Just what everyone needs- a treatise on tone production and the proper stroke. Feel free to gag. Let's face it- you hit it, it sings. I reject the notion that there is such a thing as "the proper stroke." (Although, an improper stroke is one that breaks the bar). There are a multitude of strokes and an even greater multitude of musical contexts. Problem is, many think they know all about this subject and have already written about it and argued about it without really being able to hear and see exactly what technique makes for the good tone they are trying to produce. What is really needed is a machine able to precisely reproduce specific motions, and a recording device that can interface with graphic software to show every aspect of the sound envelopes produced. This technology exists, contrary to what some may think. I have often heard people say that the ears are more sensitive than the machines, and musicians and scientists do not hear in the same spectrum. Don't ever say that to a recording engineer or sound designer whose livelihood depends on a combination of the listening skills of both the musician and the scientist. (You'll never work with a decent engineer again). No, I don't have this equipment or the software, but I welcome the chance to try it out. I am also prepared to change my mind if necessary. In the meantime I'll have to rely on what my ears tell me. And as with anybody else's ears, how the brain interprets the information is subjective at best.
For this discussion I will relate the phenomenon of four vibrating surfaces: a timpani head, a practice pad with a plastic head, the floor, and finally a marimba bar. Let's start by observing a timpani. Position yourself so that you can see the reflection of a light source off the timpani head. It is easier to observe what follows if the light source is fluorescent light fixture mounted from the ceiling, and if the timpani is 32 inches in diameter. At the timpani's lowest pitch, strike the drum and observe the reflection of the light until the head stops vibrating. You can see how rapidly the head is vibrating. Do you think you have the ability to move your hand or the stick at that speed? Now pedal the drum up to its highest pitch. It should be almost impossible to see the individual cycles of vibration. This should demonstrate that even the quickest upstroke cannot match the speed of vibration of a relatively slow moving surface. Conversely, it should be possible to move the hand or the stick slower than the timpani head. After striking the timpani, you can either get out of its way so it can freely vibrate, or you can inhibit the rebound and change its sound. By the way, if you use both hands and play a really tight flam (not quite popped) you might begin to approach playing both strokes within a single cycle of the head's vibration. But even though it will produce a specific overall sound, our ears are not equipped to separate the sound of each hand.
Now observe a similar phenomenon using a practice pad with a snare drum stick. Since the vibration of the head on the practice pad is impossible to observe visually, you'll have to rely on your ears. Holding the stick with only the fore finger and the thumb in your normal fulcrum position, play several freely rebounding strokes on the pad. Do the same while holding the stick down at the very end. Totally different sound, yes? There are actually two sounds happening simultaneously: the sound of the pad, and the sound of the stick. To hear the sound of the stick, try the same experiment on the floor. It should be easily surmised that the difference in the sound is the result of a freely vibrating stick vs. a restricted vibration of the stick. Now, back to the practice pad, same procedure: using a freely vibrating stick and one whose vibration is restricted. Notice the effect of the head not being allowed to vibrate freely. The pad and the stick are contributing to the changes in the sound. If the stick can freely vibrate, so can the head, and if the stick's vibration is restricted, the head's vibration will also be restricted. This same phenomenon can be heard on a snare drum, but it is more difficult to filter out the other sounds of the drum to observe the contrasting sounds.
It should be obvious that the stick itself is the only other body that can move at the same frequency of vibration as the head on a practice pad. Our hands can either get out of the way, or restrict the two vibrating bodies. Playing with an exaggerated upstroke may be useful for learning to allow the instrument to vibrate freely, but practically a player cannot move more quickly than the instrument. Here's what follows: exaggerated upstrokes are a difficult motion to repeat. The faster the repetitions, the more you realize you have to rely on the instrument and the stick to produce enough rebound to repeat the cycle. The player moves the stick toward the instrument and the instrument responds in kind.
How does this experiment work on the marimba? Not well for several reasons: first of all, try to observe the cycles of vibration of the marimba bar. I mean try really hard. If you can move your hands that fast, let me know. Also, the marimba bar doesn't produce nearly as good a rebound as a drum head or the floor, so the player has to execute both the downstroke and a large part of the upstroke. The designs of the various models of sticks produce widely varying amounts of rebound and different timbres based on the amount of yarn, mass and durometer of the core, and whether the mallet head has additional materials such as latex tubing. Wood dowels and rattan do not have the physical characteristics that allow timpani sticks and snare drum sticks to bounce as freely. There is also the issue of where the player is holding the stick. For those of you who use Steven's technique, realize that you're holding the stick on the very end. That renders the experiments with the timpani, practice pad, and floor less applicable to the marimba- you only have one choice of where to hold the stick, so one important parameter of the experiment is omitted. One thing I will assume based on the previous experiments: a player cannot move the stick faster than the bar can vibrate, but the player can restrict the vibration of the bar. I will relate this theorem to how a player can alter the timbre (upper partials perceived) by changing the stroke.
I use marimba sticks with latex tubing around the core, relatively tightly wound yarn or cord, and wooden handles (rarely birch, usually maple). I also wrap most of my sticks so the shape of the head is more round than oval so there is not a great deal more yarn on top of the stick than around its middle. This design aids in the rebound of the stick off the bar. I also adjust the tension of the cord suspending the bars to help produce a better rebound. All these factors help to produce predictable and repeatable strokes. And that is the key: predictable and repeatable strokes. I'll return to that concept later.
If you want to reproduce the following experiment you'll need sticks similar to mine. Sticks with excess yarn on the top will color the different sounds. Sticks with loosely wrapped yarn or two-toned sticks will spoil the experiment as well. The best bet is to grab that set of sticks that should have been rewrapped ages ago and cut off the yarn. Hopefully, there is also some latex tubing (or felt, or similar material) around the core. If not, why do you like that sound!? Or, don't you dare use those on my marimba.
Please accept the fact that volume is produced by stick velocity and that timbre is altered by several other factors that we will demonstrate. Find a specific pitch on the marimba that is not too low to listen to the attack without being confused by the sustain that follows. It should not be so high a pitch that the decay is too short either. Try middle C. To take your responsibility for the stroke out of the equation, hold the stick with your fore finger and thumb at the very end of the shaft. Keep your hand and the end of the stick about 1-2 inches higher than the keyboard, and at a distance so that the mallet head is resting right on the sweet spot you always aim for so carefully. Raise the mallet head to its full vertical position, perpendicular to the bars. Let the mallet fall to the bar without adding force as best as you can- you want gravity and inertia to do all the work. You'll have to be ready to catch the stick after its tiny rebound with your other hand- you don't want to add another parameter to this motion such as trying to inhibit or aid the rebound with the hand that is holding the mallet. You'll have to try this several times to get an overall effect of the sound as it will be somewhat inconsistent in volume. (We really need a machine to do this consistently).
Notice that when the mallet head is making contact with the bar the shaft is parallel with the surface of the bars. I call this "flat sticking." Once you have listened to the results of this setup, raise your hand and the mallet shaft so it is high enough for the mallet head to make contact at the point where the latex tubing ends. That is, up on the corner of the mallet head- the stick will not be parallel to the bars during contact. The elasticity of the tubing is affected by the smaller diameter of the core on the top. Also, since the mallet head has a little farther to travel (greater acceleration) the volume may be affected, however most people's ears are not sensitive enough to measure the difference. But the rebound is different, and the timbre is also different. Depending on the particular materials of the mallet head, the elasticity of the tubing, and the material of the shaft, you will get unexpected amounts of rebound that may be more or less than when your hand is at the same level as the keyboard.
How you characterize the difference in the timbre (I hate the terms, "bright and dark") will be different if you use a stick with a different design. Actually, this experiment may only prove that what you expect to hear may be totally different from what you actually hear and will surely be different if you use different sticks in subsequent experiments. When you use sticks with different wrapping methods, various yarns or cord with various amounts of wear, or different shafts, you have an experiment that should show that any number of timbres can be produced just by changing the angle of attack which affects the resulting rebound. Can the player aid the stick in getting out of the way so the bar can vibrate freely? To some extent, yes. But, can the player move faster than the rebound? Hardly. The whole point is that you have to listen very objectively and without a preconceived notion of what will happen. Then, you have to be ready to admit that the results are different with contrasting stick designs.
The one factor that confuses players more than anything else is relative volume. Many players claim that they can make a bigger sound by generating the stroke with the forearm. Without taking the angle of the attack into consideration and ignoring just how much the wrist and and fingers are assisting in their stroke, they have a totally uncontrolled experiment. In general, they usually only produce more volume by virtue of increased acceleration. They also claim that they have added mass to the mallet itself when all they have added is velocity. (By the way, for you Einstein fans, since the velocity will probably never approach the speed of light, we'll assume the mass will remain constant). If that assertion was true, large people would produce more sound than small people. Such demonstrations of the forearm generated stroke are also performed in isolation: widely spaced repetitions that are relatively easy to reproduce, and outside the context of performing a musical passage.
A change in the relative velocity of the stick and the angle of attack brings another factor into play: the consistency of the mallet head. The latex tubing abound the core has an inconsistent amount of elasticity because of the shape of the core. The greater the stick velocity, the more the sound of the core can speak through. Since the core is generally harder than the latex, more upper partials are audible at higher volumes. Add yarn wrapping to the formula and the sound becomes even more complex. The second most confusing factor is timbre and its psychoacoustic affect on ring time.
Simple thought experiment: two different sticks with identical mass, one hard, one soft. Use the same procedure as before- two fingers at the end of the shaft and an unaided drop toward the bar. If the masses are equal, and the velocities are equal, the same amount of energy will produce the same volume of the fundamental. Which is going to sound louder? The one that produces more of the upper partials. (It will in fact produce a higher decibel level, but only because of the addition of upper partials to the overall sound). Those upper partials add to the initial attack of the envelope and die out very quickly. They also mask the sound of the fundamental during the attack. This may give one the impression that the sound is shorter because the upper partials are so fleeting, and the ear is not able to readjust to perceive the fundamental. However, the total ring time of the fundamental will be virtually the same with the hard and the soft stick. The design and material of the bar will greatly affect its ability to sustain any of the partials including the fundamental, especially in relation of the sustain to the attack.
So what do players do? Some play with short snappy strokes because they think "quick" relates to "short," "sharp" relates to "staccato" and "fluid" or "slow" relates to "legato." Nope. Remember, you can't move quicker than the bar can vibrate. Besides, "quick" or "sharp" usually translates into increased velocity which allows the hardness of the core to speak through the outer layers of the mallet head. To execute these short snappy strokes the players invariably keep their hands low so the angle of attack is on the flat part of the mallet head: "flat sticking." Conversely, to make a fatter, more fundamental laden sound, they use a forearm generated stroke that may or may not make contact with the bar at the same angle. This is a totally uncontrolled experiment which ignores so many variables, especially stick velocity, and players describe the two effects as staccato and legato, neither of which is applicable because of obvious differences in volume. With the particular design of my sticks, I get more attack sound (higher ratio of upper partials to fundamental) while flat sticking, and less attack sound (lower ratio of upper partials to fundamental) with a greater angle of attack.
Other players who realize they can inhibit the free vibration of the bar attempt to "leave the stick on the bar" just short of dead sticking. Good luck in reproducing that stroke. Increased pressure in the grip when holding the sticks by the very ends hardly changes the properties of vibration in the shaft, and therefore has little or no effect on the natural rebound of the stick. Although, if the grip is tight, the wrist is also tight and so the upstroke may be restricted. This may result in a slight difference in the ring time, but it could also result in tendonitis or carpal tunnel syndrome.
The player can control the velocity of the stroke and the angle of attack. Since the stroke is an acceleration/contact/deceleration, the fingers, wrist, and forearm contribute in varying degrees to the velocity. The angle of attack is controlled by how high or low the player's hands are when the mallet head makes contact with the bar. The only other thing the player can do that affects volume and timbre is to control the particular location on the bar that the mallet head contacts. [Suddenly, all bets are off- so far I have only discussed sound production relating to the primary target- the "sweet spot." There are radically different sound envelopes available as the targets include other playing areas on the bar. See the article: Just What Do You Mean, Legato]. So, in the production of timbre, it is irrelevant whether the stroke is generated by the fingers, wrist, or forearm, as these motions determine specific stick velocity which affects the volume. Velocity can contribute to timbre only by accessing the various materials that make up the mallet head: the yarn, the latex tubing, or the core. But the change in timbre will be in direct proportion to the change in volume. Sustain time is in direct proportion to attack level or volume.
What does all this have to do with specific technique? I'll have to agree with Leigh Howard Stevens: the bar only knows how hard it was struck and where it was struck. Notice that in the proceeding experiments, we are dealing with isolated sounds produced one at a time. It is the player's ability to reproduce consistent strokes in a specific context that affects the overall sound. Again from the Method of Movement: predictable and repeatable strokes...
Hand height at the time of contact affects the ability to execute the upstroke as does the relative tension in the fingers and in the wrist. The mass of the mallet head and its durometer (hardness) have at least as much affect on the tone as the stroke does.
Copyright © 2002 Tyler-Rounds Productions