Djembes are drums originating from west Africa. A little bit like congas there are a variety of sounds gained but based on 3 pitches (striking only the skins, a 4th one if you include the side and then the iuse of various mallets and methods of striking the hand on the skins adds more nuances). Anyway, I hope to write more but for now I have a recording of the piece done quite well at the music school I am at

You need to be a member of Composers' Forum to add comments!

Join Composers' Forum

Email me when people reply –


  • Hi Chris,

    Really nice to hear someone pick up on this instrument and place it in a context that brings the best from it. Traditionally (over here anyhow) the djembe is the drum of choice for the drum circle, where it presents an easily accessible medium for people of all backgrounds to feel and take part in the joy of making rhythm with others. The down side of this is, from a percussionist’s perspective is that it’s voice can sound a little clichéd, and one has to work hard to find an original perspective for it.  I flirted with it briefly in a folk quartet; I was drawn to the wealth and the warmth of bottom end that you can draw from centre of the drum, and to the nuances that the edges produce. Reminded me somewhat of the extremes of the tabla, so this was the route that my small excursion took me on. The difficulty with it is that, the warmth in the drum is best drawn in an intimate environment, with close reflective surfaces, here it sings and responds organically to the briefest of touch. In a large concert hall it’s response tends to flatten out and you’re left with the mid to top stuff. I never did find a way to mic it well to try to compensate for this.

    I also found that you had to search hard to find a good instrument capable of ‘singing’ in this way, as there are so many cheap tourist drums flown in.

    But back to your work. A very enjoyable piece indeed. A fine performance from the players. On first listen I thought to myself why aren’t we getting the usual African polyrhythmic interplay, with the bass locking a repetitive cycle, but then I realised that this is generally the usual setting, and the piece is all the stronger for the scarcity of it, and a freshness of approach.  It does however set up a tension from the expectation that any time soon we’re going to be propelled forward into dancing around the room to some lovely warm African groove. What you’ve created instead is a really satisfying and interesting demonstration of the drums voicing, and something, which must be a pleasure both to perform and to watch and hear live. Nice work!


  • Thanks so much and yes I so agree that drumming needs to be in an intimate space with reflective surfaces. in the concert hall the bass does get a little lost in the reverb and longer delay. When I heard it in the smaller practice room the piece was like electricity or 100 cups of coffee I was so energized by it. I hope to write more but get it recorded in a smaller space with more acoustically reflective surfaces.

    I also do want to write a piece that does go into a dance but again still try and avoid it becoming too cliched (I don't think you can avoid it entirely). 

    One unusual technique used in part of this piece was serialization of rhythm which Milton Babbit pioneered (along with Messaien). Babbit's method is very simple and one wonders why it had not been though  of earlier. It can lead to some cool intricacies.

    First, take time signature say 5/4 to generate a number of beats., say 15.  Now use triplets as your basic rhythmic unit. So you will have 15 beats within the 5/4. Now for ease, draw a number line from 1 to 15 or 0  to 14.   You can develop rows of time points, that is when a beat will be accented, omitted, or preceded by silence.  Say 2, 5, 7, 9, 11, 13, 14 is one rhythmic row.  These beats which lie at various points of your triplets can simply be the ones accented, or they are always preceded by a rest.  The other row 1, 3, 4, 6, 8,10 ,12  could be the points in time where you have rests.

    Serialization of rhythm might offer greater complexity that is easier to manage than pitch because you do not have the restriction of using 12 pitches or added complexity of expanding the pitch classes to greater than 12 by using 1/4 tones and alternate tuning systems using more than 12 pitches.  of course, the the rhythmic base you use to create rows runs the same risk - if you choose 52 beat lengths to create 4 rows of 13 rhythmic points, well the composer and listener will find it easier to subdivide the beat into smaller more manageable units.

    If you want to read more about serialization of rhythm and Babbit, Mead has written a great book about Babbit's composition procedures. Not an easy read but manageable and a good intro to a great composer's methods.

    PS. Where do you reside?

  • Hi Chris,
    An Interesting expanation, and maybe you've achieved a first in writing this way for the instrument? Look forward to hearing the next one, and how you manage the dance element. Sounds fascinating.
    I'm based in the UK in Nottingham.
  • Bob, before you read further, where or from whom did you study African drumming and are you still doing it? Also, I think some of the dissatisfaction might come from the fact the performance was in a large hall and the recordings is a somewhat compressed mp3 - I swear it sounded far richer in concert and electrifying to me in a room with acoustically reflective surfaces.  So though a live performance, I think some compression and the venue take away a few details of the work because I have been on the fence about the piece at times due to this.

    Now Bob, as you know, much Western classical music is derived from dance form too, Sarabande from Spain, the common Minuets, Waltz (of which Brahms' 3rd is one of the most novel uses of the waltz dance form), etc.  And these dance forms have been turned into pieces that have stop and go motion.  One reason for the stop and go motion in this piece is to add substance (in addition to the serialization of rhythm), it is a binary form piece with intro and coda that is continuous. The Intro that has the most stops and starts highlights the serialization of rhythm and the sounds to be explored - finger rolls, knuckle rolls, alternation between stopped and open sounds. A transition of continuous rhythmically active music followed by a diminuendo through finger rolls leads to the second section which is continuous rhythm roughly playing 5 against 4, then there are two solos with the rolls underneath providing accompaniment with a coda. So Intro, A, B, Coda - an elaborate binary form.  

    My explanation is to counter and show my disagreement to your broad statement "So much modern music seems to be about setting some kind of mood, but without any substance to back it up."    I think I get what you mean and a famous example is Beethoven's Septet - his most popular work by the way (not saying everything that is popular is vacuous) but there is NOTHING that challenges the listener and which the composer is just filling out the form. It is disposable easy music and - after the first listen gets a bit dull - the melodies are catchy and the instrumentation is nicely done which I think keeps this piece in the repertoire.. 

    Finally African rhythms while they become extremely complex can be notated - in fact when western composers do achieve a very good approximation of such rhythms and use it in for their own ends in their music  (for example Ferneyhough or Ades) as well as do things such as introduce time signatures not based on 2, I often hear people say it is needlessly complex and unplayable.  So 1) African rhythms can be notated  2) Western composers have found notation that approximates such rhythms and have employed them indirectly in their works - check some of Fernyhough's String Quartets or Ades String Quintet..  Substantial work that is satisfying.

    And in fact in some world musics, non-Westerners find the employment of rhythmic techniques extremely simple - Steve Reich uses Javanese interlocking of rhythmic pulses that are far more complex in Java.  I imagine they have the same reaction as we do when we do hear musicians highly trained in non-Western music practices try their first hands at say the style of the First Viennese school - it ios then you realize such "direct, approachable" music such as Mozart's Linz is actually as complex as Milton Babbit or Schoenberg who in in their best works cloak their complexity!


  • Thanks for the clarification Rob!

  • Hi Bob,
    Good to hear you had so much fun with the instrument, and your tuition. On your suggestion of twin mics, bottom and top, I've been there, and whilst it's not an entirely fruitless search, one can occasionally get close, it's problematic on many levels. Any player of an acoustic instrument that has to perform in multiple and widely varying acoustic environments, is aware of the additional inspiration, or magic that appears when playing in a space that resonates sympathetically with the instrument. When placed in a sympathetic space the djembe reveals colour and texture that remains largely sonically invisible elsewhere. I eventually came to the conclusion that much of what I was seeking to capture through mic's existed only in my personal experience of playing the drum, and that the resonance was transmitted through my body, setting up specific vibrations within me. Such stuff was as much physically percieved as heard. Add to this that the space is also influenced by many other factors. Setting up carefully in soundcheck, before anyone has arrived is all well and good, but by the time you've added an audience in their clothing, the reflective surfaces of their skin, the temperature and humidity changes, affecting both the tension in the head, and the dynamics of the room. Well, trying to mic the instrument in a classical sense that retains all of those illusive nuances that I found so interesting is frustrating, and largely pointless in reality. I wasn't even sure that others were hearing what I was, in my body, anyhow. But they do benefit from the lift in performance that a good space imparts in me, so fairs fair I figured.

    Drum circle stuff, and generally louder dance performance whether backing other musicians or playing solo is much easier to mic. Twin mics work well, if you have a good guy on the desk that knows what he should be capturing, and is sensetive to the balance between the two sources. The throat of the drum produces so much level, and wild, boomy bottom end that you're better to avoid using a Bass drum mic. My particular gig involved Voice, Harp, Acoustic Guitar and Ullean pipes, the playing was much more intimate and dynamically varied. I reluctantly gave up and reverted to the darbuka and hand played floor tom, which was much more successful. I did carry the djembe, just in case I occasionally found such a space to bring the best out in it. The one gig where it really worked well was entirely acoustic, no mics, small audience, small warm dry room, shiney reflective stage, lots of drapes, could have heard a pin drop. Heaven :)
This reply was deleted.