Programming African Rhythms
There has been an enormous upsurge of interest in African music of late. Artists as diverse as Paul Simon, Peter Gabriel, Talking Heads and Brian Eno have acknowledged its inspiration on their work. But there is little information around that explains the fundamental concepts behind it and how a musician can incorporate the sounds and ideas into their own work if they wish to. Kofi Busia suggests a few ideas.
Africa is a huge continent with a bewilderingly diverse array of peoples and music in it. In spite of enormous stylistic differences, there are broadly common instruments, musical conventions and a commonly accepted vocabulary of harmonic possibilities. Around the major cities of Africa definite types of urban music have developed. These are a synthesis of Africa's own ancient traditions and contact with the Church, rock and roll, soul, pop, jazz etc. The most well-known examples of this are 'highlife', 'jit' and 'Townships music'. Here I intend to discuss only the more traditional (more 'ethnic'?) musical forms. The work of groups like The Bhundu Boys, King Sunny Ade, Ladysmith Black Mambazo, etc, is already a synthesis between African and Western music. Thus, it seems of greater value to outline one of their sources so that each of you can likewise create your own synthesis.
I have geared my explanation towards those who use drum machines, because one of the major advantages of a drum machine is that once it has been programmed to play something, no matter how difficult a pattern may seem, it will carry on playing it faultlessly giving a chance to either play along or just sit back and assimilate what is going on. Drummers can, nevertheless, also gain something from what I have to say as the principles are still sound.
African music developed completely different kinds of musical instruments from those used in the West. Western music evolved harmonically/melodically. It became necessary to develop instruments where pitch was predictable, controllable, and variable, and which allowed harmonic structures and overtones to blend suitably with each other. Composers and instrument makers therefore constantly modified both their music and the instruments in order to achieve the right, compatible, kinds of textures. As new harmonic conventions gained favour or fell out of fashion, so also instruments were adopted, adapted, or fell into disuse. The instruments used emphasised certain limited portions of the harmonic spectrum (basically those with whole number overtones). Western music developed an arsenal of pitched instruments with definite harmonic structures capable of sustaining notes. Although melodic/harmonic sophistication grew apace, rhythm did not.
African music went in a different direction. It showed little concern with melody or harmony, or with compatible overtones, and developed a wealth of unpitched instruments with disharmonic overtones that were non-sustaining. African music concentrates on using percussive textures, and emphasises the role of rhythm in musical form.
Instruments are played over a given period of time, known as the time span. Pulses within this span are used to create rhythm and make music. These pulses are derived by dividing the time span into two and multiples of two, or else into three and multiples of three (see Examples 1 and 2).
All of these patterns occupy the same time span, ie. they are equivalent. Wherever we have four pulses occupying a section of time (Line 1B) we could, if we wished, substitute six pulses (Line 2B). If we wanted to have nine pulses instead, then taking Line 2A, which has three pulses, we just replace each single pulse with triplets. Hey presto - 3x3 = 9! You will have to wait to see how we could construct, say, a 15 pulse structure.
Alongside each of the example patterns above, you will see that quavers, crotchets, etc, have been drawn. These are for the purposes of illustration only, to show how one might approach trying to write a given pattern into a drum machine. (Line 1D would be programmed into the pattern as semiquavers, Line 1C as quavers, etc).
It is enormously tempting to think of Line 1D as being 'twice as fast' as Line 1C, which is again 'twice as fast' as 1B, and so on. After all, that is what the notational convention behind the use of quavers, etc, is all about - isn't it? It is most important that this line of reasoning be discarded, otherwise one of the most important principles of African music will be missed, and with it a way to approach rhythm differently.
The actual truth of the matter (as far as we are concerned) is that the rhythm pattern on Line 1D is simply twice as dense as the pattern on Line 1C, which is again twice as dense as the pattern on Line 1B. It is for this reason that I have avoided using traditional notation in the examples - such as crotchets, quavers, etc - and resorted to the more neutral • symbol instead.
Line 1D represents a fairly dense structure. We can tell this because the 'notes' are so close together. Line 1A, on the other hand, is a fairly spacey structure. In a performance, Lines A or B are the ones most likely to be clapped and used as the conductor or reference point. Higher density structures (such as Lines C or D) tend to form the basis for soloing and improvising.
It is important to grasp that it is density - not speed - that is the major difference between Lines A, B, C and D.
'Hemiola' is simply a fancy musicological term meaning the substitution of three beats for two, or two for three. If Line 1B represents a bar of 4/4 (four pulses), we can easily substitute a set of triplet crotchets (Line 2B) for it. It is more common to substitute triplet quavers, as in Example 3.
Such substitutions can add a lot of interest to a rhythmic figure, and by means of it, practically every seemingly complex time signature can be reduced down to a simple one.
This forms almost the entire basis of Western music. It is what bar lines are all about. Bar lines and the time signature (eg. 4/4) set out a definite division of the time span, and lay out a regular way of counting beats that is to be followed by all instruments in the ensemble. 4/4 time means just that: 1,2,3,4, and then start again - 1,2,3,4. The beat can, though, be divided up into smaller and smaller notes — quavers, semi-quavers, hemi-demi-semi-quavers and the like. But whatever instrument is playing these smaller divisions must play them sufficiently fast enough, and regularly enough, to ensure that the main beat is not upset. Quavers, and so forth, are not regarded as entities in their own right, but simply as smaller divisions of the main beat.
It is because the beat can be divided up in this way, into 'smaller' and 'smaller' notes, that the scheme is called divisive rhythm. The whole point of a semi-quaver is that it is intended to be conceived as being four times as fast as a crotchet. The semi-quaver's essential character does not come from the fact that it is used for musical structures that are 'four times as dense'!
A couple of interesting observations can be made here. Firstly, of course, another reason for the notational scheme used in Western music is that since the instruments used can sustain notes, it is important for the player to know how long they should be held for. A semi-quaver, no matter where it is met, is not held as long as a minim. With percussive instruments, this consideration is not an issue. You hit them and that's it! Their sound decays instantly (although things like gongs can take a long time doing so). In fact, it is for this very reason that many drum machines and drum synthesizers do not bother to recognise MIDI Note Off messages, as they have no effective need or use for them.
The second interesting point is that, in Western music, semi quavers are 'small' notes (being smaller divisions of the beat). Crotchets, therefore, must logically be 'bigger' notes. Western music sees itself as going from 'low' notes to 'high' ones. African music sees things quite differently: it is the low pitches that are considered 'big', because they generally play the more spacey, less dense, foundational structures and figures. (Think of a big bass drum.)
High pitches are considered 'small', because they usually play the more dense, quicker figures and cycles. Where Western music goes from low to high; African music goes from big to small. Similarly, Western musicians see notes as quick and slow; Africans see time as being thick (dense) and thin.
The purpose of additive rhythms, however, is to free us from the obligation of having such schemes and note groups. Additive rhythms allow us to combine different rhythmic pulses. 5/4 time, for example, is merely a section of 3/4 time immediately followed by a section of 2/4 time, followed by a section of 3/4, then 2/4 again, repeated throughout the whole piece. By taking one simple rhythmic pulse (3/4 time) and adding it to another (2/4 time), we have created an apparently complex time signature. (Incidentally, the way to fill a time span with 15 pulses is to use 5/4 time and replace all the beats with triplet quavers.)
Consider two bars of 4/4 time. With conventional divisive rhythm this would be viewed as 8 crotchets divided into two groups of 4. (Count 1,2,3,4 twice). With additive rhythm, however, the same 8-note group can be arranged any way we like. It could be 4+4. It could equally be 5+3, or 3+2+3, or 6+2,or 1+3+2+2, or any other combination of 8 we choose.
There are over 100 different ways in which these eight basic beats could be realised. And once variations in density, hemiola, tonal contrasts and so forth are taken into account, the possibilities are infinite. And that's just for two simple bars of 4/4 time!
To summarise, the distinguishing feature between additive and divisive rhythm is that with divisive rhythm (Western music), although the beats can be divided up into smaller notes, the same beat and counting scheme is used for all instruments; ie. they all follow the bar lines. With additive rhythm, however, although pulses can be divided (ie. made denser) there is absolutely no reason why these subdivisions should follow an overall counting scheme. Pulses are strung together to make the desired rhythmic structures, and any rhythmic figure can be longer or shorter, less dense or more dense, than any other. Conventional bar lines are thus virtually meaningless. They may be useful for notating the cyclical rhythmic figure for any one instrument, but may well be irrelevant for others.
Performers have to be able to 'internalise' the time span and follow it without error, no matter how much the particular figure they are playing may be divided, added to, or varied in density. It is, after all, the only major reference point. If this doesn't happen, the music very rapidly falls apart.
In African music, because it can be difficult for musicians to internalise the time span, it is usual to 'externalise' the basic pulse in some way - either with handclaps, a cowbell, sticks, or some other percussive instrument. This figure is known as the time line, and is the foundation of the music. It is analogous to the drone used in Indian music, or to the chord sequence that a jazz band improvises over. The time line can be simple and straightforward (regular clapping), or it can be a more complex repeating rhythmic sequence. Generally, though, it is kept short (4 or 5 notes) and played by one or two easily heard instruments, such as handclaps, cowbell, or tambourine.
In African music, an instrument can be assigned a different or similar rhythmic figure to any other, and is related in some way to the time line. Rhythms and instruments can be arranged so that they interlock. For example. Instrument A may start playing one pattern. A few moments later Instrument B will start playing the same pattern, followed by Instrument C. This creates a 'canon' or 'round' effect like that produced by groups of children singing 'Three Blind Mice'. Although each group is concentrating on singing its own tune, eventually a rudimentary harmony is built up. By the same token, giving the same rhythmic figure different points of entry creates similarly interlocking rhythmic structures. This procedure is called spacing, and gives rise to two important effects - 'crossrhythms' and 'polyrhythms'.
Crossrhythms arise when different pulse divisions of the basic time span are set against each other. The very simplest kind of crossrhythm is that based on two beats set against three (Example 4). At the beginning of the time span the triplets in Line A 'beat against' or 'cross over with' the duple in Line B. Immediately after that the reverse occurs, when the triplets in Line B are playing against the duple in Line A. Very much more complex crossrhythms arise when divisive and additive rhythms are set against each other, and when there are lots of different kinds of note substitutions taking place.
Now for polyrhythms. Consider a situation where one instrument plays in 3/4 time, a second plays in 7/4, and yet another plays in 9/8 time. This is much less complicated than it might at first appear.
A cycle in 3/4 could well be made up of beat 1 on a low tom, beat 2 on a medium tom, beat 3 on a high tom. No problem there. Set that going and consider it our time line. It is nothing more than a three-figure cycle of moderate or medium density.
Now the 7/4 cycle. This could be umm, cha, umm, cha, umm, cha, cha played on (say) snare and cymbal. No problem there either. This is a seven-figure cycle of similar density (crotchets). Get this one going simultaneously with the above 3/4 cycle and let them run. You will notice after a while that you have created a composite rhythm pattern that repeats itself - every seven bars of 3/4, or three bars of 7/4 in fact!
The 9/8 cycle is a more dense, more busy, pattern than the other two. All we need is a regular 9-beat cycle played on any one, two, three or perhaps four instruments of our choosing. I'll leave it to you to devise a sequence of your own. Perhaps conga, cowbell and handclaps - maybe sticks as well? Anything, as long as it is a regular 9-beat pattern.
Now let all three cycles play together. There is nothing conceptually difficult about this arrangement. It is, in fact, quite a simple set-up. All it takes is a different approach to the idea of drums, rhythm and time. Polyrhythmic effects come from the spacing (or offsetting) of rhythmic patterns in terms of their start points, as described earlier.
Both crossrhythms and polyrhythms are often simultaneously handled by one and the same performer. Every good African drummer learns to play different figures in the left and right hands (whether they be crossrhythms or polyrhythms) early in their career. In fact, quite often the left-hand part interacts (perhaps playing a crossrhythm) with the patterns of the drummer on the left, while the right-hand part intersects (perhaps chasing polyrhythmically) with the drummer on the right. Each drummer can thus interact with neighbouring drummers until every drummer in the group is interacting with everyone else. (This, again, is not as hard to grasp as it might seem. After all, in an orchestra, every player has a different part to play but they are all contributing to the same symphony!)
Although African music is not overly concerned with definite and repeatable pitches, nevertheless, one of the most important ingredients in percussion arranging is achieving an effective tonal contrast between different drums. Having drums tuned to different pitches enables the arranger to highlight and contrast crossrhythms, polyrhythms, and the like. Little tunes and figures can be switched from one drum to another, or shared between different drums. Rhythm may well be the primary focus, but attention must be paid to pitch and tone. The aesthetic appeal of drumming lies in the effective organisation of both rhythmic and melodic elements. The rhythmic figures and instruments must be graded according to their density and complexity. Whether they are lead instruments or accompanying instruments, 'call' instruments or 'response' instruments, low-pitched ('big') or high-pitched ('small'), such things must all be considered when deciding who is to play what. Here are a few guidelines to follow for an effective approach to drum arranging:
• Decide on a suitable 'time span'. This does not have to be one bar of 4/4. If you are playing a ballad, why not think conceptually of a rhythmic scheme that stretches languidly across two, three, or even four bars? This will allow you to create slow moving, spacey structures that reflect the nature of the song. For a more up-tempo number, why not use a time span that is only two or three beats long? You could then create increasingly dense structures to highlight the more frenetic and 'busy' nature of the music.
• Choose a 'time line'. It should be a simple, short, repeating pattern - a three, four or five-note figure works best. It needs to attract the listener and be catchy; it is, after all, your 'hook'. It could be played across two or three different percussion instruments if you wish, but keep the basic feeling and nature of the song in mind. If it is a slow, romantic number, you do not really want the time line played on a 'hard' sounding instrument. If it's a hard-edged song, perhaps a tough, tight-skinned snare sound would suit. With a dance number, it is usually a good idea to have the time line clapped, or else played on sticks or a cowbell that dancers can hear and relate to easily.
• Having chosen a time line, you need to think of what other percussion instruments you might want. What goes with the mood you are creating, and either complements or augments the instruments and figures on the time line? If you have already used cymbals on the time line, for example, it can be confusing to use them elsewhere for other patterns.
How dense do you want these other structures to be? Instruments like bass drums, being 'heavy' (producing 'big' notes in the African way of thinking), tend to be used for slow or moderate densities. Cabasas, tambourines and the like, being 'lighter', can be used for much more dense, faster, structures.
How many different rhythmic figures in total do you want? And how long or complicated do you want them to be? Be careful here, because if you have too many cycles and structures going on in your song at once, the song will lose all sense of identity. It is best, in the early stages, to go for a few short rhythmic figures that can be easily arranged, controlled, understood and identified. It's like arranging vocals - nothing should obscure the main vocal. A backing singer may well have a really great line to sing, but he or she should never ever drown the lead vocals. Their job is to support and enhance, to add harmony, texture and/or counterpoint, not to try and hog the limelight. This is how you should approach other rhythmic figures and cycles. Why not introduce a little polyrhythmic, canonic effect? Short three or four-beat cycles chasing and echoing each other can be extremely effective.
• The most important thing is to stop thinking of drums as simply 'percussion' instruments. Avoid leaving them until the very end of the song, when everything else has been set out and you just want 'a few backing beats'. They should be an integral part of the arrangement from the very start. Drums, as any African will tell you, are effective and valid musical instruments too. They make music just as effectively as a guitar or piano, although Western music has not used them in that way for centuries.
Having thought of a few rhythmic patterns and figures, there comes the major headache of actually playing everything! This is where drum machines can be a real life-saver, but also where you have to exercise a bit of patience.
Sit down with several pieces of paper and work out, bit by bit, what instrument is playing what and when. It can seem a drawn-out process, but if you have a drum machine that will only record in either 3/4 or 4/4 time (as is so often the case), and where patterns can only be programmed in step-time, there is little alternative. The average drum machine was not made with additive rhythms, polyrhythms, or African percussion in mind! If you have a sequencer that you can record into in real-time, things can be a bit easier. You can play each rhythm part individually and record it. Then you can use the editing facilities of the sequencer to fine-tune the arrangement. You can even use the MIDI note assignment facility on the drum machine to access each drum voice from a keyboard, or else play direct from a drum pad or whatever.
Using the basic principles outlined in this article it is very easy to create a four minute 4/4 song which plays for 100 bars without repeating a single bar! And yet, if you lay out your cycles with care, your audience will still be able to follow you through every move. I hope you'll agree that, musically, this is quite an exciting prospect.
I have not on this occasion been able to say much about the actual percussion instruments and what they can best be used for; similarly, I have only talked briefly about the cyclical figures that can be created and the real art and significance of placing different length cycles against each other. (As an example: octaves 'fit' when played together. In the same way, a 2-note cycle is obviously going to 'fit' with a 4-note cycle. Unless, of course, we make one of them polyrhythmic. If 2:4 is the rhythmic equivalent of the consonance of an octave, what is the rhythmic equivalent of a musical third or a dominant ninth?)
If you would like to actually hear these African rhythm ideas in action, you could try getting hold of my own album, Oh Africa, which puts everything mentioned here into practice. It might give you some extra ideas, as well as help clarify matters further. Practical demonstrations are worth thousands of words.
I do hope, though, that I have been able to pass on a few useful new ideas for your own rhythm-making. African music is the world's most ancient type of music, and it has a lot to offer those willing to find out about it and experiment with it. Happy drumming!
Kofi Busia's LP, 'Oh Africa', is released by African Records International, (Contact Details).