The question that forms the title to this page might seem a strange one.
After all, a violin doesn't seem to be a complicated contraption with moving parts, except for the tuning pegs, and it's fairly obvious what they are there for.
However, when they are bowed, the strings of a violin obviously do move, at least a very short distance; not only is this visible, but in the absence of vibration, a violin would make no sound.
And the body of a violin is a hollow box which allows the violin to produce a louder sound than the strings would produce by themselves. This already needs some explanation. After all, a regular acoustic violin doesn't need batteries, so without a source of power, how can it amplify a sound?
Of course, it doesn't. In electrical terms, what the body of a violin behaves like is a transformer: it performs impedance matching between the strings and the air, so that more of the energy of the strings' motion can impart motion to the air. The strings are small; they move the much larger surfaces of the violin, so now they're grabbing hold of more air.
When the strings are bowed, they move from side to side, in a pattern closely resembling a sawtooth waveform. And thus, on a simple synthesizer - or the SID chip of a Commodore 64 computer, as I can vouch for from personal experience - one chooses a sawtooth waveform when one wants to make a violin-like sound.
The strings move from side to side, with the violin beneath them. How, then, does the bridge of a violin do anything more useful than slide along the belly of the violin from side to side?
The rather crude diagram above attempts to answer that first question.
The tension of the strings holds the bridge firmly against the body, so the feet can't easily slide from side to side. But the shape of the bridge is chosen so that the side-to-side motion of the strings tends to cause the bridge to tip over from side to side. Thus, its two feet move vertically, in a direction perpendicular to the surface of the belly of the violin, thereby being effective in making its whole area push on a large amount of air.
There's just one problem. While one foot pushes down, the other one is going back up. So it looks as if a lot of this motion, although it's in the right direction, is going to cancel out.
To see how this problem is dealt with, we need to take a look inside the violin.
Although a violin looks nearly symmetrical from the outside, on the inside the two feet of the bridge are treated differently.
One foot has the "bass bar" beneath it.
The belly, or front surface, of a violin is made of a soft wood like spruce or pine (Stradivarius always, or almost always, used spruce). The bass bar is made of a hard wood, like maple, and it runs across a large portion of the height of the belly of the violin.
This helps the foot of the bridge that lies above the bass bar to move a large portion of the belly of the violin.
The back of a violin is made of a hard wood like maple, sycamore, or pearwood (Stradivarius always, or almost always, used maple). Poplar, or some types of willow (while willow is usually thought of as a softer wood, there are harder varieties, and willow and poplar trees are closely related), is another possibility: although only occasionally used for the back of a violin, these woods are more often used for the back of a cello, and Stradivarius did so himself. The other foot of the bridge sits (with the belly of the violin in between, of course) almost on top of the "sound post", a wooden peg made of a soft wood, like spruce or pine, which connects the belly of the violin to the back of the violin.
Since the back of the violin is made of a harder wood than the front of the violin, that foot of the bridge now won't move as far as the other foot, which pushes on only the front and not the back. So the two feet don't cancel each other out.
The back of the violin does move, though, and in the opposite direction to the front of the violin. So together they either expand or contract the interior of the violin at the same time, with a bellows action.
The only problem with this otherwise ideal arrangement is that the sound directly radiated from the top of the violin will, therefore, be opposite in phase from the sound coming out of the f-holes in the top of the violin. Given the usual manner in which a violin is held by a performer, however, this is more of a problem for the cello and the double bass than for the violin or the viola.
This analysis is an oversimplification; the speed of sound is finite, and given the frequencies of notes played on the violin, this is significant for its acoustic behavior; so instead of being opposite in phase to sound from the belly plate, the phase relation of sound from the f-holes to sound directly radiated from the belly plate will depend on frequency.
Also, given that the sound post's two purposes are to transmit vibration to the back, which is made of maple, or another hard wood, and to reduce the excursions of the motion of the foot of the bridge near it, by transmitting the stiffness of the back forward, one would have thought that the sound post should be made from the same hard wood as the back, if not possibly one even harder.
The reason that instead a soft wood, like that used for the belly of the violin, is used is that the sound post is not glued in place, and so it must have the ability to be squeezed so that it can hold itself in place.
Based on this understanding, I wondered if the violin could be improved by replacing the f-holes on the belly with an aperture of the same size elsewhere; for example, on the upper bout, near the neck of the violin, on the side opposite the chin rest.
I see, however, that a similar idea has been had elsewhere, although in a case where the circumstances are different: the German luthier Thomas Ochs produces his own version of the Kasha guitar design by Michael Kasha with the sound hole placed on the side of the instrument.
Also, in the immediate postwar period, Julius Zoller made violins which had a set of small sound holes in the side instead of f-holes; however, these violins differed from conventional violins in other respects as well; their shape was different, and they had an extra string for sympathetic vibration.
To assist with the discussion that is to follow, I am adding two diagrams which illustrate the different parts of a violin. Incidentally, rather than attempt to draw a picture of a diagram myself in a paint program, I have used illustrations from old books in the public domain as my starting point. It may be of interest to note that the photographs of a violin used in the diagram immediately below are of the Earl Stradivarius, named after the Earl of Westmoreland, and they came from a book with photographs of several of the violins in the collection of Royal De Forest Hawley.
In the first diagram above, one thing to be noted is that two completely different parts of the violin are called the "saddle" of the violin by different sources.
According to a diagram of the parts of the violin I saw on one site, the curved portion of the neck of the violin, near where it joins the body of the violin, is the saddle.
Other diagrams use that term to refer to a small rectangular piece of wood which protects the belly of the violin, and its purfling, at the bottom from the pressure of the tail gut. This part is sometimes also called the "rest", not to be confused with the chin rest, not shown in these diagrams.
Speaking of small pieces of wood, the image of the ribs and back of a violin included in this diagram shows a number of small square pieces of wood along the central seam of the two-piece back. These are called cleats; normally, they are not part of the original construction of a violin, but are instead glued to the back when it is being repaired for that seam having come apart.
As we've already noted, the body of a violin has a back carved from a piece of maple, or sometimes another hard wood, and a front, called the belly, carved from a piece of spruce, or sometimes another softer wood.
Around the sides of the violin, maple is also used. In this case, thin strips of maple are bent using heat, to form the upper bout, the two center bouts, one on each side, and the lower bout. These are held in place by carved solid blocks of maple, the corner blocks.
As well, the upper end block and the lower end block brace the upper and lower bouts.
The top nut is a piece of metal against which the strings rest as they come from the tuning pegs and go to the bridge. This minor part serves two important functions: it defines the lengths of the open strings (that is, it defines how long the vibrating part of the string is when the fingers aren't pushing the strings against the fingerboard to shorten them for playing a higher note), and it ensures the strings are a distance above the fingerboard right from the start on their way to the bridge.
The tailpiece, which holds the far end of the strings, isn't permanently attached to the body of the violin. Instead, it has a metal rod, bent into a U-shape, and also bent towards the back of the violin, called the tailgut, which hooks on the tail button of the violin. The tension of the strings holds it in place, against the saddle at the bottom of the violin.
The fingerboard is usually made of ebony. It does not have frets, unlike the similar part of a guitar; one of the features of a violin is that, like a trombone, it can play notes that are continuously varying in pitch, with no striking change in sound quality between pitches that correspond to the piano keys and those that are between them.
Incidentally, note that the corners of the violin, due to the fact that the corner blocks are solid, are not part of the shape of the interior of the violin. This has been used in the shape of some modern violins made of carbon fiber, which omit the corners in their external shape - and, as well, in the nineteenth century experimental violins made by François Chanot had this form, as well as having f-holes of a simplified shape.
I was able to find an old image of a Chanot violin, but it had a heavy Moiré effect, so I had to process it somewhat heavily: