Sautille 'impulse'

December 22, 2016 at 03:13 PM · Hi,all,

I have been recently working on the sautille bowstroke.I notice that it seems very hard to start the first note of a group (say 4 notes),from a static,not bowing start point.I read in Nathan Coles

website that the sautille stroke is started by dropping the bow onto the string from 'above',depending on the sound desired.I presume that this is the 'impulse' that starts the sautille chain-reaction of bounces.Does anyone have any thoughts on this?Also I notice that it seems harder to play groups of 3 notes,like triplets with sautille bowing.Any suggestions on how to handle this? One last it easier to play sautille nearer the fingerboard where the strings are more flexible?

Thanks in advance,


Replies (52)

December 22, 2016 at 03:28 PM · Sautille is not a stroke I find easy to do. Some people start it on the string - others, apparently as Cole does. Fifteen years ago I had an adult violin student who was getting back to violin playing and seemed to be doing everything right for sautille, but it did not work, she had a middle of the road (about $600 today) German pernambuco bow. I let her try a cheap Glasser composite I had recently acquired for about $100 and her sautille just sprang out. So she bought a different bow - and stopped the lessons. So the bow you use can have a lot to do with it. Some bows act as though they have a "sautille motor" in them, but not many. Among the cello bows I have tried I found this property only about 5% of the time (that's from trying a lot of bows). I must admit, I have only found this in one of the violin bows I have tried - an d you also need very fine hand and arm control.

If you listen to enough music by great artists you will hear that runs of many fast notes appear seamless - they flow like a mountain stream - so triplet runs should not be any different than sixteenth-note runs in execution - the beat just comes in a different place - just don't emphasize it.

December 22, 2016 at 03:36 PM · In Sautille', the bow "hops" from the string. The bow is active, the hand is passive, in controlling the stroke. This differs from Spiccato -- For Spiccato, the bow is dropped from the air in a U or saucer shape, depending on the speed. In spiccato, the hand actively controls the stroke, the bow is passive.

For Sautille'...

First, make sure you are around the balance point of your bow, usually slightly below the midpoint.

Then, start by playing a short, quick detache' stroke at the balance point

As you accelerate the tempo and press into the string, the bow will naturally want to "hop" or rebound.

I think one misconception is that when we are playing off the string strokes, we are lifting. Not necessarily true. Do you lift a basketball? No, you hit it into the ground and allow it to rebound. Such is Sautille'.

December 22, 2016 at 03:42 PM · I agree with Andrew on the bow being a big influence. Andrew -- I love the term "Sautille' motor"!

Some bows have seemingly endless Sautille' potential, almost doing the stroke on their own. Others, I have to coax along.

December 22, 2016 at 03:57 PM · You need a methodical approach, starting with examination of your finger and wrist action, and using something like Sevcik Op. 2.

In spicatto, the hand action describes a U shape, but as the change to a fast sautille is made, the hand travels up and and down.

December 22, 2016 at 04:10 PM · I had the same experience in my first few months of playing when I was learning sautille. I had a cheap $100 bow and couldn't get off the string even with the right technique, but when I used my teacher's bow it was fine. My teacher, who's the concertmaster in a local chamber orchestra, wasn't able to do sautille on the bow I had either, so I guess sometimes the flaws in the bow aren't able to be overcome.

December 22, 2016 at 06:00 PM · Sautille should just happen when the bow is moving quickly enough and the player is relaxed enough. I don't feel like I really get it started, like I would a spiccato.

December 22, 2016 at 10:57 PM · This is the way I teach it:

I use these 2 terms:

Down Bow Position: (suppination) fingers curved, balancing the weight of the tip of the bow with the little finger.

Up Bow Position: (pronation)The position the fingers take if you move the bow hold to the very tip of the bow: much weight on the first and second finger, fingers almost straight.

Sautille´ (very fast spiccato)

1. Performed a few inches above the balance point of the bow.

2. The fingers are in down bow position. The hands is at right angles to the bow. 3. Use an up and down wrist motion. The motion is the same as tapping, or knocking on a door. There should be no arm movement, except for changing string levels.

4. The student should practice the tapping motion with his hands on a table, without the bow, and then with the bow.

5. Hold the bow firmly.

6. The bow remains on the string, but the stick bounces. If the bow stroke is done correctly the tip of the bow will move up and down.

7. Beware of lifting the elbow, this will put you into up bow position.

8. The principle of this bowing may initially be taught with the hand in up bow position. If done properly, the tip of the bow will not go up and down, but will remain on one plane. This bowing is called the eraser stroke.

If this doesn't work then I give them my "magic sautille bow" to try which is purported to always work to achieve the bow stroke. :)

December 23, 2016 at 03:41 AM · Hi Malcolm, check out this thread from a while back:, near the bottom

In it Nathan and I talked about how a colle motion can give an incisive start to sautille. In most ensemble playing, both sautille and spiccato are started from on the string in many contexts for the sake of precision. It's possible to start from off the string for both strokes as well, and in some contexts that can be desirable in chamber music (almost never in orchestral contexts.) Here's a link to a thread on colle motion we referred to in the previous thread:

So you can drop the bow, a sort of controlled crash, to start sautille, as in the 3rd movement of Tchaik. Or you can start with a colle motion, as in the Scherzo from Schumann 2 Nathan demonstrates so well for us.

As for triplet playing, it's difficult when strokes are uneven. Usually we play heavier and slightly longer down bows than up. Practice duples starting with an accented up, e.g. take Kreutzer 2, starting up bow, at first in a short detache. Make sure someone listening with eyes closed can't tell which direction your bow starts. Play K2 at various speeds starting up. As you shorten the stroke for sautille (try to get as short a stroke as possible, e.g. 1/8 inch strokes) make sure your down bows don't get longer or heavier than up. You can also play musically with phrasing and agogics starting up and try to make the exact same phrases as when starting down. All the bow divisions will feel weird and you may have to master playing at the frog if you have yet to do so. Next play K2 in spiccato and sautille, starting both up bow.

When you play triplets in detache, accent the first of every triplet and release (shorten length of bow used, not length of sound, for the last two triplets.) Make sure every accent sounds like every other, and all the released notes sound the same as well. Then play dead even (in length of bow, sound, articulation, finish, middle) triplets in detache. Play same with very short detache. Speed it up. Play spiccato. Sautille.

In terms of mechanics, make sure your elbow is opening and closing, smoothly alternating, no matter how minute the motion. A locked elbow (any locked joing for that matter) can kill off-string playing. As for speed and bow point, experiment with different tempi at different parts of the bow, faster toward the tip, slower toward the frog. Each bow differs in it's bounce frequency throughout the length of the bow. A good bow will have a vary large bounce range that sounds good. For a given dynamic, if you can find the best bounce point for the tempo, sautillee becomes much easier--truly the stick bounces for you. Of course you can and often will have to compensate with an extra kick here or there in real music. Contrary to popular belief you can play a very slow sautille close to the frog. It just sounds terrible. It's also possible to play quite a fast spiccato as cellist David Finckel demonstrates:

His whole series is great.

Sound point is similar to on string playing. Experiment with tilting the bow more, using less hair, closer to fingerboard.

December 23, 2016 at 12:33 PM · Hi,all,

Thanks for all the advice,I'llkeep (patiently) working at it.


December 23, 2016 at 04:05 PM · Yes, I agree that it can be started from on or off the string; that shouldn't matter much to the result, as I do either depending on the context and how aggressive it needs to sound.

And I may still be convinced otherwise, but I hope that the distinction between spiccato and sautille eventually goes away. I just haven't seen the practical benefit for those I've tried to help. I see the strokes as different points along a spectrum of hand/arm activity.

Yes, triplets are a bit tougher in the beginning but Jeewon's advice looks good!

December 23, 2016 at 09:35 PM · Nathan,

I think you are correct in that: "I see the strokes as different points along a spectrum of hand/arm activity."

I view that performing off the string bow strokes requires a technique similar to that in dribbling a basketball. When the player runs full tilt down the court he uses the whole arm as one unit. When he stops to ready for a shot, the ball is mostly controlled by the upper arm and the lower arm becomes less involved. When the player dribbles the ball very near the floor, the hand provides most of the control and provides the primary impulse for the activity while the rest of the arm just follows the action in a passive way.

December 23, 2016 at 11:44 PM · Isn't every stroke a "different point along a spectrum of hand/arm activity"?

I maintain that these are quite different, though related. Learning to play a fast spiccato and a slow sautille', and how to make a transition between them, opens up a world of expression.

Ivan Galamian says this about that:

Spiccato: "In this type of execution, the bow is dropped from the air and leaves the string again after every note. In doing so it describes an arc like motion. In length, spiccato can run the gamut from very short to very broad. A general characteristic of the spiccato is that the bow is thrown down on the strings for every single note."

Sautille': " distinguised from spiccato by the fact there there is no individual lifting and dropping of the bow for each note. The task of jumping is left primarily to the resiliency of the stick. It is best played around the middle of the bow. Contrary to common belieft, the sautille' stroke may be played from very fast to quite slow, although in slow tempos the spiccato will generally be a more practical choice."

December 24, 2016 at 07:53 AM · I'm glad you pull the Galamian quotes, because he's certainly an authoritative voice. If he were here, I'd have some questions for him though! :) It sounds from these descriptions as though spiccato is only a slow stroke. That's the only way I can understand the bow being thrown down on the string for every note.

According to this definition, if the stroke gets fast enough to act as a true "rebound", it's then sautille. For me, that happens at what I would consider a fairly slow speed. So I've found it most helpful to lump it all together as the spiccato "spectrum" since it's difficult to describe to a student what to do differently as the stroke gets faster.

For instance, in this video where I demonstrate what I call different speeds of spiccato, would you also define the very beginning of the video as spiccato, or sautille, or both? This is a question I'm always trying to get different opinions on:

And to go along with Bruce's post, because I first learned spiccato thanks to my experience with a basketball, here's the master himself, Pistol Pete, demonstrating his amazing dribbling technique! There's a parallel here, of course: where does Pete transition from arm dribbling to hand to finger?

December 24, 2016 at 10:53 AM · I feel that for advanced players the distinction isn't all that important...kind of like how the basic grips for tennis players are really just starting points for less experienced players of the game. They'll all start someplace, but advanced players adapt the basics to fit their personal needs, and what they end up with is less defined by category but their ability to flow between all their options at the time each one is most useful.

But I think there is a significant difference between the spiccato bow stroke a beginner learns, moving the hand in a "U" shape, and activating muscles to both drop and lift the bow, vs. the sautille they pick up first, which usually results from one of the "tricks" like the string-crossing ellipse, where the stick bounces but the hair does not, and sounds distinctly different from the basic spiccato. Once they can do both, then it's about exploring the range between them?

December 24, 2016 at 02:07 PM · Nate: For spiccato, the bow tip will describe the same U shape as the hand. For sauitlle', the hand moves in an oblique motion, opposite the up-down motion of the bow tip. My impression from the video is that you're starting with a spiccato and moving into a sautille' rebound for the faster part. Great video!

Gene -- For professional players who also teach like myself, it is important to know and communicate every detail :-D It may not be as important a distinction for playing as teaching -- as players we tend to have things become automatic, but there is value in developing a fast spiccato and a slow sautille', as they can have different effects.

December 24, 2016 at 02:07 PM · double post

December 24, 2016 at 04:27 PM · Interesting Doug, that sounds like something I used to know and then forgot! Maybe one day I can make a slow-motion video that would show that clearly.

December 24, 2016 at 06:04 PM · A video of this would be great.

It adds the the confusion that we often play sautille' when it isn't specified, or when spiccato is actually indicated. Some of this is because of the style of the times -- baroque and early classical bows couldn't sautille' (at least, not well).

If we look at the evolution of bows with regard to the baroque, classical, and romantic periods, we can start to understand what direction we should take in the music (or at least know how far we are deviating from what might have been intended). Bows probably didn't sautille' well until FX Tourte made his late-classical bow designs to introduce more spring into the bow. But spiccato was likely possible before this.

December 24, 2016 at 07:04 PM · And of course tempo choice has so much to do with it! The stroke that works in one tempo won't in another.

December 25, 2016 at 05:33 AM · It'd be interesting to see your bouncing stroke in slow mo, Nate!

I'm not convinced Nate goes into a rocking motion (what some of us are defining as sautille motion) as he goes faster. The tell-tale sign of that kind of jumping bow is in the motion of the hand, i.e. the hand throws the bow down, flexes at the wrist, on a down bow, thereby causing the tip to lift, and reverses on an up bow, causing the tip to drop. It's generated by a fast colle type motion which is amplified by the hand with a counter-motion at the elbow. Like this:

or this:

Nate's wrist starts neutral and doesn't flex much as he speeds up. Instead most of the bounce is generated by a kind of flippy motion: the elbow rises as the bow is thrown onto the string for a down bow, through a rotation in the upper arm, and reverse. At top speed, as the muscles oscillate quickly, the hand starts to kick in more (especially for the higher bounce) but the rotation of the forearm stays constant thanks to the simultaneous rotation of the upper arm (i.e. the forearm stays pronated as the upper arm rotates internally.) I'd bet slo mo video would show 'spiccato' motion in Nate's 'sautille.' For the rocking motion to occur there has to be a slight supination of the forearm on the down stroke.

I was taught the now classic (though fairly recent, in my estimation) spiccato v. sautille distinction also but Nate's argument is compelling. What is the purpose of these labels, which just add a layer of ambiguity? I like Rolland's (and Nate's) approach, describing actions instead of defining terms. That way we can communicate a direct set of instructions for a specific sound/stroke, and in most cases offer several ways of achieving the same musical ends.

December 26, 2016 at 03:40 AM ·

Teachers always win in the classroom. Students rarely question things, and when they do the teacher is always right, because a student doesn't have enough knowledge to oppose them with.

In a forum the rules change; we are not yes men. You will be question, and possibly be proven wrong. It is a place where teachers can learn, if they have an open mind.

The only thing these strokes have in common is that they are a form of staccato, that's it.


-notes can be slurred

-it is more of a forearm movement with some finger movement

-no movement in the wrist

-full control of weight on up and down strokes(even note possible)

-up stroke can be accented

-strokes are separate and change of bow direction is done off the strings


-notes cannot be slurred

-lots of finger movement in the stroke

-lots of movement in the wrist(the hand makes a circle like movement)

-uneven strokes: the down stroke is generally louder

-an accent on a upstroke is improbable

-sound is constant and bow direction change is done on the string

A very slow sautille compared to a slow spiccato are completely different motions and they use completely different muscle groups. To say Sautille is a fast spiccato, is wrong.

To the poster.

It's best to learn sautille with a straight wrist and the hairs flat on the string. You loose bounce when the bow is tilted and the wrist has a bend in it.

Another point, the letters U and C produce the wrong mental image for spiccato. The motion is more like an arc ) or smile :) . There isn't any vertical, or straight movement. Telling a student to make a C or U motion is definitely going to through them off.

December 26, 2016 at 06:56 AM · Hi Charles, I don't think anyone here is trying to win. Hopefully everyone is offering information and discussion. And I'm enjoying reading descriptions of different motions of the bow arm and hand, many of which I don't think about on a regular basis.

I do still question how much this distinction matters in a musical context. I certainly don't want to split hairs about terminology or anything, but I'm not convinced there are really two very different techniques involved. If someone would make or point to a video demonstrating slow sautille vs. slow spiccato it might start to clear things up.

Ideally you want to be able to show any desired character off the string: any dynamic, strength of articulation, etc. That seems to be what the various descriptions of spiccato and sautille get at. My interest in this subject as a teacher comes from the fact that everyone understands these terms in different ways, so I'd like to get a sense of what everyone has learned so far.

December 26, 2016 at 12:43 PM ·

Nathan, you just need to tilt your head a bit so you can see the bow nut when playing these to different bow types. The nut makes an oval pattern when sautille is played; whereas, the nut makes a back and forth arc ) pattern when spiccato is done. A close video of the frog when doing these bow types will show this.

Yes, its not a contest, but it is a debate. If a person has an open mind the winner is always them. I am sure the shoulder rest debate IS a contest.

December 26, 2016 at 04:49 PM · It's not a contest, but I think is an important resource for students and for that reason I like to see real supported information posted here.

Galamian not enough evidence?

Here is the great Kurt Sassmanshaus demonstrating the differences for his "Violin Masterclass" series. He begins by saying "[Sautille] requires a totally different technique [than spiccato]".



Here is David Finckel, famous cellist from the Emerson String Quartet, discussing the difference and transisitioning:

December 26, 2016 at 04:49 PM · double post

December 26, 2016 at 06:25 PM · Charles, I find the direction you're taking this thread a bit weird, but whatever... I will debate a point when I disagree, especially if I think the point made is misleading, or incomplete, but I'm here to discuss and share information and find new ideas. I get satisfaction from the exchange of ideas, not winning debates.

Douglas, I was arguing your very point, that sautille motion, where the hand or arm rocks the bow like a see-saw on the string, is different from spiccato motion, which swings the bow in an arc, reverse to the curve of the bridge. But I've learnt from Nathan's bouncing technique that you can simply use an arcing motion at all speeds to get the same sound as the rocking sautille. Finckel himself demonstrates this in the video I linked to above, calling it 'extreme spiccato,' and he uses it where there is mixed slow and fast bounced bowing. I've not yet tested it much myself, but it may indeed be a simpler, more direct way to teach bouncing strokes at all tempi. The caveat is that one may have to relearn how to coordinate the fingers/hand/lower arm/upper arm to get the timing right.

Off the top of my head, without having studied them, I think you might find people like Szeryng, Midori, and possibly Francescatti to use such a fast spiccato.

P.S. Paul Watkins is the new cellist of the Emersons, another amazing cellist and musician.

December 26, 2016 at 06:55 PM · There are different ways to achieve the same result. For instance, my fellow violin teacher at my university uses more of an arm motion for sautillé than my wrist generated variety. She studied with a Russian school teacher, Kevork Madridossian. Sassmanhaus studied with D. Delay so is in the Galamian tradition.I studied with Galamian, but he never said anything to me about sautille. I got my best information about the stroke from a Dounis pupil, George Neikrug. Neikrug said, by the way, that to equalize the up and down bow, practice starting up bow, so that the up bows are on the beat.

December 26, 2016 at 09:37 PM · Bruce -- that's interesting. My college professor was a Galamian student, and I recall he was given very specific instructions from him on spiccato, sautille', and blending them.

For anyone who isn't sure of the difference (or if there is one), I really encourage watching the Sassmanshaus videos above. Then, make your on conclusions.

December 26, 2016 at 09:37 PM · double post. again.

December 26, 2016 at 10:47 PM · I can't get this out of my head today for some reason! I agree that a video of frog/tip during the strokes would show a lot. They're not going viral on the internet at the moment though, so I may make one with some super slo-mo. :) I'm curious to see what it would show for my own off-string strokes (whatever we'll call them). As I transition from slow to fast, as I do in the beginning of my video above, would the shape change? And if so, is it in the end useful to teach it as two different techniques?

If we want to say that sautille is faster than spiccato (as Sassmanshaus says in his video), I can agree. Or that in sautille, the hair often doesn't leave the string (while in spiccato it always does), I can buy that. But I still don't know what a slow sautille would look like, or what practical applications it would have.

In addition, just so I have a starting point when I try a video on this, would you say in my above video that I start with a spiccato and transition to a sautille?

December 26, 2016 at 11:11 PM · Nathan -- it is my impression that your starting spiccato and going into a sautille'. Your right arm, right wrist, and the side of the bow tip would give a better indication of where the transition happens, but they are absent from the video.

I agree with Galamian who says "contrary to common belief, a sautille' stroke may be played from very fast to quite slow, although in slow tempos the spiccato will generally be the more practical choice". As we speed up there is no time for the individual controlled motion of a true spiccato and we likely end up playing fast passages sautille'.

By the way, there is a lot more detail in the Galamian book, several pages on spiccato vs sautille', too much to post. If everyone here doesn't have it already, I highly recommend getting a copy of "Principles of Violin Playing and Teaching" by Ivan Galamian. Read it front to back occasionally, and use it as a reference when needed.

December 27, 2016 at 12:20 AM · I have the Galamian book somewhere, as I used to read it quite often. I have Simon Fischer's books closer at hand these days. The quote you pulled is interesting, since I'm very into "practical" and I'm still not sure what a slow sautille would be used for.

My goal in all of this is not so much for my own playing, although I hope never to stop learning! But there is so much miscommunication in teaching/learning these strokes that I would like to simplify the process.

December 27, 2016 at 12:51 AM · I'm glad you brought up Simon Fischer. He agrees that these are different, listing them discretely. He says:

"In staccato, as in the other lifted or bouncing strokes, such as spiccato, sautille and ricochet, the fact that the bow wants to bounce is the single most important factor"

I'll give you one "practical" benefit: when I tell my students to play spiccato, they play spiccato as defined above. . When I tell them to play a sautille' stroke, they play sautille' as defined above.

Nate -- I thought of another way to check what stroke you are playing. You probably can't play spiccato without the pinky on the bow, because it holds the weight of the bow through the lift. You probably can easily lift the pinky from the bow for sautille' though, since the bow rebound lifts its own weight.

December 27, 2016 at 01:02 AM · Haha, I'm glad to see we're both thinking about sautille during the holiday break! I already sent Simon a note to see if he would like to clarify any of this. He's impressed me in the past as upholding the great teaching traditions while being willing to say when they didn't make sense to him. If he gets back to me, and is willing to share what he says, I'll post it here!

December 27, 2016 at 02:20 AM · That would be interesting to hear his input, Nathan.

Here is a video that shows the basic rocking motion of sautille' pretty well, from Todd Ehle aka. "Professor V".

Sautille' Part 1 and 2:

For the sake of completeness in this thread, here is his separate video showing a spiccato bowing:

December 27, 2016 at 04:48 AM · So I've been experimenting with 'u' motion fast spiccato and it's totally doable for me. You can easily tell which direction the bow is arcing (u motion v. see-saw) by looking at whether the tip of the bow dips beneath the plane of the string. It's less efficient when there are a lot of string crosses, but there's no reason not to just switch to a see-saw motion to make crosses. And I think that's a good reason to learn both types of motion, but there's no reason to favour one as a true sautille and the other as 'extreme spiccato' or what have you.

To keep the u motion, I have to make my pinky and third finger more firm than I usually do (as I think Finckel implies) and keep my wrist fairly neutral (i.e. hand doesn't drop, infact hand is often slightly raised in certain contexts.) Also, I tend to want to keep my hand more leaning than square. I think you can see this in Nate's demos too. In the dribbling video the wrist stays neutral, the hand stays fairly leaning, forearm pronated. In the Schumann and Saint-Saints videos, where there's lots of string crossing, the hand is much more square, the forearm supinated. Square hold makes the hand flap up and down, rather than hinging in the direction of the bow, and so makes it easier for string crosses. But rather than throwing the hand down from neutral, which would create the see-saw motion, the hand stays slightly raised and the hand is thrown to neutral. Tell-tale sign, then, of violinists who do fast spiccato more like a slower spiccato is a higher elbow and a raised hand, i.e. the wrist is lower than elbow and hand. Violinists who rock the bow tend to have a lower elbow and dropped hand, i.e. the wrist is higher than elbow and hand.

As for speed, clearly violinists who play with a low wrist have no problem with speed. As for the hair not leaving the string, that has no relation to what type motion is used to play fast spiccato; the hair can be made to stay or jump depending on how much the fingers/hand get involved, which can be done by throwing the hand below the wrist, or lowering the wrist/raising the hand and throwing to neutral.

Slow sautille is just a see-saw string cross, as opposed to maintaining a 'u' shape before and after the cross (why would you do that) e.g. Wieniawski Concert Polonaise 1, Op. 4. If we were to start analysing every motion involved in an off-string passage from a real piece I think it becomes clear the spiccato/sautille distinction gets severly blurred. If we start to analyze the relative motions of the various parts of the fingers/hand/arm, again, it becomes kind of silly. To get the hair to jump for a more percussive sound and for lots of string crossing we want the hand to flap more vertically, but why do we draw an arbitrary line (the plane of the bow) to decide to call one motion sautille and the other spiccato.

As for what the bow 'wants' to do, it always wants to bounce. That's what the camber in the bow is designed to do. Actually a sprung stick is by it's very design very bouncy. Much of bowing technique involves learning how to make the bow not bounce. And so learning bounced strokes has much to do with unlearning what we do to keep the bow from bouncing: leverage, weight transfer, pronation, leaning the stick, wrist leading the stroke, etc.

December 27, 2016 at 07:13 AM · Thanks for the discussion so far, and I'm going to take a break until I have new information to contribute.

I'm definitely aware of the various definitions that are out there for these strokes! As I said, I consider myself a permanent student, and I've been at this a while. But this is one area where I've found the traditional categories limiting for high-level playing.

So far, my proof is that if these are truly two totally different techniques, it shouldn't be possible to make a seamless transition between them as I do in my video. And I haven't yet seen a practical demonstration of a slow sautille.

Wishing you the best holiday break and into the new year!

December 27, 2016 at 09:03 AM · Hi,

Thanks for all the input,everyone,I kind of feel like Ive opened

up a 'can of worms',lol,It seems a little like sautille is a bit like vibrato....hard to quantify,because the physics are variable.

I also wondered if its any easier to learn sautille using a

composition/fibreglass bow?

December 27, 2016 at 06:10 PM · Nathan -- You're welcome, glad to debate this. We'll agree to disagree. Even if you have new information, I'm unlikely to be swayed since we have Ivan Galamian, Kurt Sassmanshaus, Paul Rolland, Todd Ehle, and many other great teachers describe different techniques for these. Your "proof" is a logical fallacy -- you state that because you can make a seamless transition they cannot be different techniques. If you can seamlessly transition from A to B does not make A=B or B=A. Anyway, agreeing to disagree, and appreciate your contributions.

Malcolm -- I think you've touched on an important topic, and one that can be very fun for the student, opening up a lot of great sounds. I hope the videos and quotes I have posted open up a new avenue for you for expression with your instrument. Have fun with these strokes!

Will a composition bow bounce well? Some of the better CF bows bounce very nicely indeed. I have an Benoit Rolland CF bow from when he made them in the 1990's, and it has an amazing "spring" effect. Of the newer lot of CF bows, I find the Jon Paul bows to have a nice springiness to them, much more than many other makes which I find rather stiff.

December 27, 2016 at 06:10 PM · double post.

December 28, 2016 at 05:07 AM · I wouldn't say a can of worms per se... maybe Pandora's box :) But seriously, as Bruce said, there's always more than one way to achieve our ends and that's what's great about this forum. I think the more tools we have at our disposal the more we're able to express what and how we want to.

I just got back from playing Pirates of Penzance and was using Nate's fast spiccato all over the place. I for one am glad to have this extra tool to pull out when I need or want it. I find it's a more refined stroke than my usual rocking motion. Thanks Nate!

That's not to take anything away from the rocking motion. It too is a great tool to have, especially for a soloistic sound. But I wonder if it's the best way to be introduced to sautille. The rocking motion, because it adds an extra kick, enables the player to cheat the bounce a bit, by which I mean it enables sautille at points other than the best bounce point for the given tempo. Many sources suggest starting sautille with an extremely short fast detache to discover where the bow starts to kick, the best bounce point, and gradually let it kick by undoing the 'detache hand.' At that juncture some would advocate rocking the bow. But I now believe rocking is not necessary, it's not fundamental to the sautille sound. It's an added motion.

Regarding carbon fibre bows, I don't think it's the best material to learn bowing technique. For the money it might be the best option. But if you're able to spend a few thousand, I would recommend a good Pernambuco stick that handles well. The problem with synthetic materials is that they're kind of coarse tools. They bounce well but they're harder to stop suddenly. They're kind of on or off, without gradation. Some are too edgy. Others are kind of rubbery or laggy. They lack the immediacy and subtlety of wood, if that makes any sense.

December 28, 2016 at 05:07 AM · Double

December 28, 2016 at 05:07 AM · Triple!

December 28, 2016 at 11:57 AM · Hi,

Reading through this makes for very interesting reading, including the videos! Thanks!

In the traditional Franco-Belgian approach, there is a difference between spiccato and sautillé. Basically, the spiccato is an off the string stroke, while in the sautillé, the stick bounces, but the hair doesn't leave the string (probably already mentioned, but worth maybe re-mentioning). The actual italian word for sautillé is saltando (you can see that used in the last movement of Wieniawski's second violin concerto, where it implies a sautillé as Wieniawski studied in Paris). With respect to Galamian, since he studied with Capet, that is probably why he distinguishes between the two.

Someone mentioned Szeryng, and he had an interesting take on this. He had some reserves in that spiccato often was used in a percussive way. His thought was that the strokes should be lateral in motion primarily, and keeping the lift to a minimum.

To answer the question above above a possible slow sautillé... In the way I was taught, the location in the bow changes depending on the speed of the stroke - therefore the faster, the higher up in the bow, and conversely the slower/the lower. The sautillé begins not as off the string stroke, but as a détaché in the area in which the bow will bounces at a given speed. It remains primarily therefore a lateral stroke. So, I do find that a slow sautillé is possible personally, and I do use it at times. I find that for the various sautillé strokes to work, the lighter the bow hold and letting the bow do the work, so figuring out where a bow can/will do the stroke at a given speed, all bows beings different and some easier than others, are the most important.

Anyhow, very interesting discussion for the most part, and thought I would throw in some thoughts.


December 28, 2016 at 09:12 PM · Christian -- Great input! Wieniawski 2 is a fantastic example.

It's intriguing that you bring up Capet. I have his "Superior Bowing Technique Treatise" around here somewhere -- I will have to find it and see if he has any specific thoughts.

Could you elaborate on Szeryng? Was he saying that spicatto was too often played by default with too much accent / articulation?

December 28, 2016 at 10:02 PM · Capet considered spiccato a variant of sautille. I used to think he was backwards in many ways, but now I wonder. He does not consider them completely different strokes, but like Nate, views them as being along a continuum, belonging to the group of what he calls "the light strokes."

We forget that even a legato stroke requires the arm to be suspended, as does on-the-string-sautille. It's not as if the suspension of the arm for on-the-string playing is vastly different from how it functions for lifted spiccato playing. Rather the difference is in degree, coordination, timing and leverage, and how we combine horizontal and vertical components. Also, much of how we conceive of bowing is an analogy. There is no real passive motion, as all motions are generated by the arm. There is no motion of the bow, rebounding or otherwise, without motion from some part of the arm. No part of the arm can move without a countermotion from some other part of the arm. If we try to prevent the countermotion a seizing of muscles and joints occurs.

As has been acknowledged, we've all heard the terminology and (if not before, now) know the different concepts of executing various strokes. It's great to have various resources linked to in one thread. But the question remains, in Nates words, "is it in the end useful to teach [spiccato and sautille] as two different techniques?" From my brief experimenting I no longer believe you have to significantly change coordination between fingers, hand, lowerarm and upperarm to transition from slow to fast spiccato, as I used to. It's not really even something I believed, just something I did without questioning. You can simply focus on bouncing height as Nate demonstrates. Or you can focus on bounce point of the bow, etc. You can use a U shaped motion throughout, or a rocking motion throughout. It's also possible to change from a U to rocking motion and vice-versa.

Christian, I was wondering about your Szeryng reference too. Do you have a source for that? I have some notes my teacher took from a seminar with Szeryng I should try to find. It's interesting you mention that as you can hear it in his playing, so clean and refined. I think he also prescribed always playing a few inches away from the frog, as he thought the sound was too harsh there. Not sure where I read that. But that might be a function of his bow hold. Oistrakh clearly had no difficulty playing cleanly right at the frog.

December 29, 2016 at 05:23 AM · I haven't looked at Capet's Superior Bowing Technique in a long while. But skimming through the bits on sautillé reveals a lot in common with Nate's idea of 'points along a spectrum,' except that Capet uses sautillé as the more encompassing term and reserves spiccato for a slower, biting, bounced stroke.

From p.57, The Sautillé

135. In the same way that a rubber ball thrown to the earth rebounds because of its encounter with the ground, in the same way the Sautillé should have as a goal to throw the bow on the string and to elicit from this action the rebound, owing to the simultaneous springiness of the hair and of the stick. However, a factor enters into these principles, which, in our opinion, is of greatest importance: that is the absolute submission of this rebounding to our will.

It is necessary to be able to control this rebounding and to give it the slow, moderate, or rapid tempo required by the musical elements in which it must manifest itself. For this, it is thus absolutely necessary to be master of all the tempi of one's bow, and consequently, to have a means of constant control of these movements; it is also necessary to be able to introduce in this bow stroke the most powerful and the most delicate dynamics; because the Sautillé should not be only and definitively: light and animated by a movement determined by the habits of the wrist, at all times. One must be able to accentuate certain notes in this bow stroke (moderate or rapid) and for that, one must be master of one's Sautillé to the point of being able to do with it exactly what one desires; it is certain that in a slow tempo the Sautillé becomes Spiccato; however, it is necessary to give the Sautillé the possibility of accenting it like the spiccato; one needs a biting Sautillé in certain cases, despite a fairly rapid tempo.

Sounds a lot like Nate's video, "Master spiccato at any speed."

He reserves the term Spiccato for a biting bow stroke, which never gets very fast. So in that regard I don't think he really cared whether the bow was moving in a 'u' shape or like a see-saw. It's the articulation which distinguishes Spiccato as an accented Sautillé, which itself can also be a biting stroke when faster than Spiccato, what we might call 'percussive' sautillé.

In summary, the Spiccato is a more biting Sautillé than the on-the-string Sautillé or the rebounding Sautillé; it cannot and should not arrive at too rapid a tempo, but on the contrary should be reserved for certain passages in which the character of will holds the principal place. It is a type of Sautillé of which the goal is to develop Accentuation. Its preparatory practice is, in our opinion, one of the best for the muscular study of the right hand fingers, it is at the same time an excellent preparation for the Staccato.

Anyone who has worked through Sevcik Op. 3 will have gone through such a "muscular study." Capet's use of the terms are accurate: sautillé = jumping; spiccato = detached, separated, with every note performed in a distinct and pointed manner. Our common usage, from Galamian's definitions, is not precise. It seems like either he or his writers confused the terms somewhat.

One last point, though the rest of the world has lumped together the French and Belgian traditions, I don't think that is how French violinists would identify themselves. Capet is of a French lineage going back to Baillot through his teacher Maurin.

Edit: so ingrained is our incorrect use of the term spiccato that in the 1991 English translation of Baillot's The Art of the Violin, by Louise Goldberg, the term "détaché sautillé" from the French is translated as "The Spiccato."

December 29, 2016 at 02:26 PM · Hi,

Jeewon: very interesting posts!

To answer the questions about Szeryng, I read that either in that segment on him in the series "The Way They Play" or in an interview-article in Strad magazine from the 1980's. It's been a long time, and I remember more the thought than the source. And yes, you can hear that concept in his fabulous playing!

I think that the reason terminology gets skewed up is that in the late-19th & early 20th century, many traditions melded together. Galamian fused elements of Russian/Armenian playing with ideas from Capet and the Franco-Belgians into his own approach; so we maybe getting his interpretation of the terms rather than exact reference to others. Even the "Russian bow hold" for example was something first labelled by Carl Flesch in the Art of Violin Playing after looking at Auer's students and assuming that Auer was teaching it (apparently, not directly - there is a discussion on this in The Great Masters of the Violin by Boris Schwartz). The French and Belgian traditions kind of came together through the Massart family who were teaching in Paris and Brussels, so you are right Jeewon in that lineage plays a role.

As for the rest - Nate's way is a great way of looking at it, no doubt!!! In my own mind at this the present time (who knows, I may change after this thread!), I still find there is a difference, since in essence in a sautillé, the hair never leaves the string, whereas as with the spiccato there is a lift or a moment between the strokes where the hair doesn't touch the string.

Have to run for now...


December 29, 2016 at 02:44 PM · Hi again,

And just because my mind is now on this also, here are some approaches I ran across on video as a result:

James Ehnes discussing spiccato:

And a video of Pinchas Zukerman teaching the last movement of the Tchaikovsky Violin Concerto (labelled about position, but involves a lot about strokes including the spiccato in this context):

Hope this adds something useful to the discussion...


December 29, 2016 at 03:49 PM · Jeewon Kim said: "There is no real passive motion, as all motions are generated by the arm. There is no motion of the bow, rebounding or otherwise, without motion from some part of the arm."

That is true, Jeewon. When we refer to "active" vs "passive" we are usually referring to how much the spring of the bow participates or accomplishes the stroke. Take ricochet, for instance. Of course the hand is controlling the stroke, but is a passive element -- the bow spring is active and doing most of the work. Sautille' is like ricochet, with the bow spring very active, while in true spiccato the hand is more active (like an off-the-string detache').

I did once see a cello play itself entirely passively, the bow with no interaction for the right arm, but this was at the Wizarding World of Harry Potter in Universal Orlando :-D

Thanks for pulling out the Capet quotes! That really adds to the information in this thread. It's clear that he thought spiccato and sautille' were different as well, although our definitions may vary.

I agree with Christian -- I still find a clear difference.

Christian -- thanks for posting more videos. I played for Zukerman in a master class once -- it was fantastic. A kind man and a heck of a player!

December 30, 2016 at 04:18 AM · Christian, I agree the terms are all skewed up, and I think it's our job to deskew them :D

Thanks for those videos. I think they demonstrate the old French terms and ideas perfectly. I think Ehnes and Nate are in agreement and, except for using the term 'spiccato' instead of 'sautillé,' like we've all been doing, probably since bowing technique started to be discussed in English, they're also in agreement with Capet. In both their videos they demonstrate sautillé at various tempos. They do not demonstrate a more accented sautillé, nor a proper spiccato, which is an even more accented sautillé at slower tempos (you can see some of that in Nate's Schumann, Saint-Saëns and Tchaik demos.) You can get a biting sautillé or spiccato by adding a collé motion to actively pluck the string with the bow.

I think one of Galamian's greatest contributions to bowing technique is in his description of the collé stroke. I can't find a source prior to Galamian, so I'm assuming he was the first to describe collé in detail, which I think the French just called biting or accentuated staccato or sautillé. They were interested in sounds and character, rather than motion.

While I find Zukerman inspiring to watch as a player, I've always found it frustrating to watch his masterclasses, live or on video, unless he doesn't have much to complain about (when he can focus on interpretation.) His demonstrations are great, but he just relies on a handful of catch phrases (pun unfortunate but intended... groan) to communicate with the student. If only he would talk about collé instead of repeating 'catch' over and over.

The kid in the video can't do what Zukerman wants because of poor timing, inconsistent application of pressure and inability to add and release pressure abruptly. His arm often shoots too fast, before his hand catches the string because he can't feel the cushion of the bow nor the resistance of the string, what Capet calls 'penetrating the string with the bow,' before moving the bow. He doesn't feel flexion at the wrist and baseknuckles, before pulling a down bow, i.e. he doesn't have a collé motion. Instead he often breaks the wrist (extends it) on a down bow and leads with the wrist like you'd do in a smooth détaché. He doesn't feel his first finger push against the stick on up bows to catch the string, with an up bow collé motion. He probably also has no control over his roulé, to control how much hair he has at his disposal. I'd get him to put aside the Tchaik for a couple of weeks and give him Kreutzer 2, 7 and 8, and a few variations from Sevcik Op. 3 until he developed absolute control over his detaché, sautillé, collé and spiccato.

He's a perfect example of an apparently advanced student who lacks skill in fundamental bow strokes, but what's more, he doesn't understand how one stroke is connected to another. He can only think in binary terms, on or off the string. He doesn't see the progression and relatedness of the variety of strokes within the détaché/sautillé family (using the French definitions,) those "points along a spectrum." He needs to understand how to use the elasticity of the stick in detaché, how to turn detaché into an on-the-string-sautillé (or perlé, as Baillot called it) using shorter and shorter strokes, how to transform perlé into an off string sautillé by employing the spring of the hair, and finally how to 'bite' the string by adding a collé motion. This stuff all goes back to the original French school.

Baillot, in his The Art of the Violin outlines bow strokes produced with the bow on the string:

1. the grand détaché

2. the martelé

3. the staccato;

and strokes produced using the elasticity of the bow:

1. the light detaché

2. the perlé

3. the spiccato, which is actually the sautillé in the original French

4. the ricochet, or thrown and rebounding staccato

In the French he has the heading, "Détachés Élastiques" and proceeds to list the types:

1. Détaché Léger (Light Détaché)

Separate each note, holding the bow very lightly on the string, taking advantage of the elasticity of the stick to give an imperceptible and slightly elongated bounce (sautillement imperceptible et un peu allongé.)

2. Détaché Perlé (Perlé Détaché)

Separate each note in the same way [as for the light détaché], taking advantage of the elasticity of the stick. Give very little length to the bow stroke because of speed.

(In a previous section he adds, "all the notes must be equal, separated, round as pearls--which why this bow stroke is called the perlé.)

3. Détaché Sautillé (Sautillé Détaché) [Spiccato in the English translation. I'm beginning to think the French school never used the term spiccato at all.]

Make the bow bounce lightly in the same place, leaving the string a little.

4. Staccato à Ricochet ou détaché jetté [sic] (Ricochet Staccato or thrown détaché)

The ricochet is played down-bow or up-bow, but usually down-bow. The player throws the bow at the lower end of the middle third, and from about two inches above the string; the bow rebounds and "bites" several notes by itself. When playing up-bow, the violinist must lift the bow quickly off the string after each note.

Capet says:

In reality, the Sautillé is a Détaché in a very reduced division of the bow. So that, before letting the bow rebound from the string, one must do preparatory practice by pressing the bow on each note as for the Détaché, but in divisions which are shorter and shorter. Practice slowly. The bow should not leave the string between notes, and profit from the springiness of the stick... The principal point is to thoroughly accustom oneself to penetrating the string with the bow; this exercise has as a goal to establish the springiness of the stick alone without participation of the hair in this springiness. ...This method ... allows ... master[y] of Sautillé. One could call this the on-the-string Sautillé and it is in practicing this... that one succeeds in giving more liberty to the bow and that one can allow oneself to make the hair leave the string.


Douglas, I don't find the idea of the hand as passive very useful. How do you let the bow do most of the work? What does it feel like to let the hand be passive? In order to have full control over our bow strokes we have to always be aware of the elasticity of the bow, it's aliveness, I would argue, for all strokes, even long, on-the-string strokes. As soon as we load the sprung stick, it is activated. It doesn't suddenly become active when we move it back and forth in short strokes, or throw it onto the string. How we apply that load, the effective weight added, the forces employed, the angles used, the changes in direction and load, the rate of change, is what determines the activity of the bow. Whether we're damping the kick, or letting it kick, or injecting extra kick, with just the fingers, or the hand, or the forearm or upperarm, we're always regulating the spring of the bow. We can't just think on/off, active/passive, like the kid in the video, it's a continuum. The finer the gradation of action we have with every part of the arm, the more control and nuance we have over the variety of sounds. In a ricochet, as the stick bounces on the hair after the initial throw, you might conceive of the hand as passive, and the bow acting on it's own. But actually we dampen the rebound with the fingers to get the exact number of bounces and speed of rebound we want, or we raise the upper or lower arm to allow a higher, slower bounce, or we move closer to frog or tip to control the speed, or we flatten or tilt the bow to increase or decrease the 'bite.' Even if we consider the least interference from the bow arm, i.e. somehow we get the arm to exactly parallel the motion of the bow after it's initial bounce, we're still suspending and moving the arm along the trajectory of the bow, and still regulating the firmness of the wrist and fingers. Passiveness is just an analogy. Control is always action.

Capet literally thought of the spiccato as a kind of sautillé, very closely related to the biting sautillé. It's right there in black and white. I don't understand how you can come to the opposite conclusion. Our definitions don't vary. We have it backwards, because the French invented the definitions. Some anglophone, somewhere along the line came along and just wrote spiccato in the place of sautillé. Now it would be correct, if a bit redundant, to coin the term détaché-spiccato as an equivalent of perlé or on-the-string sautillé because of the meaning of spiccato, but spiccato does not by itself include the notion of bouncing, jumping, or throwing, so unless you include it in the family of sautillé, you may as well just use the word staccato to represent any stroke which sounds detached and short, which most non-string players do. As for Nate's question I now think spiccato is truly at the end of a single sautillé-spectrum, which includes a light détaché at the opposite end. There are always vertical and horizontal components in every type of bow stroke, whether on or off, long or short, slow or fast. As with the bouncing ball, which gets faster as it loses height, so too all the horizontal and vertical motions get smaller as the stroke gets faster, and bigger/slower, for sautillé, given the same dynamic and articulation.

December 30, 2016 at 12:57 PM · In an earlier thread started by Nate on this subject I said the following. When I have more time after New Year's Eve I may expand upon it:


Raphael Klayman

September 27, 2016 at 07:46 PM · I don't remember how I learned it; I think I figured it out for myself. My approach is similar to Christian's. I view sautille physically, more as an outgrowth of the detache, even though sound-wise it more resembles spicatto.

If we put the bow to the string at its balance point, angle the hand more or less at right angles to the string, rather than slanted, hold the bow somewhat loosely and play fairly rapidly, a sautille should emerge, where the bow springs but doesn't leave the string. That said, with a fairly good mastery of spicatto, we can get to sautille from that direction as well - i.e. by using shorter strokes, moving faster and keeping the bow more on and at its balance point. But I usually come at it from a detache direction.

According to Dounis/Leland as I recall, the cut-off point between spicatto and suatille tempo-wise is about 100 to the 1/4 note when playing 1/16ths. I keep the movement for sautille mostly in the hand and fingers

This discussion has been archived and is no longer accepting responses.

Facebook Twitter YouTube Instagram Email is made possible by...

Shar Music

Shanghai Isaac Stern International Violin Competition

Pirastro Strings

International Violin Competition of Indianapolis

Yamaha V3 Series Violin

Dimitri Musafia, Master Maker of Violin and Viola Cases

Metzler Violin Shop

Gliga Violins

Corilon Violins

Meadowmount School of Music

Find The Song You Want To Play Next: StringClub

Anderson Musical Instrument Insurance

Bobelock Cases


Nazareth Gevorkian Violins

Los Angeles Violin Shop

Violin Lab

Wangbow Violin Bow Workshop