Some Aulos Techniques

Thoughts on Overblowing

Thoughts on Microtones



Posted in Uncategorized | 1 Comment

Further thoughts on reed making

This tool makes short work of cutting tubes in batches, all the same length. The end stop is adjustable because the length of reed varies from aulos to aulos.

Wolfcraft ‘megaCut S’ precision cutter with an adjustable clamp added by Barnaby Brown after the example of Julian Goodacre.

Wolfcraft ‘megaCut S’ precision cutter customised by Barnaby Brown after the example of Julian Goodacre.

After cutting the tubes, I now use a sharp scalpel blade to remove the bark, first slicing consistently with the grain, then scraping lightly to remove small ridges. This achieves a smoother finish and you can be more precise than with sandpaper.

Scalpel action removing the bark and golden fibre layer.

Scalpel action removing the bark and golden fibre layer.

To constrict the waist, I use bicycle gear cable with a wire rope grip at each end.

Apparatus for constricting the waist.

Forming the waist, I use bicycle gear cable with wire rope grips at each end to form loops. These are FIRMLY fixed to something at waist height, beside a pan of boiling water.

I rotate the tube back and forth, turning it with my fingers while increasing the pressure by leaning back.

Leaning back with hands free to rotate tube.

With the gear cable fixed to my belt buckle, I rotate the hot, wet tube back and forth between my fingers and thumbs while leaning back to tighten the cable. For its hardness and lower friction, it helps to have the shiny protective bark under the cable. The tube must be freshly boiled for the waist to form easily and evenly.

I don’t use a metal rod any more: it gets in the way and is more of a liability than an aid. For the batch that Barnaby took to Berlin in March, I tightened the waist until it was completely closed. The waist opens up again a little on releasing the cable and binding with waxed thread. This way, I can control the final waist size more precisely.

With some of the reeds in this batch, the internal fibres frayed, lifting up on either side of the waist. In Berlin, finishing reeds for the Pompeii aulos, Stefan Hagel removed these loose fibres and advised me not to constrict the waist so far. Cleaning internally above the waist is a difficult operation! After this Skype conversation with Stefan, I cleaned out  loose fibres on other reeds and found this greatly improved the sound.

A finished waist interior (above); removing the bark and golden fibre layer (below).

A finished waist interior (above). Removing the bark and golden fibre layer (below).

If the waist is too tight, the blowing pressure is too high; if the waist is too open, the air flow is too high – it is about achieving low airflow without having to burst blood vessels. What we want is equilibrium between the thickness of the blades and the constriction of the waist, so I tried adjusting the inside diameter of the waist using beeswax on a heated needle. I ran the needle along beeswax until a drip formed, then poked it inside the waist and ran it around to build up wax, reducing its size.

A temporary bridle to help break in the reed.

A temporary bridle made of cane and waxed hemp.

Wrapping thread around as a bridle opens up the lips, making the reeds harder; when cane slips are inserted under the thread, however, pressure is applied at the centre of the blades which closes the lips, making the reeds easier. These bridles are highly adjustable – you can pinpoint where you want to apply pressure, and they don’t slip or gouge into the cane like wire, forming a rut.

If you go for the softer cane (the ‘deep pith’ layer), the reeds form more easily but you loose sound quality – the tone is duller. Closer to the bark, the cane is springier and harder to blow, at least initially. But I believe that perseverance will create reeds that last longer, sound better and are more reliable.

Previously, my success rate was about 50% (6 out of 12). I was heartened to discover that this was also the case in ancient times:

[Aulos cane] generally differs from other types of cane by a kind of inherent thriving condition, being fuller and more fleshy and overall of female aspect. It also has broader and brighter leaves, but a smaller plume than the others, some having none at all, which they call barren-cane. From these, some hold, the best reeds are produced, though few work out right during the process of manufacture.

Theophrastus, Hist. Plant. 4.11, trans. Stefan Hagel

Does this mean that using ‘barren’ cane (i.e. with no flower) reduced success rate but might be worthwhile to achieve a superior quality of reed?

Increasing the blade length on 8 March increased my success rate to 10 out of 12. Was this improvement a result of: 1) a longer blade length, 2) variation in the nature of the cane, or 3) me becoming a more proficient reed maker?

Batch formers

Batch formers made of cane.

These clamps are quick and easy to make, but perhaps not as good as split dowel rods, or half rounds, which press on a larger surface area of the blades.

Thanks to Barnaby Brown for the photos and for co-writing this instalment. See also:
Part 1  |  Part 2  |  Part 4 coming soon…

Posted in Reeds | Tagged , , | 4 Comments

The Seikilos Song on the Pompeii aulos

Here is the best-known ancient melody, played, for the first time since antiquity, both on a reproduction of an ancient instrument and in the original pitch (a = c. 480 Hz, so to say).

The finger span on the lower pipe (Pompeii Pipe 2) is absolutely cruel; but that’s evidently how they fingered that pipe in its lowest playing position.

Posted in Ancient scores, Learning to play, Pompeian tibiae | Tagged , , , | 1 Comment

Circular Breathing Tutorial 1

STEP 1. “Balloon Cheeks”

Breathe in and out normally through your nose. With your lips sealed and your cheeks fully puffed out, continue breathing in and out normally through your nose. Maintain an even pressure behind your cheeks while breathing normally. Everything should feel relaxed apart from the stretch around your inflated cheeks and the lip muscles, which stop air from escaping out of your mouth.

Use a mirror to check that your cheeks remain constantly puffed out. Experiment with how to increase the volume of air by relaxing muscles, expanding all the cavities in your mouth, filling up available areas in front of your teeth as well as to the sides. Seek to reduce effort to a minimum. Only a few muscles need to work; all the others can relax.

It feels like inflating a balloon; the bigger, the better. When you are used to breathing normally, with your cheeks puffed out and feeling constant pressure, almost to bursting, then you are ready for a variation.

Change the rhythm of your breathing to a short, sharp in-breath and a long, slow out-breath. Flare your nostrils for the in-breath, reducing resistance so that the air reaches the bottom of your lungs as swiftly and quietly as possible. Your cheeks remain puffed out throughout: the pressure behind them is completely independent of your breathing.

Practise both rhythms for a few days, finding ways to reduce the effort. How many muscles are working unnecessarily? Your neck? Shoulders? Switch off unnecessary tension, muscle by muscle, till only the ones you need are switched on. As Joseph MacDonald advised in about 1760, a piper wants to discover “a gracefulness of carryage in Feature and Attitude” and to be “as aggreeable to the Eye as Ear”.

STEP 2. “Raspberry Sound”

Begin with the second rhythm of step 1, but increase the air pressure behind your cheeks a little more. Now, simultaneously with each quick in-breath, make a short raspberry sound between your lips. To replace the air you expel when making the raspberry sound, you will need to discover how to top up the air behind your cheeks.

Do this topping up during the slow out-breath by using the back of your tongue like a pump. This pumping action transfers pockets air from the throat into the cheek cavity. There are two phases to the pumping action:

  1. A horse-shoe shaped seal is formed by the tip of your tongue, touching all around the hard palette immediately behind your teeth. The toungue is cup shaped, with the root of the tongue holding the position for “ng” against the soft palette while the tip is sealed all around the hard palette.
  2. As you raise the cup of your tongue towards the roof of your mouth, as if to make the sound “ee”, you break the seal at the tip of the tongue allowing air to pass towards your teeth, topping up the cheek cavity.

Think of pressure arrows pushing out your cheeks: those arrows need to stay equally strong throughout the breathing cycle. Don’t be disheartened if it takes you a few days to discover how to do the topping up. It may help to say “tongue yee tongue yee tongue yee” without breaking the sound to identify the muscles involved.

STEP 3. “Dummy Reeds”

Take a reed that consumes no air (when your finger seals its base) or a dummy object of similar shape and size. This is to train your lip muscles to form an airtight seal around an object with the same feel as the real thing. However, it should not allow any air to escape.

Put this silent dummy reed between your lips and repeat steps 1 and 2. Don’t allow the presence of the reed to change anything. It should neither sound nor consume any air.

When you are comfortable repeating steps 1 and 2 with a single dummy reed between your lips, introduce a second one.

Steps 1-3 may take several days. New muscles are being trained and new neural pathways established. It is best to practise a little every day as the deep learning happens while you are asleep.

STEP 4. “Blowing Bubbles”

Now for the fun bit! Find a straw that you can squeeze shut and fill a glass not more than a third full. The straw can be plastic or organic, but you must be able to adjust the flow rate by squeezing it between your fingers. You should be able to shut off the bubbles completely: practise blowing through the straw into the water but with no bubbles!

When you can do that, repeat steps 1 and 2 with the straw squeezed tight, no bubbles.

Next, very gradually allow a tiny stream of bubbles to appear. Keep steps 1 and 2 going. Your job is to make that tiny stream of bubbles perfectly steady.

If you can do that, then you are over the crest of the hill. There’s a long journey ahead, but it is downhill from here. It took me a couple of years before I could circular breathe on autopilot, seamlessly, without it disrupting my concentration while playing.

Good luck!

The video was filmed by Peter Holmes in Olga Sutkowska’s office, Berlin, 26 March 2015. Please leave a comment sharing your experience learning to circular breathe. What helped you most?

Posted in Learning to play | 3 Comments

Some sounds from Pompeii

After a day in the workshop, adjusting the ring friction of the Pompeii 2+3 aulos so gloriously reproduced by Peter Holmes, Neil ‘Spike’ Melton and Martin Sims, here are some first experiments, using reeds by Callum Armstrong, scraped down to playability. The mechanism is wonderfully airtight!


And with the sýrinx pulled up (or is it down?):

Posted in Pompeian tibiae | Leave a comment

First working reproduction of a tibia from Pompeii

This tibia has been reproduced from one of several pairs of pipes buried at Pompeii in 79 AD. It was made by Peter Holmes with the help of Martin Sims and Neil ‘Spike’ Melton in a collaboration between the European Music Archaeology Project and Middlesex University. Stefan Hagel measured the original (which survives complete) and is directing this reproduction project.

Three cheers to Peter, Martin and Spike for making an instrument that really works! The precision engineering is spectacular. Its first sounds were heard in Berlin on 23 March 2015:

Stefan used reeds made by Callum Armstrong. Our hostess was Olga Sutkowska and I was behind the camera. We are grateful to the Deutsches Archäologisches Institut for bringing Peter, Stefan, Olga and me together on the eve of a meeting on the future of the International Study Group on Music Archaeology.

Posted in Pompeian tibiae | Tagged , , , , | 2 Comments

Lecture on the Roman tibiae on-line

I am pleased to share with you my lecture on the Roman tibiae which was given in September last year in the frames of the Short Summer School “Music Archaeology”  organised by Berliner Antike-Kolleg in cooperation with the EMAP – European Music Archaeology Project. The lecture was part of a block devoted to music in ancient Rome which I had a pleasure to share with Jutta Günther.

You will find here plenty of iconographic examples of auloi/tibiae and several archaeological evidences for the pipes with metal keywork which enabled modulations during the performance and, as mentioned by Stefan in his previous post, might be used in the Hellenistic and Roman theatre. I also show archaeological finds of tibiae with side-tubes which are subject of research for my on-going PhD project at the Berlin University of the Arts.

At the end of the presentation you will find one short musical example on the Louvre aulos (see 00:47 minute).

I’d like to express my warmest thanks to Adje Both and Jana Kubatzky for organization of the summer school, to my co-lecturer Jutta Günther for a very satisfying cooperation, and to those who made this video possible: Henrike Simon and Winny Henkel from Berliner Antike-Kolleg, and CeDiS of the Free University Berlin with Karsten Pruehl, whose enormous patience during the production process helped to achieve the high quality results.

Posted in Iconography, Louvre aulos | Tagged , | Leave a comment

Theatre auloi

Barnaby asks me about my thoughts on possible uses of the Louvre aulos in the theatre. Well, in my view the Louvre and the Berlin instruments belong to a different world – perhaps that of dinner parties.

Theatre music from the late fifth century BC on was characterised by modulation. So the texts imply, and some of the oldest scores show a highly modulating style, employing a wealth of chromatic notes. In the Roman period, theatre music always appears notated in proper ‘keys’ (tónoi), while I have argued that the type of pipes the Louvre and Berlin exemplify was not so, but employed transposing notation. Already the pseudo-Aristotelian Problems imply a number of different ‘keys’ being used for chorusses and actors. Nothing of that kind might be achieved on the Louvre aulos.

So I am pretty sure that the Hellenistic and Roman theatre always used pipes with metal keywork. But what about Classical theatre, especially before Euripides? Actually, we do know very little. Was the same set of pipes used thoughout a play, or a trilogy, or might the aulete have switched pipes? The Orestes papyrus, at least, arguably containing the earliest readable melody, uses but a single mode. However, if it’s natural signs are applied to the Louvre in the transposing manner, the melody becomes unplausibly high. Anyway, all the archaeological evidence for pipes of that date looks very different from the Louvre/Berlin, and even my speculative retrojection of the basic design, building on a passage of Aristotle and late constructions of the harmony of the spheres, takes us back no further than the fourth century BC.

Therefore I’m afraid I don’t think our wooden pairs would get us anywhere near ancient theatrical musical reality…

Posted in Louvre aulos | 1 Comment

More thoughts on reed making

This is the second instalment of Callum Armstrong’s reed-making journey — the first is here.  We made these notes and photos before a London Aulos Group meeting on 6 January 2015.

Why do reeds crack?

1. End unbound when constricting the waist. 2. Fibre band not removed. 3. Lips too thin.

This photo shows three different causes of cracking. Here are our suggested solutions:

  1. Before constricting the waist, tie a few winds of waxed thread around both ends (proximal and distal).
  2. Above the waist, remove the bark and golden fibre band below it to reveal the whiter pith.
  3. Leave an even thickness of pith up and down the blades, filing with a technique that does not leave a gradient. The tips should not be much thinner than the centre of the blades.

Callum made the following points:

I think there are two reasons for cracking at the lips: either too much of the fibre band is left on, or the lips are uneven in thickness and crack at the thinnest point. A couple that would have cracked I saved by scraping either side of where I thought it was going to crack, evening out the surface.

If you need to constrict the foot to fit the reed seat, then insert a rod to tighten around in order to prevent buckling like this:

buckled feet

The problem I have still to solve is air efficiency: they take too much air. After a while the reed settles into a position where squeezing is no longer required to make it playable. Until broken in, however, the reed wants to return to its pre-boiled shape. As soon as you wet the blades, they start to pop back. A ‘cold former’ is therefore essential for the first few days, perhaps weeks. The only alternative is a permanent bridle, but this contradicts the ancient evidence presented by Olga Sutkowska.

Hot former 1. Hot former 2. Cold former for breaking in the reed.

Wooden ‘hot former 1’.  |  Wooden ‘hot former 2’ to reach base.  |  Cane ‘cold former’ for breaking in the reed.

This photo shows two finished reeds and the formers I used to persuade the blades into their new shape. Hot former 2 has a shallower gradient than former 1 in order to reach the base of the blades. If the blades are going to crack, they crack in hot former 2. The cold former is for breaking the reed in over the first few days or weeks, with the reed drying out in the former after each playing.

I thought toasting might help the reed to keep its shape, but there must be knack which I have not yet discovered. I practised toasting several cracked reeds, but when I tried with a good one, it cracked in the usual place. These are much more delicate than duduk reeds!

The crescent shape of musette scrape didn’t work; a more even scrape, leaving thick blade tips, and using the cold former to flatten the base of the blade is what I find works. You loose tonal colour by scraping the tips and it doesn’t increase the air efficiency.

You can tie the waist, form the blades, and reduce the foot in direct succession, but you must then leave the reed to dry overnight before any further work.

Part 1  |  Part 3  |  Part 4 coming soon…

Posted in Reeds | Tagged , , | Leave a comment

My first aulos reeds

3 reeds made in Barnaby's kitchen, 17-18 December 2014

Barnaby Brown invited me to join the aulos revival when we performed together in London last month. These notes build on what Barnaby learned from Stefan Hagel at reed-making workshops held earlier this year in London and Tarquinia.

1. Cut and mark pairs of tubes (45mm for Louvre aulos, 40mm for Berlin aulos)

  • Barnaby fitted a Wolfcraft precision cutter with an adjustable end-stop and filed a semi-circular cradle to facilitate the neat cutting of identical-length tubes.
  • We used cane of diameter 11.5-12mm supplied by Medir.
  • Next time, we would mark the distal ends of each pair with a unique sign, located to record the position of the bud, just in case this turned out to be relevant. Use permanent ink on the end-grain because anything on the bark rubs off and ink on the proximal end can look like a split.

2. Remove bark and fibre band from proximal ends

  • Leave bark under the waist: only remove the cane’s natural outer protection from the end that will be flattened.
  • Carefully remove the bark and the golden substratum, just revealing the whiter layer of cane below. Cane is made up of three layers: a waxy bark, a thin fibre band, and the ‘pith’ (see

See Part 3 for photos

3. Thin proximal ends by filing inside

  • Let the cane tubes soak in water and file them wet.
  • Thin the whole part that will be squeezed above the waist, keeping the file in contact with the internal surface of the tube. If the blades become too thin at the tips, they will split when flattened: an even thickness of the pith (not the fibre layer) must remain to provide lateral strength and elasticity.
  • File as evenly as possible, laterally and longitudinally, frequently checking the wall thickness by eye. Observe the translucency by holding the tube up to a strong light.

4. Boil and form waists

  • Wind a few turns of waxed linen thread around both ends of each tube to prevent splitting.
  • Simmer the tubes for at least 2 minutes.
  • Prepare waxed linen thread with a loop made from two half hitches.
  • Remove one tube from the boiling water and immediately constrict its waist using bicycle gear cable tied between something solid and your belt. This allows you to have both hands free, which is vital (see Part 3 for photos).
  • We inserted a 4mm rod periodically to gauge progress, but next time might try constricting the waists to an internal diameter of 3.5 or 3mm. We found no advantage in keeping a metal rod inserted while constricting the waist; in fact, the rod was a hazard, increasing the likelihood of splitting.
  • Once the rod is tight, constrict the waist a little further before passing over the two half hitches and tying tight. This allows for slight enlargement of the waist when you slacken the cable.

5. Boil and form blades

I was taught by Francis Wood how to heat and squeeze a plastic drinking straw to form a reed (he makes a ‘disposable’ bagpipe). The result looks quite similar to the drawing of the lost Berlin aulos reed, so, instead of pushing the cane into a metal former, or clamping it in a screw press, I thought I would try squeezing the blades by hand, using the feel of my fingers.

I used four fingers (thumb and index of each hand) to find where the sides would come together most comfortably, rotating the cane while repeatedly squeezing and relaxing it. If any area seemed stiff and likely to cause a split, I thinned that area using a file.

Once I had decided on the orientation of the mouth, I brought the lips closer together by squeezing the corners between my fingers and thumbs: this avoids putting pressure on the central area.

When I felt the cane had cooled and was stiffening, I dropped the reed back into the boiling water. Each time it went back in, the mouth returned to a near-circular shape, but the next pressing was easier. I didn’t need to mark the corners of the mouth: they were relocated by feel and by eye. I repeated this two or three times, until the mouth closed sufficiently to fit comfortably inside my forming block.

Once partially in the forming block, I dropped it back into the boiling water in the block. Every half minute or so, I picked it out, removed it from block and massaged it with my fingers to encourage it to form a symmetrical aperture. I then pushed it a fraction further into the block and dropped it back into the water. If the lips were not coming together evenly, then I filed the relevant area from the inside to improve their symmetry.

On the second or third inspection, depending on how well the mouth was closing, I made two incisions from the corners of the mouth towards the waist, following the grain. These were not deep enough to cut through the cane, but served to encourage any splitting to occur  along the edges of the fold, rather than on the flat surfaces.

I found 5-6 little pushes were sufficient to close the lips. I then removed the block and continued squeezing the blades with my fingers.

Part 2  |  Part 3  |  Part 4 coming soon…

Posted in Reeds | 3 Comments