreshing freedom of execution” or “creative workmanship” or “panache” when an old violin is admiringly described in marketing speak. The most illuminating example being of course Guarneri del Jesu, many a violin by whom wouldn’t have survived the first round of a modern violin making competition.

As a corollary we have to swallow the fact that the mild disregard for exact execution in those days formed absolutely no obstacle to creating a great sound. And the smaller the scale on which we regard these creations, the more obvious their individuality. When we come to the tenth’s of millimetres we describe plate thicknesses with, the results are markedly individual and chaotic. And so to a degree that it is unimaginable that the result can be considered the outcome of geometrical intentions. Also to a degree that, taking into account that the chaos differs from violin to violin, we may consider this quality the result of another process than trying to make a plate according to some geometrical prescription. The thickness distribution must be the unintended result of quite another action. No coming around that.

In other words: the old masters not only disregarded the possibility of making bench copies, but treated the geometrics of their work partly in a way that has never been intended to be copied, even is to be regarded as essentially uncopyable. And this leaving totally aside the problem of variability of wood etc. At this point I suggest as a tentative conclusion that we can learn a lot from studying old instruments, but not how they attained their sound by carefully copying or realizing a geometrical recipe of some sort.

That said, let’s now move on to what we can learn from studying the acoustics of the violin, and let me start to admit that modern violin making, mine again not excepted, is unthinkable without acoustics. But I do think that the violin as a whole will, also for some time being, defy complete acoustical description. Broadly speaking the level to which computer modelling is able to handle violin acoustics, is not markedly better than computerized weather forecast. We understand chunks of the matter, but the whole is a little too much. A violin consists of a reasonable limited collection of parts. But, these parts do not function, i.e. vibrate, independently from each other. They function together in very different kinds of ways as coupled systems. That may be overcome still, but the strength of the coupling varies markedly according to the pitch. And even these coupled systems interact with each other at some point.

This picture leaves us with a formidable array of interdependent variables. So many so, that if we want to come to a scientifically clear understanding of how a violin works, the task begins to look utterly daunting, and, if we are honest, may well be considered far beyond the capabilities of an individual maker or an individual acoustical laboratory. The reason being that, if we want to research a system properly in the scientific sense, we can only move on surely by altering variables one by one. And that again not even speaking of the inhomogeneity and unpredictability of wood.

Even if one day we would be able to extract from our acoustical labours a geometrical recipe for a great sounding violin, using wood that has been tested and selected by computer analysis, who is going to make it? Not me, nor you, nor any living violin maker. The only way to realize such a recipe properly, that is exactly, will be by CNC machines, or even still more refined mechanical contrivances to come.
So here I propose another tentative conclusion, that trying to understand the violin by acoustical research to the point that you are able to predict the sound, is a rather unpromising approach in terms of time and energy involved by itself. But, granted that we succeed one day, it would mean the end of violin making as a craft practised by living makers.


I hope to have lent some credibility to the following thoughts:

  • violin making as it was, is not violin making as it is;
  • the old masters didn’t bother to copy anything and they did surely not try to figure out the ultimate recipe for a great violin, to be realized with the utmost precision. They made violins in a superb but still perfectly human way;
  • to figure out said recipe is, even with help of modern technology, an undertaking so complex, that, for the time being, it reaches well into the superhuman;
  • provided we succeed one day, only machines will be able to realize the recipe.

Provisional hypotheses

  • To try to learn how to make a great violin, by studying and copying the geometry of existing ones, new or old, and, doing so, try to improve the ones to come, is a dead end, as regards the creative aspect of the craft.
  • The old masters succeeded clearly in making great violins without any recourse to painful exactitude of execution.
  • Allowing the conclusion that in what I called before the “essentially uncopyable” of their work, must reside the clue to their succes.

Proposition for a new and practical way of making violins

Though the transition is slightly rough, I can only recommend to continue by reading “The last secret of Stradivari III” from the preceding section of this website.