Larry Krantz Flute Pages: Mike MacMahon Flute Corner
Mike MacMahon

Mike is Professor of Phonetics at Glasgow University; he is also a Council member of the International Phonetic Association. He plays flute and piccolo in the Glasgow Symphony Orchestra.

Tongues, Gums, Teeth and that letter T
This article was published in 'Pan', The British Flute Society Journal, Summer 1996,and is reproduced here with permission.
Throat resonance, vowel sounds
Flutter Tongue
How to Pronounce BOEHM
Head Tilting
Head-voice "sound" while playing?

Tongues, Gums, Teeth, and That Letter T
Mike MacMahon
The best advice I was ever given about learning to play a flute was to 'breathe warm air' into it. This seemed such a natural thing to do. Then came the instruction that I had to 'say the letter T', or a 'TEE syllable', and 'attack' the note, and watch my 'articulation'. Was this flute-playing, I wondered, or was it something to do with talking into a flute or using it like some weapon of war? Now, years later, I want to offer some phonetic comments on what flautists do with their tongues when they 'articulate', 'attack' etc the note. Even though I still prefer to think of 'breathing' a flute, I have to confess that I do wonder if the way in which French people speak French gives them an advantage when they play a flute, and if, in consequence, some awareness of how a French [t] differs from an English [t] may actually help English-speaking beginners to get a good articulation early on. I emphasise beginners, because once a flautist becomes reasonably competent, he or she quickly learns, often instinctively, to use a variety of tongue strokes.

All instruction books attempt to shed light on what a beginner should do (and what an advanced student playing a Baroque flute, for example, should do when playing in the 18th-century manner.) The literature is full of comments about 'the letter T' (or 'the letter D'), or avoiding the 'letter L' (when double-tonguing), or flicking the tongue for the 'ra' syllable when practising the 'ta-ra' sequences in the Taffanel-Gaubert exercises, and so on. Essentially, descriptions like this may not always be as helpful as they seem. There is, I believe, a more useful way of explaining what to do, which will give the player greater insight into what the tongue should be doing and encourage easier and more confident playing right from the start.

Do instructions to follow the orthography ('Say a T sound') help?

Instructions like 'Say the T in TOO or TEA' beg various questions. For example, are the [t]s the same? The answer is no. The [t] in TOO has, for most speakers of English around the world, some protrusion of the lips (caused by the influence of the following vowel sound), whereas the [t] in TEA doesn't. Look in a mirror and watch your lips as you say TOO and TEA.

Other subtleties to note are, firstly, that when we say words like TEN, TRY and WIDTH, usually slightly different places along the top of the mouth are touched by the tongue. The [t] in TRY could well be slightly further back than the [t] in TEN; and the [t] (it may sound like a [d]) in WIDTH is very probably made on the back of the front teeth. Secondly, if you say the words TEA and TAR, you'll probably feel that the middle part of the tongue is higher up in the mouth for the [t] in TEA than for the [t] in TAR. This has to do with the tongue preparing for the following vowel sound. For the 'ea' vowel in TEA, the tongue has to be quite high up in the mouth.

English and French [t]s: some of the differences

The sort of lip-protrusion we use in TOO is not the same as the lip-protrusion needed for playing the flute - as we all know. Much more helpful is to say 'imitate the French [t] in TU' - but make sure it's a native French speaker's pronunciation, not an English imitation! (French TU uses a different style of lip-protrusion from French TOUS.) If possible, get a native speaker of French to say the words TU and TOUS as you watch their lips.

There are other differences. The English [t] is usually made with the tongue touching the area of gum behind the front teeth (not the teeth themselves), whereas for many French speakers it's the front teeth that the tongue touches.

Many speakers of English - it doesn't matter which variety it is around the world - use the blade, not the tip, of the tongue for [t] and related sounds. (The blade is the half-inch or so behind the tip). French speakers divide themselves into two categories: those who use the blade (on the front teeth) and those who use the tip (on the front teeth).

Something else that I think is relevant is the way that the French 'set' the tongue when they speak; it's quite different from what's used in most varieties of English. For French, the whole body of the tongue is positioned further forward in the mouth: it's as if it swings from an anchor-point behind the lower front teeth; in English, the anchor-point is much further back: the sides of the tongue touch the upper side teeth and the tongue 'works' from that position. This has big implications for how we tell people to start tonguing notes on a flute. The popular view that the English don't open their mouths when they talk, but the French do, contains a lot of wisdom. Again, watch native speakers of French when they talk: their tongues are very visible. Now compare them with English tongues.

Summing up so far: the French [t] uses the tip (or blade) of the tongue on the back of the upper teeth, whereas in English the [t] usually has the blade on the area of gum behind the upper teeth.

What else happens in an English [t], compared with a French [t]?

The way we finish off a [t] sound is important. If you say the words TEN and DEN, you're likely to find that for the [t] of TEN the finish is quite noisy compared with the [d] of DEN. (Technically, there's 'aspiration' on the [t] of TEN, but not on the [d] of DEN. What this means is that for a brief period of time between the [t] being released and the beginning of the vowel sound, the lungs are continuing to pump air out of the mouth. This can be visualised as something like ThEN - ie TEN with an [h] after the [t]; this is not the same, however, as the 'th' (a single sound) in THEN. (It can sometimes help to think of aspiration as an extra 'puff of air' after the 'main' sound has been released.)

Alternatively, for some varieties of English, the finish to the [t] is noticeably sluggish, and the resulting sound is like [t] followed by a brief [s] sound: as if it were written TsEN. (Technically, the [t] is 'affricated', not aspirated.)

In French, on the other hand, the [t] is finished off much more cleanly, without aspiration or affrication. If possible, get a native French speaker to say the French words THÉ and TABLE and compare them with an English pronunciation of TAY and TAB. Allowing for differences in the vowel sounds, there are still other noticeable differences in the way the two languages pronounce the initial [t] sounds.

Speakers of Korean or Thai or Hindi (and certain other languages) consciously use lack of aspiration, as well as the presence of it, in ways that we don't in English - or French. It would be interesting to know what they recommend when tonguing a flute.

What do beginning flute players need, then: a quick neat sound or a slower, more drawn-out one? Presumably only the former. In which case, we shouldn't be teaching beginners to say an English [t], but something closer to a French [t] (but not totally identical to it - see below).

[t] or [d]?

If you say a word like STOP, you'll produce a [t] which is very close to the French one - compare it with TOP. However, the sound of the 's' can obscure, for many people, the precise sound of the [t] in words like STOP. The solution is to emphasise that the tonguing sound should be more like a [d] - I say 'more like' deliberately, since there is still one crucial difference. Try saying STOP as if it were SDOP, and you'll be producing something closer to the 'right' sort of sound, but it's still not correct.

The explanation lies in what is happening down in the larynx (the voice-box). In a word like DAN, the [d] is voiced: that is, the vocal folds (or vocal cords) in the larynx vibrate. What happens during voicing is that the vocal folds go close together and then vibrate. The result is less air gets between them for [d] than for a [t]; for [t] the vocal folds are kept wide apart all the time. But, the sound of the [d], with its characteristic buzzing quality from the larynx, is not what's wanted when playing the flute. (The occasional 'throat sound' one hears from flautists is the result of the vocal folds vibrating - when they shouldn't.) What we want is a more controllable jet of air but without the rapid movements of the vocal folds produced by voicing.

The solution is quite simple: whisper the [d] in DAN, or better still, the two [d]s in a word like DAD. What this does is to keep the folds quite close together - the transglottal area (ie the area of space between the vocal folds) is quite small - so that the air-flow is more controlled (and controllable when it reaches the upper teeth area) and, critically, the vocal folds do not vibrate. Now whisper DAD again, but with the tip of the tongue touching the upper teeth. This, I believe, is the quality of [t] (or should we not be calling it [d]?) that beginners need to aim for when articulating the flute.

Another relevant factor about a whispered [d], rather than a [t], is that the amount of aspiration (the 'puff of air') in [d] is minimal: a few thousandths of a second only or none at all. For a [t], by contrast, it can be well over a tenth of a second.

The rest of the tongue

The point about control of the air-flow as it reaches the closure between the tongue and the front part of the mouth raises another point: what to do with the rest of the tongue (especially the middle) in order to help channel (or adjust the character of) the air-flow as it moves forward towards the upper teeth and lips. As pointed out earlier, the [t]s in TEA and TAR are not the same: the first one has the middle part of the tongue humped close to the roof of the mouth, whereas for TAR the middle of the tongue is lower in the mouth. There are, of course, many other positions the tongue will go into when we speak, depending on the following vowel sound. Beginners need to experiment to find the best positioning for the large area of tongue behind the tip and blade when they are playing the flute. They could be asked to experiment by whispering words such as DEED, DEAD, DAD, DOD, DUD etc to discover which is the best sort of [d]-ish sound for them.


The features that have been identified, then, as relevant for tonguing are:

  • (a) The part of the tongue that makes contact with the roof of the mouth;
  • (b) The area of the roof of the mouth that the tongue touches;
  • (c) Producing the sound without aspiration or affrication;
  • (d) Whispering a [d], with its much smaller transglottal area, rather than trying to use a [t], which will always be accompanied by far more air.
  • (e) The position of the rest of the tongue.

So my recommendation would be to use what phoneticians calls a 'whispered apico-dental plosive with no aspiration or affrication' - in other words, a whispered [d] in DAD but with the typical French feature of touching the back of the upper teeth with the tip, not the blade, of the tongue. However, in loud passages, where a lot of air has to be released quickly, it is a [t], rather than a whispered [d], sound that's needed - because the transglottal area is much bigger, and so more air can be called up from the lungs and pressurized more quickly.

Of course, there are other factors that can influence the choice of articulation: eg the shape of the tongue, the angle of the front teeth, any gaps between the teeth, the width and vaulting of the roof of the mouth, whether or not there's a dental plate in the mouth, and the constraints imposed by a speaker's familiarity over many years with a particular style of English pronunciation. This latter point can involve the speaker in 'setting' his or her tongue to operate within a particular area of the mouth and having to break that pattern when playing the flute. Speakers from the North of England, for example, often have the body of the tongue set further back and higher up in the mouth than South of England speakers, thereby giving a rather 'darker' quality to their speech. Try talking with a so-called 'effeminate' or 'little girl's' voice quality and you'll discover that the body of the tongue has to move further forward as well as noticeably upward in the mouth. Again, the importance of adjusting one's normal 'articulatory setting' as it's called when one switches from speaking to playing the flute needs to be consciously recognised.

More advanced players learn, of course, to use a variety of tonguings when playing the flute. For the Taffanel-Gaubert 'ta-ra' sequence, for example, the 'ra' is very similar to the pronunciation of the 't' in words like ATOM and LATER by many American speakers. And for different auditory affects, the tip of the tongue can be shaped in various ways. Jazz flautists utilise an even wider range of articulations.

  • This article was published in Pan. The British Flute Society Journal, Summer 1996, and is reproduced here with permission.

  • Mike is a phonetician in the Dept of English Language at Glasgow University; he is also a Council member of the International Phonetic Association. He plays flute and piccolo in the Glasgow Symphony Orchestra. He can be contacted at; his URL is

Return to Top

Throat resonance, vowel sounds

At the risk of boring the pants off you all with yet more phonetics in relation to flute-playing, the following points may help to clarify various things that have been raised directly or indirectly by Alexa, Thomas, Lisa, John and others. Either hit the 'trash' key or read on!

1. When 'vowels' are mentioned, *don't* think in terms of A-E-I-O-U-(Y). We're dealing with vowel-SOUNDS. It's crucial to get away from the idea that vowels are just 5 or 6 written things and "we know all about them because we learned to read at school, and anyway it all happens in the mouth and throat - somehow." The difference is that vowel-sounds are, quite simply, sounds made with very little or no obstruction to the air-flow when we speak. Compare the vowel-sound in your pronunciation of AH with what happens when you say a 'p' or a 's' sound. (Try it! Don't just sit mesmerised at the screen!)

The A-E-I-O-U-(Y) vowel-letters (in HAT, LET, etc etc) will correspond with vowel-sounds, but altogether there are hundreds (yes, hundreds) of vowel-sounds that can be made. They all depend on the shape of the tongue and the lips.

Almost everything that's said (in the books at least) on the tongue when playing the flute seems to have to do with 't' and 'k' (or 'd' and 'g'). I've thought for a long time - and particularly when listening to Jimmy playing - that we've got to consciously bring the *whole* of the tongue into consideration when talking about making different sorts of sound on the flute; it shouldn't just be slipped in as an after-thought or apologetically. (A couple of weeks ago, someone mentioned that Jimmy had been talking about vowels and flute-playing at the NFA. Great!)

2. When you say any vowel-sound, you have to shape the middle and back of the tongue in a particular way. Hold a long AH and look in mirror. Now do the same for OO. The tongue will go back and up for OO, and the lips will change position. Now say EE. The lips will straighten out and the tongue will go well forward but keep quite high up. Now say AH, OO and EE again, but this time *silently*. This helps to feel where the tongue actually is. Say EE again, followed by the word EYE. In EE, the tongue is either stationary or nearly so. In EYE, it moves quite noticeably. Feel the difference and check it in a mirror. (In case you're wondering how we get from the three AH, OO, and EE sounds to the hundreds I mentioned earlier, phonetics has the equivalent of a longitude/latitude grid method for plotting the location of vowel-sounds in the mouth; it also has a whole load of funny-looking symbols to write them down with!)

By changing the position of the tongue-shape a very small amount - and we're talking *millimetres* here - a different sound can be produced. In flute-playing terms, this means that a slight adjustment of the tongue could affect the characteristics of the air as it flows through the mouth to the lips and then hits the chimney/riser.

3. Notice that when you go from EE to AH, the jaw goes down. If you try this on the flute, you'll hit problems. But it's possible to compensate, by manoeuvering the mouth into what is sometimes a non-speech shape. Ventriloquists do this sort of thing all the time, of course. (OK - that'll start another thread: do ventriloquists make good flute-players?)

4. So where does this 'throat resonance' thing come in? There's more to the tongue than the bit we see when we look in a mirror. The root of the tongue (as it's called) is out of sight and forms the front wall of the throat. But by changing the position of the root, we can change the quality of the vowel. The acoustic affect of shaping the *whole* of the tongue (including the root) and the lips in particular ways is to set up a series of resonances, in the throat and in the mouth. There can be up to 5 of them for a spoken vowel. It's these that I would argue can be used to create different sounds/tone-colours/nuances on the flute.

And it's quite easy to hear two of these resonances: WHISPER the vowels 'ee', 'ay' (as at the beginning of DAY), 'e' (as in LET), 'a' as in HAT, 'a' as in PSALM, 'o' as in HOT, 'aw' as in DAWN - by now the Americans, the English, the Scots will be arguing as to how many vowels they have! - 'o' as in HOME (but make it like a French or a German vowel; almost all speakers of English around the world have a non-stationary vowel here), 'oo' as in FOOD. You should hear the pitch of the vowel going progressively downwards. What you've identified is, broadly speaking, the resonance area between the tongue and the roof of the mouth; we might as well call it the 'mouth-pitch'. In technical terms, you've identified the second formant (F2) of the vowel. These pitches are usually always there when you say these vowels - except you can't hear them because of the general hubbub of noise coming from the voice-box.

The next bit is a slightly trickier, but only because you've got to put your beloved flute aside for a moment. Turn your head slightly from the vertical and flick a finger against the side of your voice-box. Put the finger-nail against your thumb first of all, so that you get a good, hard flick. This time, don't whisper the vowels: just MOUTHE 'ee', 'ay', 'e', 'a'... 'o', 'aw'...'oo' etc as you flick your finger against your throat. You'll hear a quite different series of pitches. The pitch for 'ee' is low down. It then rises up to the vowel for PSALM, and comes back down again. This is the throat-pitch (or throat-resonance). The flicking movement has caused the air in the throat to move about. Technically, it's the first formant (F1). Again, it's there when you say these vowels normally, but it's covered up by the sound from the voice-box.

The pitches can be worked out in musical terms and put on staves, or they can be analysed using computer programs. Either way, you finish up with F1 and F2 expressed in terms of Hz.

5. The first and second formants of vowels will vary depending on the shape and size of the vocal tract, and on the variety of English (or Dutch, or Italian, etc etc) you speak. Also, there are gender and age differences. Women tend to have shorter vocal tracts than men; children even more so. This means that their resonances will be a bit higher in terms of Hz than men's. So we can't generalize and say 'this is how you must set your tongue for this particular note'; our differing vocal-tract geometries have to be taken into account.

6. How all this ties in with flute-playing is as follows:

    (a) I reckon we have to become more conscious of what the rest of the tongue is doing when we're playing; it's not just a case of articulating a 't' (or a 'd') and forgetting about the other 95% of tongue. And we need to start feeling these subtlely different positions of the tongue and lips for vowel-sounds. Just saying them silently and introspecting is as good a way as any to become aware of what's going on inside the mouth.

    (b) To get a 'good' sound, we should be consciously, and eventually unconsciously, adjusting the tongue for the notes we're playing. I can do a wonderful, Galwayesque bottom G if I put my tongue in position for a vowel close to my normal 'oo' vowel in GOOD (but without the lip-rounding - because that would interfere with the embouchure). If I move the tongue forward to the 'i' sound of HIT, it sounds rubbishy. What's happening is that the basic pitch (the fundamental) of G (about 395 Hz) is 'triggering' the throat- and mouth-pitches (ie F1 and F2) of my 'oo' vowel. The result is a truly resonant sound. For a Galway-like imitation of A, I have to move the tongue a fraction further forward.

Remember the Moyse exercise for keeping the tone the same as you go up and down a scale? I reckon the key to it is making very small changes in the shape and position of the tongue - not just attention to the embouchure. Once you accept that there are hundreds of vowel-sounds and they differ from one another by often very small amounts, then the idea of resonance and flute-playing falls into place.

Well, at least this post hasn't involved you squirting water into your bath tubs! Just flicking a finger against your throat and going around for the next few minutes with a red mark on your neck perhaps!

Mike MacMahon

  • This article was posted to FLUTE and is reproduced here with permission.

Return to Top

by Mike MacMahon

There are two sorts: (a) using the tip of the tongue against the ridge behind the front teeth (sometimes the back of the teeth as well as or instead of); (b) using the uvula (the 'blob' of muscle hanging down at the back of the mouth) against the back of the tongue. Whichever way it's done, the critical thing is to make sure the beating movement is slight and loose and relaxed: tensing the tongue definitely won't work. In no cases - unless you're singing and fluting simultaneously - would the vocal folds (the things in your voice-box) be vibrating.

Spanish examples can be a bit misleading, unless one realises that a pair of words like PERO and PERRO use different sorts of 'r'. (In the first word, there's a single tap of the tip of the tongue against the ridge; in the second, there's a series of fast taps (a so-called trill or roll). For flutter-tonguing, it's the PERRO word that will work. The nearest you'll get to the front type of flutter-tonguing in an English word would be an exaggerated pronunciation of words like PRAISE or TRUST or CRUST, with a very definite trilling-movement of the 'r'. Think of someone trying to do an over-the-top imitation of a Scottish pronunciation of these words - you'll probably need to think of kilts, whisky, heather etc. etc to get in the right frame of mind for it!

The 'Oriental problem' with flutter-tonguing derives from the fact that what Anglophones think of as 'r' or 'l' (or even 'n') is produced differently in many of these languages. At the end of the day, these flautists should consciously try to make a nice, soft, gentle, relaxed trilling motion of the tip of the tongue against the ridge (or the uvula against the back of the tongue). Watching in a mirror for starters will help. For such players, the sound will still be a 'foreign' one to them - as it is to some speakers of English.

This article was posted to FLUTE and is reproduced here with permission.

Return to Top

How to pronounce BOEHM
by Mike MacMahon

For many speakers of English, the German word BOEHM can seem difficult to pronounce. For native German speakers, of course, it's very straightforward. It's similar to the vowel sound after the 'g' in GEHEN (to go), except that the lips are pushed forward. The word for 'flute' in German, FLÖTE, has the same sound as BOEHM, as does the word for 'Bohemia' (BÖHMEN). (BOEHM is sometimes spelled BÖHM.)

There's a similar sound to it in French -- the 'eu' in FEU or PEU -- except that the German sound has to be rather longer than the French one. On the title-page of music published in Germany etc, you sometimes see the word SÖHNE, as in 'X und Söhne' (='X & Sons'). The vowel sound is the same as in BOEHM. The name of the German poet GOETHE (as pronounced in German) also has the same vowel sound. The Scandinavian languages also have similar vowels to this German sound.

But the GÖTTER- element in GÖTTERDÄMMERUNG contains a different vowel sound! It's like the 'e' sound in GET, except that the lips are pushed forward. The word BRÖGGER, associated with features of Brannen flutes, has the GÖTTER- vowel, not the BOEHM one, when it's said in German. (The reason that German has these 'extra' vowel sounds has to do with changes that took place in the language hundreds of years ago.)

You can hear sounds fairly similar to the German BOEHM/FLÖTE/GOETHE vowel in some varieties of English. Especially to North American and West European ears, the pronunciation of the 'ir' and 'ur' sequences in words like FIRST and NURSE in some types of South African, New Zealand and Australian English will seem close to the German vowel.

If you want to imitate the German vowel, try saying the word SHAME, but, as soon as you hit the vowel sound after the 'sh', keep the lips pushed forward and try to stop the tongue moving about at the end of the vowel.

Now do the same with a new word BAME -- ie lips pushed forward for the vowel and no final 'wobble' of the tongue. You won't necessarily get a native-like German pronunciation of BOEHM, but you'll come pretty close to it.

Depending on what sort of English you speak, you'll probably find yourself opting to use one of the following pronunciations in conversation -- unless you consistently adopt the German vowel: BEAM, BAME, BERM, BOAM, BOA-EM, BOA-IM. American English appears to have more pronunciations of the word than British English has. (I say BOA-EM.) Once people see the word in print, they usually stop fussing about how it should be pronounced and accept that it just looks odd!

In fact, the vowel sequence 'oe' can be pronounced in various ways in English, none of which is remotely like the BOEHM vowel. This is one reason why it's a difficult word to pronounce when you meet it for the first time. Try PHOENIX, MONROE, SHOE, OESTROGEN and (O)ESOPHAGUS.

In the technical language of phonetics, the BOEHM vowel in German is lowered and retracted from Cardinal Vowel 10, the front close-mid (half-close) rounded vowel; it is also long. Its phonemic symbol is /ø:/.

Return to Top

Mike MacMahon

Various people have raised the question of how to teach (or just improve) double-tonguing. Asking the learner to say a 't', then a 'k', or a 'd', then a 'g' won't necessarily work. Here are some ideas that might be worth considering:

1. The tongue is far larger than most people think, and its volume is about three-quarters that of a tennis ball! It goes much further down in the mouth than it seems to when we look at it in a mirror, and the very back of it, facing the throat, is well out of sight. Its muscular composition and its links to other parts of the head are quite complex. Yet, when we double-tongue, there are only two relatively *small* areas of it that we should be using.

2. The tip of the tongue is much more sensitive than the sides and, particularly, the area further back. This is why one finds it relatively easy to feel what the tip is doing during 'normal' tonguing ('d'), but not so easy during the 'g' part of double-tonguing. Many people imagine they can't double-tongue, simply because they can't feel the 'g' part of 'd-g' properly. They shouldn't blame themselves. The explanation lies in the way the nerve-endings in the tongue have evolved in human-beings.

3. Learn to become aware of the sides of the tongue. This is very important and I don't think it's emphasized enough. Even to do normal 'd' type tonguing, there has to be an air-tight seal between the sides of the tongue and most of the upper teeth - otherwise the air can't be pressurized to create the attack for the 'd'. Consciously feel the sides of the tongue pressing against the side teeth and/or the roof of the mouth as you whisper a series of 'd's whilst you're reading your way through this email. Do the same for a series of 'g's. (During a 'g', you'll probably feel the sides better than the middle part of the tongue that momentarily blocks the air for a 'g'.) Because of differences in the shape of the mouth and teeth, there'll be some variation between individuals as to how much of the tongue makes contact with the side teeth and precisely where the contact is. It's worth 'comparing mouths' with another flute-player - and don't feel embarrassed about poking your finger around the sides of your own (!) mouth and moving the upper lip out of the way to get a better view of the sides of the tongue and the upper teeth.

4a. *Say* the words WIDTH and GEESE slowly. The sounds to concentrate on are the 'd' and the 'g'. You'll find that you don't need to move the tip of the tongue very much for the 'd', nor the middle of the tongue for the 'g'. Furthermore, the distance between the two places on the roof of the mouth where the sounds are made is really quite small, only about 4 to 5 cms. It may feel more, but this is because you can sense a lot of fine muscular changes taking place elsewhere in the tongue as you move from a tip-of-tongue to a middle-of-tongue movement.

Now bring your jaw up a bit further as you would when playing the flute, and *say* WIDTH and GEESE again. It probably feels a bit cramped - this is OK. However, don't imitate the normal 'speaking' versions of 'd' and 'g' in their entirety. Instead, go for much lighter, more subtle, more precise movements. Think of the way ballet dancers have to control very precisely what they do with their feet, particularly the tips of their toes. It's this same precise control that's needed for double-tonguing. Make your tongue 'tip-toe'; don't push it around your mouth as you would for speaking. Remember as well that all you need to do is to pressurise the air (for a 'd', then for a 'g') without letting air escape over the sides of the tongue at any point. You certainly don't need the larger-scale movements of the tongue that we tend to use when speaking. Compare the delicate 'd' and 'g' movements in WIDTH and GEESE with the way we move our tongue and jaw around in words like FAR, HAT, or SLASH.

4b. It might also help to practise saying the words 'eenie-meenie' two or three times - but as a young child would say them. This is so that one feels how the tongue can be bunched up in the mouth, and well forward, with the sides preventing any escape of air. Many people naturally favour this sort of positioning of the tongue when they single-tongue and double-tongue.

4c. Pretend to be a ventriloquist and make 'd' and 'g' sounds without moving your lips or jaw - but *don't* tense the lips and jaw. Whisper DIG, DEED, EAGER, GIG. As you do so, *feel* (a) the relative lack of movement of the tongue, and (b) how close the 'd's and 'g's are on the roof of the mouth.

5. Some people block the air in their throat when they tongue - and especially when they try to double-tongue. What they're doing is closing the vocal folds (vocal cords) in their larynx (=voicebox) just when they're making the 'd' and 'g' sounds in their mouth. (One reason for this is that they transfer across to their flute-playing certain actions they instinctively use in their pronunciation of 'd' and 'g' sounds in their own language or accent.) To avoid this happening, practise making a series of 'd-g-d-g-d-g' movements with air flowing absolutely unhindered from the lungs and into the mouth. In other words, *breathe through* the 'd's and 'g's. But deliberately use far less air than you would for speaking; this will also help to concentrate the attention on those precise movements of the tongue.

6. If you want to try the old-fashioned 'd-l' type of double-tonguing, you can work out how to do it by saying the words GLOVE, GLAD, GLEE with 'dl-', not 'gl-', at the beginning of each one. (Make sure you say DLOVE, DLAD, DLEE, not DuLOVE, DuLAD, DuLEE, though!) Now *whisper* them in their DLOVE, DLAD, DLEE forms and feel what's happening in the 'dl-'. (The 'l' is equivalent to the 'g' of modern 'd-g' double-tonguing.) There are only two drawbacks to 'd-l' for double-tonguing. One is that the air-jet for the 'l' isn't as precise as for the 'd'; the second is that the air for 'l' doesn't come out quite in the mid-line between the teeth - it cascades over one or both sides of the tongue. (This may suit some people, of course, depending on the arrangement of their teeth.) On my Boehm flute, I use 'd-g'; but on my early 19th-century six-keyed flute, which has a rather different embouchure hole and no lip-plate, I find 'd-l' tends to work better.

7. See if you can get away with practising double-tonguing (of the 'd-g' variety) on a train, plane or bus (fluteless, of course!) without anyone noticing. The ultimate test, I suppose, would be to sit in the audience at a concert and double-tongue the solo in the 'Midsummer Night's Dream' scherzo whilst the flautist on stage was playing it. If your neighbour glares at you, you'll know your tongue movements have been too strong!

Mike is a phonetician in the Dept of English Language at Glasgow University; he is also a Council member of the International Phonetic Association. He plays flute and piccolo in the Glasgow Symphony Orchestra. He can be contacted at; his URL is

Return to Top


I think too much can be made of the apparent disadvantages of tilting the head slightly to one side -- although I appreciate the benefits of the insights about posture provided by the Alexander Technique. If one looks at the shape and contents of the neck, it's quite clear that the size of the channel that the air has to get through at its *narrowest* point in the neck is relatively small, compared with the overall width and volume of the neck. And critically, this narrowest point is below the level at which the head tilts. Once the air's past this point, there's much more room for it to 'spread out'; and any tilting of the head is very unlikely to interfere with its movement forward to the lips. 'Constricting the neck' by tilting to one side will certainly not interfere with the position of the vocal folds. (However, bringing the folds closer together is a quite separate issue, as is tightening the very top of the larynx. These movements can -- and do -- interfere with the aerodynamic properties of the sound. But this is not the same as tilting the head.)

By way of explanation: the air from the lungs has to pass through the space between the vocal folds located in the larynx (the voice-box). The dimensions of this space are really extremely small. Imagine a triangle, two sides of which (in men) are about three quarters of an inch in length; the other side is about half an inch. In women, the dimensions are even smaller: two of the sides are about half an inch in length, the other one about a third of an inch. I suggest you draw these dimensions on a piece of paper to appreciate just how small the gap is.

Once the air has got through this gap, a much larger space is encountered. The throat area above the larynx is enormous compared with the space between the vocal folds. Look at the back of your throat in a mirror to appreciate the width of the space and compare that with the limited space between the vocal folds.

When one tilts the head playing the flute, the larynx (despite the fact that it can move) would normally stay still. The air-jet, having passed between the vocal folds, would then move into the voluminous area above it -- and, even if the head is tilted, there is still plenty of space for the air to pass through without hindrance. Thereafter it would be directed into the mouth, where, depending on the position and configuration of the tongue, it would aim for the space between the lips.

I really don't see a problem -- apart from possible neck tensions for some players, depending on their personal anatomy and physiology -- in tilting slightly to the right.

Mike MacMahon
From FLUTE list March 1999

Return to Top

Head-voice "sound" while playing?

It's easy to work out whether the vocal folds (aka vocal cords, vocal chords) are vibrating or not. It can be done whilst you're playing the flute -- but only with the left hand. To begin with, try it without the flute.

Place your index and middle fingers of one hand on either side of the Adam's Apple -- the bit of the larynx that sticks out at the front of the throat. Say a long 's' sound; you'll feel no sensation in the fingers. Now do a long 'z' sound; you WILL feel the vibration. The vocal folds are vibrating together very rapidly (eg at least 100 times a second in men; usually at least 150 times a second in women).

An alternative way to work out if the vibration is there or not is to stick your fingers in your ears as you say the 's' and 'z' sounds. You'll hear the vibration for the 'z' sound. (The vibrations from the larynx are transmitted to the skull.)

The humming sound that some flute-players report is because air is resonating in the nose at the same time as the vocal folds are vibrating. In other words, two things have gone wrong: the vocal folds have started to vibrate, and the soft palate has been lowered. (An alternative explanation is that the 'humming' air doesn't get into the nose, but resonates towards the back of the mouth and in the throat itself.)

To experience the nature of the humming sound (at least through the nose), try a long 'm' sound, comparing it with the sound of just breathing in and out through the nose -- whilst you're grasping your throat or you've got your fingers in your ears.

The 'rotten sound' that Lesl mentions is because at very slow speeds of vibration the vocal folds tend to move less rhythmically and a so-called 'creaky' sound results. I'd expect this to be very unusual indeed in normal flute-playing -- but there's no reason why it shouldn't be used as a special effect in some modern styles of playing, eg jazz.

The thing that causes the folds to vibrate together is essentially the characteristics of the air-stream coming up the windpipe from the lungs. One moment the air pushes the folds apart; the next it causes them to be sucked closer together (the Bernoulli effect). If when playing the flute, the volume and velocity of the air-stream aren't quite right, then the folds can be sucked together and vibration will start up.

There's a useful website on laryngeal anatomy at the University of Pittsburgh. Do have a look at the series of colour photos of the vocal folds in different positions.

Return to Top

Return to my homepage