Klarinet Archive - Posting 000760.txt from 1999/04
From: Tony@-----.uk (Tony Pay)
Subj: [kl] "All that stuff about the diaphragm" (long)
Date: Fri, 16 Apr 1999 16:39:26 -0400
As a contribution to the discussion of the diaphragm and our control of
it that I see going on, I thought I would reproduce here an essay I
wrote several years ago. It was published in the Clarinet and Saxophone
magazine in Britain, and then later elsewhere a couple of times. The
central point is also put forward in my chapter in 'The Cambridge
Companion to the Clarinet'.
In essence, the essay is a description of a discovery I made for
myself -- in the process of giving a lesson, actually -- rather than a
statement of anything that anyone taught me. (Of course, this may
reflect my lack of understanding of what was said to me while I was a
student, but I think not.) And as I say, I haven't found the central
point put this way anywhere else in the literature. It will be
interesting to see whether anyone here has.
I've also found that the essay seems to provoke rather a strange
response among players. I'd expected people either to say, well, of
course everyone knows that -- or, to say that it isn't true, and why.
But what occurs instead is a sort of embarrassed silence....
With regard to what scientific content there is, I'd be interested to
have a professional evaluation. But for reasons that I try to make
clear during the article itself, I don't think that really matters all
Anyway, here it is. I haven't edited it, though perhaps it could do
with some editing.
ALL THAT STUFF ABOUT THE DIAPHRAGM
Books about how to play wind instruments often talk rather vaguely about
the diaphragm, and teachers of wind instruments seem to be agreed about
its importance in playing. However, in my view it is not so often
related to a player's actual experience, nor indeed explained to a
student's satisfaction. This essay suggests a reason why this is so,
and makes a connection between the action of the muscle and our
experience of playing, via the ideas of 'opposition' and 'support'.
If we fully understand the paradox that the action of the diaphragm is
in a sense outside our experience, yet nevertheless under our control,
our playng becomes simpler, and we are better able to trust ourselves.
CAN WE FEEL OUR DIAPHRAGM?
It seems that we don't have a direct experience of our diaphragm, the
muscle that we use whenever we take a breath. Anatomists tell us this
is because the diaphragm contains no efferent nerves. That is, no
sensory nerves run from the diaphragm itself to the brain, so we don't
know when we're using it except by noticing the things it does; and the
muscle can't feel tired to us, for example.
But we can get a bit closer to it if we flex our abdominal muscles and
try taking a breath whilst keeping them flexed. (By the expression
'flex the abdominal muscles' I mean the act of making the lower front of
the body hard, as if to protect ourselves against a gentle punch in the
middle ™ but without pulling our bellies in.)
Normally, when we flex our abdominal muscles, we prevent the air in our
lungs from escaping outwards by a blocking action somewhere in the
throat. In order to breathe in with our abdominal muscles flexed, we
have first to lift this block. When we do so, we may find (as my friend
and colleague Phillip Eastop recently pointed out to me) that the
experience of breathing in is very close to the experience of yawning.
We are breathing in against the resistance of our abdominal muscles, so
the diaphragm has to work a bit harder than usual in order to overcome
the resistance. It still isn't quite true to say that we can feel the
diaphragm working ™ we can feel working, and a bit of discomfort
perhaps, but this is mostly the sensation of the flexed abdominal
muscles and the solar plexus being pushed downwards and outwards; the
diaphragm, which is what is doing the pushing, is still not directly
accessible to our experience. Anyway, the anatomists tell us it can't
Most of the important actions of our bodies ™ important for survival,
that is ™ occur by themselves without our conscious control, so perhaps
it shouldn't be surprising that our diaphragm can't be felt directly.
We can't feel the working of our heart directly either, and rather
fortunately it beats without requiring us to remind it to. Choosing an
even more extreme example, we would obviously have absolutely no chance
at all of carrying out the incredibly intricate processes of our
biochemistry with our conscious intellects, even if we understood these
processes and could by some magic interfere directly. There's no
advantage for us to be able to feel ourselves consciously doing all the
things that get done in our bodies, so evolution just hasn't set it up
LEARNING WITHOUT 'KNOWING HOW'
Of course it is still an open question how much we can, by various sorts
of self-training, come to be able to influence these actions. Experiment
shows that, surprisingly, we can learn to control the speed of our
heartbeat, and even the surface temperature and electrical conductivity
of our skin; some would say that even our biochemistry is not as
inaccessible as it would appear.
This sort of learning, however, doesn't conform to our normal model of
what it's like to learn to do things. Mostly, when we learn a
technique, we have a direct experience of doing something. But there is
no direct experience associated with a change in the electrical
conductivity of our skin, for example. To be aware of such changes we
have to connect ourselves to a meter, and then we can learn to control
them and see the effect of our learning ™ the needle on the dial moves.
But we can't feel any change in our body.
I think most people feel, or would feel, uncomfortable learning in this
way. Somehow it seems strange, almost creepy. But it is important for
us to have a frame of mind which will accept this sort of strangeness.
An insistence on knowing exactly how we are doing something can
interfere with our learning to do it better. Quite independently of
whether we have an actual experience of performing a task, it is often
best to proceed as though our conscious minds were independent observers
of our actions, and to let those actions occur according to their own
logic. This is especially true if the task is a complex one. As
musical performers, we are frequently concerned with preventing our
minds from interfering with our abilities.
Breath control doesn't quite come into this category, you might think.
We simply have to learn how to blow the instrument effectively. Although
we use our diaphragm to breathe in, we don't need to experience it, and
anyway we only know about it in the first place because we've been told
about it, so ™ why not just forget it? Indeed many excellent teachers
and performers you speak to would say just that. And so would I have
done, until recently.
However, I then discovered a more useful metaphor to apply to breathing
and blowing the instrument; which metaphor is what I want to explain
In this essay, I shall say that we are 'using a metaphor' whenever we
give a picture or description of what we do on the instrument that is
not technically specific ™ in other words, a description that does not
tell us in detail what we should do. A metaphor says, rather: while
playing the instrument at this point, think of the situation as if it
were like something else. For example, I was taught as a very young
player to imagine the sound of the instrument as a smooth, solid tube
that began deep inside me, passed through the clarinet and stretched out
into the room. I still think this is a very good image for a beginning
Good teachers have always been aware that the transmission of subtle
skills involves the creation of suitable metaphors for a student, even
if only as an interim device. These metaphors work better than explicit
instruction. In fact, fully explicit instruction is actually
impossible, because even the best teacher cannot say in detail exactly
what he or she does. Indeed, usually the teacher too is operating out of
a personal metaphor. It is a question of leading the student into a
successful experience by describing the situation in such a way as to
help with a particular difficulty.
So metaphors constitute an indispensable tool for a teacher. They vary
in generality, from those designed to address the student's attitude to
her whole being with the instrument, to those concerned with the
character of a particular phrase or note. In fact, given the great
complexity and multi-levelled nature of what we do when we play, you
would have to say that even the most careful scientific description is
metaphorical too, being necessarily a simplified model of the situation.
We may not always notice that we simplify in this way. (Sometimes it is
surprising that someone else fails to understand us ™ after all, we find
the matter perfectly clear.) On the other hand, adopting the
metaphorical attitude allows us, if we wish, deliberately to go against
scientific description. It may be a better tactic to allow a student to
discover the physical solution to his problem whilst trying to do
something which is literally impossible.
A good example is what is called in brass teaching 'the no-pressure
embouchure system', which is a very helpful metaphor. This is so
despite the fact that the magazine New Scientist was able to show a few
years ago, by careful experiment using strain guages, that no
professional actually played using this system, even though some of them
said they did. The fact is that you play better by thinking of what you
are doing as not pressing the mouthpiece against your lips than you do
otherwise. This is true whether or not you actually are pressing
slightly, as shown by a strain guage attached to the instrument. Though
the 'no-pressure system' is in one sense a lie, it is what we might call
a useful lie; one that is worth telling oneself when playing, and worth
transmitting to students.
Another example would be my 'smooth tube of sound'. This tube of sound
doesn't really come from deep inside me, even though I may be better off
imagining that it does.
Yet clearly, when we use a metaphor, there is a danger of saying
something that is both wrong and not useful.
'PLAYING FROM THE DIAPHRAGM'
In fact, I believe that there are some people who talk about the
diaphragm in their teaching both inaccurately and unhelpfully. The
central fact that must not be obscured is that the diaphragm is a muscle
that can only exert force downwards, i.e. to draw air into the lungs.
As a passive membrane dividing the abdomen from the thoracic cavity it
is pushed up by the abdominal muscles in the usual action of blowing an
instrument (as contrasted with the universally condemned method of
pulling down the previously raised ribcage), and perhaps this is what is
meant when many people speak of 'playing from the diaphragm'. But this
is not the same as using your diaphragm as a muscle in order to blow,
which is a physical impossibility.
What I want to bring out in this essay is one aspect of breathing and
blowing which does have surprising and unusual experiential
characteristics when we compare it to most of the rest of our actions as
we play an instrument. The technique itself is mentioned in wind
instrument methods, and certainly employed by able players, who
communicate it with varying success in their teaching, but it is rarely
discussed in such a way as to make the situation usefully clear, at any
rate to me; and it hasn't been approached exactly from the position I
propose to take, as far as I know. Moreover I think the 'surprise'
connected with it has been almost universally overlooked.
I always liked, and often repeated to students, Paul Harvey's advice to
'keep your trousers up' when playing. He told the story of how he
forgot his belt (or was it braces?) one day when he had to play a
concert, and found that his performance was improved as a side effect of
his effort to make his abdomen as large as possible. I've also found
this a useful metaphor in practice. But there is another variable we
can control whilst we have our attention on our abdomen, which is to
what degree the abdominal muscles are flexed as we are blowing.
When muscles are working against very little resistance, we don't
experience tension in them. They work smoothly, and the resultant
motions look relaxed and fluid. But if I flex the opposing muscles in
my arm, both sets of muscles become hard, and my biceps stand out if the
effort in both is sufficiently large. If my arm is motionless, I can
tell without looking that my triceps are flexed by the fact that my
biceps are flexed in this way. (The triceps are the ones on the back of
the upper arm.) So it is with the opposition abdomen/diaphragm. I know
(if my airway is open) that if no air is going in or out, and my
abdominal muscles are flexed, that my diaphragm is also flexed.
(Indeed, this is the only way I can know it! After all, it's inside, so
I can't see it, and I can't feel it directly, as we've said.)
When we want to make precise and controlled movements, we do generally
use opposing sets of muscles, and here a danger is that we will use too
much force in both sets, since the output is unaffected by an equal
increase in tension in both. But this danger is well understood. We
may use metaphors in our teaching to avoid this danger: 'Float the
sound...imagine that your arms and instrument are balloons filled with a
light gas ...' etc. etc.
Precise and controlled movements are of course central to any art, just
as much as freely expressive ones. A perfectly executed trill, a fine
piece of handwriting, an elegant pirouette ™ all these require the
delicate balance of pairs of opposing forces, each supporting its
counterpart, under the overall orchestration of a guiding intelligence
and expressive impulse. And the word 'support' is crucial, both for
precision of adjustment and speed of response. It's why we push our
hand against a surface when we write on it (or use a special
signwriter's support stick), and why we poise ourselves ready to spring
when we wait for the serve at tennis.
How does all this apply to the diaphragm?
Normally, when we breathe in, the diaphragm encounters no opposition
other than the elasticity of the viscera. This elasticity causes the
diaphragm to return to its normal position when it relaxes. Similarly,
when we blow out air using the abdominal muscles, the diaphragm is
relaxed, and so the only opposition is the inertia of the lungs and the
outgoing air. Opposition occurs when the diaphragm and abdominal
muscles are both working at the same time.
The characteristic of muscles balanced in opposition is that both are
slightly flexed. So, if you perform the experiment of breathing in with
your abdominal muscles flexed (not too much), you will find that you can
achieve a point of equilibrium where the air isn't moving. And then,
putting your clarinet in your mouth, you can play a note, still with
your abdominal muscles flexed. (If you prefer, and indeed this is the
more usual way of proceeding, you need not breathe in with your
abdominal muscles flexed ™ you can flex them afterwards. As we said
before, being aware of our abdominal muscles working is the only way we
can know that our diaphragm is also working.) Now, you will find you
can perform a crescendo, and a diminuendo, still with your abdominal
muscles flexed to the same degree.
SUPPORT, AND THE 'MAGIC DIMINUENDO'
So what? you might ask.
Here's the point I never noticed, and which I now find makes such a
difference, not to begin with to what I do, but to how I imagine what I
do (i.e. to my personal metaphor), and therefore, in the end, to almost
everything. It is that the crescendo, and perhaps even more clearly,
the diminuendo, can occur in this situation without anything else at all
happening in your experience. You imagine a diminuendo ™ hey presto, a
diminuendo. You want a faster diminuendo? ™ no problem.
I don't just mean that the process of doing it has been submerged, in
the same way that the actions of driving a car, say, in the end become
automatic. In this case you can call up the experience into
consciousness by paying attention, even though you weren't aware of it
before. No, I mean that you can't call up any physical experience
corresponding to the change in dynamic. Everything stays the same.
Perhaps you can feel a slightly different movement of air in the mouth
as the dynamic changes, or a different embouchure. But nothing in the
You don't do anything ™ you just imagine it. The only change is in the
sound. You shouldn't take my word for this ™ you have to try it
yourself. Play about with it for a bit. Convince yourself that you
really are keeping everything else the same in your experience.
Remember the 'no-pressure system' ™ well, this is the 'no-doing system'!
The strangeness of all this is rather like the strangeness of the
experiment with the electrical conductivity of your skin. There, you
couldn't be sure what you were doing to produce the required result. In
this case, equally, you're no longer listening to whether it's coming
out the way you think you did it. Now it seems you're not doing
anything; once you've set it up, there isn't anything else to do but
listen to the result.
In other words, to imagine it (and listen to it), is to do it.
The explanation for this, of course, is that the diaphragm is resisting
the abdominal muscles (which remain at constant tension) to a varying
degree. But that is inaccessible to experience. So our only feedback
is to listen to the result, and thus we establish a direct link with our
How did I (we?) miss this? I suppose, like most things, by not paying
close enough attention at the crucial moment. Also, it's very much not
what you'd expect, and you have to be very careful to hold everything
constant to appreciate it. But, as I shall spell out in a bit more
detail later, it explains lots, like how passagework becomes even by
itself, if we listen to it, and why we can play fast dotted rhythms
seemingly without effort if we support ™ and here's the magic word!
Have you been confused, like me, by the way different people use this
term? Doesn't it help to know it means the exact opposite of blowing?
or rather that it's an opposition or complement to blowing ™ part of a
magic technique which works by your setting the only variable you leave
available to yourself (the flexion in your abdomen), at one strength and
then allowing the result to change according to your whim? ™ a sort of
black box that you can't fiddle with, only use? ™ dealer only service?
Isn't that wonderful? (Doesn't it make you want to sing and shout?)
You can see that looking at it this way is a reversal of how support is
normally envisaged. Normally, support is what stays constant while
action varies. If we assign the role of support to the diaphragm, then
here it's the support that varies, whilst the action stays constant. We
could of course say that you support with your abdomen and act with your
diaphragm outside your experience, in the opposite direction, and people
who are really clear about all this already (perhaps there are lots, I
don't know) may speak about it that way round.
At any rate, the situation is strange enough to be worthy of more
informed discussion than it gets.
Perhaps we can tie in here a well-known phenomenon on voice and flute,
looked at from this slightly different perspective. Most singers and
flute players (though few clarinettists) have what is called a
'diaphragm vibrato', which they acquire and then refine, many of them
without really knowing how they do it. Vibrato is spoken of in much the
same way as I have been speaking of support ™ as a rather mysterious and
even magical part of playing which seems very closely connected with the
innermost being of the performer. Teaching it seems to be largely a
question of setting up circumstances in which the student will ”catch
on• physically to the idea. It seems plausible to conjecture that this
is because the mechanism is a periodic variation of diaphragm flexion
outside awareness, and I find it possible to imagine that this sort of
vibrato occurs most naturally in circumstances where the opposition
abdomen/diaphragm is relatively small. (Intuitively, when our movements
are larger, or faster and more free, we want to be as relaxed as
possible for best effect. The forces 'tied up' in the oppositions
simply generate heat and tire us.) Clarinet players have a slower
airstream, need more precise control, and so tend to play with stronger
support than flautists or singers; my guess is that this minimises the
chance that a diaphragm oscillation will arise and be developed as an
expressive device. Notice that diaphragm vibrato is not unknown on the
more free-blowing saxophone, though of course stylistic considerations
enter considerably here. Clarinet players who want to use vibrato
usually employ other means to achieve it.
THE PROS AND CONS OF USING SUPPORT
When your abdominal muscles are flexed more than is required simply to
play at the dynamic you are delivering, you're using diaphragm support.
Your diaphragm is resisting your blowing, but you have the advantage of
very precise control over dynamics. However, you still have to judge
whether the effect is what the music requires. The rather 'careful'
quality of such dynamic control has a way of spilling over into other
aspects of one's playing, and this can need guarding against. I am
thinking in particular of the sort of restrictions we can make in the
air column, limiting the resonance of our playing by, for example,
closing the throat. It is easier to do this by mistake if we are
already committed to the diaphragm/abdomen opposition. So it's worth
while practising keeping the air column as open as possible with maximum
support, rather as we sometimes practise playing fortissimo with a most
delicate finger action, and vice-versa.
But it has to be said that a very valid musical effect can be obtained
by precisely controlled resistance all along the line. Debussy's doux
et pénétrant in the Rhapsodie, for example, can be played in this way.
And I've always felt that it isn't enough to play the solo in
Tschaikovsky's Pathétique merely beautifully. It must represent the
loneliness that comes from expression through reluctance to express,
which reluctance has also to find expression.
Of course, you don't have to play with support. Often, playing without
it has a light quality in low dynamics, suitable for short, floaty
phrases, and a grand, gestural quality when loud. All the other
variables of tone-colour, resonance are still available. The
appropriateness in the context of the music is always what counts.
MORE ON WHY IT WORKS
I mentioned before that tennis players use opposition when they are
waiting for the serve. Another way of describing what they are doing is
to say that they are storing energy so that it can be delivered fast,
and in the required direction, immediately they find out what that
direction is. It is as though they are springing both to the left and
to the right at the moment the other player serves, but because the
muscles that would drive them to the left exactly balance the muscles
that would drive them to the right, there is no overall effect. When it
turns out that the serve goes to the right, they simultaneously relax
the muscles driving them to the left, and begin to work harder with the
opposing set. But they have a flying start, because of the initial
working of the muscles driving them to the right.
A bow-and-arrow is an energy storage system. We do all the work of
bending the bow before we shoot, storing the energy that will be
released over a much shorter interval in order to throw the arrow far
faster and farther than we could by hand alone.
There is a useful analogy between the bent bow, which embodies a bow/arm
opposition, and playing with support, which embodies an
In this analogy, the abdominal muscles correspond to the bow, and the
diaphragm to the arm. The sudden delivery of energy common to both
would in the case of playing a wind instrument be what is required
either for a sudden change of dynamic between an adjacent pair of notes
(a sforzando or subito piano); or for a precisely controlled change of
air pressure to equalise the dynamic of an adjacent pair of notes with
different responses on the instrument. Support enables us to do both of
these things easily and elegantly, and moreover without knowing
precisely how, so that it seems an automatic and natural ability.
Support is also useful for taking a fast, inconspicuous breath. In the
bow-and-arrow analogy this would be like letting go the bow rather than
the arrow, which wouldn't be very useful ™ but clearly relaxing the
abdominal muscles with the diaphragm already strongly flexed would
result in a maximum delivery of energy to draw in air over a short
timespan, which is precisely what we want. And so it proves: to play
with support just before taking a breath guarantees both maximum
air-intake and precision of return to the playing position. How? We
simply bring our abdominal muscles back to the state of flexion they
were in just before the breath, and continue with the phrase.
The usefulness of this little discovery for me is that I find I'm now
much more able to accept and trust as rational what I often did before
instinctively, and to simplify my actions so that they have more chance
of success. The support mechanism can be calibrated at the beginning of
a difficulty (translated: you decide how hard your abdomen should be)
and the calibration then changed until the setting that produces the
best effect is reached. After practising in this way I find I often
need to do, and compute, less than I'd thought. When teaching, it's
still difficult to stop people sticking to one way of playing which
isn't working, and now some of them think you should flex your abdomen
all the time, but ™ 'twas ever thus.
A few more things to try: what we mostly did already for an upward leap
™ support on the low note ™ then, imagine the upper note as clearly as
possible, but concentrate on keeping the support constant.
Before a 'difficult' entry: breathe in against opposition, rather like
yawning, and time the top of the yawn to coincide with the moment of
entry. As well as guaranteeing precise control, this tactic gives you
something to think about other than the thought that you may miss the
In medium speed articulation (like Beethoven's Fifth Symphony: repeated
pianissimo 'A's in the clarinet register) ™ where it's often difficult
to guarantee an even response ™ support, and then ask your diaphragm to
help! The quality of this 'request' is important. I don't mean you
actually do anything, in fact, quite the opposite, as I've explained.
It's more like letting go of the worry about having it be even,
realising that there's an inaccessible mechanism at work which may know
or learn what's necessary. If you like, imagine writing your request on
a small piece of paper and swallowing it! You have to realise that you,
consciously, have really no control over it. I think this realisation
is most powerful. The idea of giving in to a wiser self has often been
held out as the key to mastery in all sorts of traditions. The
relinquishing of the idea of control is often all that's needed in this
I hope this little essay may stimulate some people to make discoveries
for themselves. As I said before, I think the subject hasn't been well
served. I've outlined one particular metaphor for playing that I find
useful, but anyone who wants to extend that metaphor is welcome to do
so, with the proviso that it should be a helpful extension. There are
many connections to be made with other aspects of playing, and the
possibility of technical detachment going hand in hand with expressive
involvement is an ongoing project for all of us.
The creation of metaphors is arguably what we are about in all aspects
of music, but it's important to bear in mind that whilst these must have
some coherence, they are above all personal.
© Antony Pay 1996
An earlier version of this essay appeared in Clarinet and Saxophone, the
magazine of CASS, the Clarinet and Saxophone Society, in 1993.
_________ Tony Pay
|ony:-) 79 Southmoor Rd Tony@-----.uk
| |ay Oxford OX2 6RE GMN family artist: www.gmn.com
tel/fax 01865 553339
Unsubscribe from Klarinet, e-mail: klarinet-unsubscribe@-----.org
Subscribe to the Digest: klarinet-digest-subscribe@-----.org
Additional commands: klarinet-help@-----.org
Other problems: klarinet-owner@-----.org