The Game of Mbira

The Game of Mbira

by Adrian Wagner

Zimbabwe, 1999


            I would like to thank my advisor, Keith Goddard, whose guidance, advice, and encouragement made this paper possible.  I would also like to thank Scott Peterson, for enabling such a rich experience of Zimbabwe, and whose structure and help in preparing for the ISP period made the experience of being completely free to explore, all the more rich.

            I would like to give a special thanks to Jonathan Mungate, whose wisdom, generosity, honesty, and spirit became the foundation of a fantastic mutual exploration.  No words can express how grateful of you I am for stopping to listen, for pushing the limits, and for sharing so much of yourself in the journey.  In addition, I thank the entire community of Seke, for their hospitality, curiosity, richness, and sadza, and for creating such a wonderful environment in which to explore.

            I also thank all those who taught, who listened, and who shared, in the experience of mbira.

            Lastly, I thank my mother and father, for forcing me to take piano lessons for 4 years, long before I knew, myself, how important music was to become.  I love you infinitely.

Table of Contents

1.      Acknowledgments                                                                                                             pg. 2

2.      Table of Contents                                                                                                              pg. 3

3.      Abstract                                                                                                                             pg. 4

4.      Introduction                                                                                                                       pg. 5

5.      Methods and Techniques                                                                                                   pg. 6

6.      Chapter I : A Quick Glance at Mbira                                                                                 pg. 7

7.      Chapter II : The Grit of Music Theory Rubbed Smooth                                                     pg. 11

8.      Chapter III :  Learning the Game of Mbira                                                                         pg. 18

9.      Chapter IV : Checkmate : The Art of Improvisation on Mbira                                           pg. 34

10.  Conclusion                                                                                                                        pg. 44

11.  Bibliography                                                                                                                      pg. 46

12.  Appendix A : The seven chords on the mbira dza vadzimu.                                              pg. 47

13.  Appendix B:  Notation and analysis of traditional shona music.                                        pg. 50

14.  Appendix C:  Rhythms                                                                                                      pg. 53


            This paper is the product of my own exploration of the mbira as a musical instrument.  It looks at western music theory, and its relation to the design of the mbira.  This paper also introduces my own form of theory, a theory that is created from the instrument itself.  My aim is to share my understanding of mbira and to create a base upon which further exploration can be made.  There are many rules to the game that have yet to be written.


            I was an hour into my mbira lesson with Ephat Mujuru and I had already spent 45 minutes of it working on a single pattern, a variation of a song called Dambaza.  My fingers were fumbling across the far reaches of the keyboard, striking the keys in the order that I tried to remembered.  The notes simply would not come.

            After a few more attempts, I stopped playing, and took a deep breath.  I picked up my mbira again and in it I saw the song.  I saw its pieces.  Pieces of patterns that I already knew from the songs I had learnt before, patterns that come from the instrument itself.  And as I put the pieces together, the entire song emerged.  The variation was mine.

            As Ephat Mujuru said, “you have passed the test.”

            Mbira is the sport of the mind.  It is the game of an apparent simplicity in which is contained a complexity that can take months to explore.  Such was my experience over a 2 month period spent in Zimbabwe, 2 months spent exploring the complex logic of the mbira.

            There have been many papers and books published on the importance of mbira within a cultural setting, on the history of an instrument that shares basic characteristics with others found all over Africa, and on the analysis of the music and lyrics that are such an integral part of Zimbabwean culture[1].  But, as my journey into the instrument progressed, and it moved further and further away from these aspects of the music;  I found nothing published on the logic behind the instrument itself.

            This paper attempts to explain this logic, to reveal it in the same way that the instrument revealed it to me.  This paper is my journey, into the game of mbira.

Methods & Techniques

Journal Entry, 11/13/99:

            “I just had a three hour lesson with William, and it felt more like an exploration of thought than anything else.  The man’s mind is like a computer.  You can see it in his eyes, as each new pattern forms on the keys.  His stare twitches until you see his thoughts beginning to hum;  his finger disappear into an apparent chaos.  And the music is incredible, like a swarm of melodies moving through all dimensions.  And I ask, ‘by what means can I reach this state?’”

            The process of uncovering the hidden complexities of mbira was one guided by observation.  In the course of two months, I took lessons with 8 different teachers.  I learnt many songs and their variations.  The answers that I sought, however, I did not find in the songs themselves, but in the bending of the songs.  The answers came from the ways in which the mbira players made the songs their own.  Each player has his own style, his own variation of a given song.  But between these differences are hidden similarities that came not from the songs, but from the instrument itself.

            So as I watched William’s hands flash before me, I did not hope of tracing their exact movements.  Instead, I looked for the similarities in each of the movements, as he jumped from chord to chord.  And when I found some type of connection, I would incorporate it into my own playing.  Then, I would play to see if it worked. 

            I came to the mbira with over 10 years of experience, playing piano.  I believe that this greatly helped my process of discovery, based on the basic understanding of music that piano has given me.  This understanding of music theory carried my own exploration, as it carries this paper. 

            Ultimately, however, I learned the mbira by playing the mbira.  I played alone, I played with friends, I played with my teachers, and I played in ceremonies.  As Jonathan says, “Every time I touch my mbira, I will be learning.”  There is not greater truth than this.

Chapter I

 A Quick Glance at Mbira.

            In Fall of 1999, I traveled to Zimbabwe to find a nation very much in turmoil.  Zimbabwe fought and won its independence from minority rule in 1980.  Yet even today the process of reforming political structures, of reestablishing economic security, of writing a new constitution that reflects the views of the people is far from complete.  But beyond the struggles, beyond the percentages of unemployment, the signs of corruption, and the diminishing hope of a better life, there is an incredible richness within the people of Zimbabwe.  It is in their pride of their native language.  It is in the traditions that people follow when eating, traditions still shared by families across the country.  It is in the customs of marriage and the ceremonies of death.  Above all, behind it all, it is in the music, a music that had endured colonization, war, and modernization, and whose melodies stretch back into the far reaches of time.

            The majority of Zimbabweans are Shona, the descendants of a Bantu speaking people that moved into the area of present day Zimbabwe around the 10th century, AD. (Berliner, 32)  They are a dominantly agricultural people.  I was once told by an old man in Seke, a rural resettlement area of Zimbabwe, that there was a time when no man would have taken a leaf from a tree unless he had profound reason to do so.  The influences of Christianity, overpopulation, and a subsequent lack of natural resources have changed this relationship to the earth in most parts of the country.  But there still remain areas where the land is still in the domain of tradition.  And it is in this tradition that we find the world of the spirits.

            It is the traditional belief of the Shona people that every man, woman, and child is born with a spirit.  This spirit lives with the people, throughout their entire lives, taking form as hope, as creativity, or as images in dreams.  When a man dies, his spirit leaves his body.  It wanders about the medium of the universe, until a year later a Guva is held.  I once attended and played mbira at a Guva in Seke.

            I  arrived at the house where the ceremony was to be held, just as the sun had slipped under the horizon.  The red light from the sky was fading, giving way to that of the fire, a light that made shadows dance in anticipation of the night.  There was no rush to start the journey, so I spent a great moment absorbing the warmth of the fire.  By then, the children had already begun the evening, singing traditional and Christian songs to the intricate patterns of a massive drum.  The voices of the two carried into the night, calling the village towards the music.

            In time, I began to play my mbira around the fire, warming my hands in the process.  As more people gathered, they filled the round traditional house that was to become my entire universe for the night.  Before long, I was summoned inside, and the music began.  The mbiras rang out into the heavens as the drums pierced the core of my body.  There was a wind inside the small hut, as what seemed like the entire village was creating a furry of dance, laugher, and tears.  And as the night progressed, I went further and further inside.  Inside, at first, my own music, as I rang my keys among the entire dream, and then inside the music of the masters, who let me all but rest my head on their mbiras as I tried to understand their genius.  In the end, I rose to dance, and kept dancing until the sun was re-emerging from its slumber.

            As told to me by Jonathan the next morning, a Guva is a “memorial service for the dead.  It is a service where we bring the dead back home, the spirit of the dead.  We will communicate with the sprit world and say, ‘Hey, look here, the one who died last year is not yet among you; he is not yet among us.’  He is still dead; he is in the grave.  Wherever he is, we don’t know, but we are trying to bring him in... within.  He can communicate with the other spirits, he can communicate with us.  Before this ceremony, he can not.”

            At the center the Guva, at the center of traditional shona culture, is the mbira.  It is believed that the mbira has the power to call the spirits.   The spirits hear its notes as they reach into their world.  The mbira is the medium through which the world of the sprits and the world of the living can connect.

            Mbira is also played in non-ceremonial circumstances.  It is played in the field to accompany the days work.  It is played at funerals, as the grave is dug, played simply to make the time more enjoyable.  The recent westernization of Zimbabwe has also led to the introduction of an electric version of mbira, which can be found in the night clubs.  Mbira is played with the spirits.  Mbira is played with electric guitars.  Yet beyond its context, it still remains a cultural icon of the Shona people (Berliner, 48).

            The mbira is a small and portable instrument, about 8 inches wide by 10 inches long.  It is usually no more than one inch thick.  The most popular form of mbira in Zimbabwe is the Mbira dza vadzimu, or ‘notes of the ancestral spirits.’  The mbira is characterized as having between 22 and 25 keys, though some have even more.  Regardless of the number of keys, however, the layout of the instrument is the same.  The keys are made out of iron or steel, and are very flexible which allows them to produce a sound. They are wedged onto the wooden soundboard, by a medal bar that runs across them perpendicularly.

            The sound that a mbira makes is most easily described as being close to the sound of a music box, although the pitches are generally far deeper.  The instrument produces a soft and soothing tone when played, which is usually amplified by means of a large resonator.   It was described by Father Dos Santos, in 1589, as such:

“... the iron rods being shaken and the blows resounding after the fashion of a Jew’s harp, they produce altogether a sweet and gentle harmony of accordant sounds.” (Berliner, 27)

In addition to the sound produced from the keys themselves, the mbira is characterized by having a set of rattles.  In ancient time, shells were attached to the sound board, but they have more recently been replaced with bottle caps.  The sound that is produced is a drone of buzzing that accompanies the music.  It sounds like noise at first.  During my first encounter with an mbira in Zimbabwe, I asked the player to cover up the bottle caps, so that I might be able to hear and enjoy the sound in its pure state.  However, after time, the sound that the resonators produce becomes an integral part of the sound of mbira.  The notes ring naked without them.  Skilled players can adjust the volume of their playing such that the bottle caps act as little drums, much like a set of snare drums, rattling along to the music in pulses.

            The mbira is played with the thumbs, as well as the right index finger.  The left thumb covers the entire left side of the keyboard, which consists of two rows of keys, called manuals.  The right hand is responsible for the right side of the keyboard.  With the following division of labor:     

The thumbs strike down on the keys, with either finger tips or nails hitting the metal.  The right index finger plucks upwards (away from the soundboard).  All other fingers wrap around the sound board, except for the right pinkie which is placed into the hole in the board as pictured in figure 1.1.

            As mentioned, rows of keys are called manuals.  As the mbira dza vadzimu has three distinct rows of keys,  it can be divided up into three manuals:  the lower left manual, the upper left manual, and the right manual. 

            The relationship between these manuals and the keys contained within them is a complex one.  A brief introduction to music theory is needed before we can proceed to learn more about the mbiras’ hidden and intricate design.

Chapter II

 The grit of music theory rubbed smooth.

            The word ‘theory’ can be quite intimidating.  When it is combined with such a simple and beautiful concept as music, it seems like it is better if the two are left alone.  After all, birds sing like sirens, and they receive no formal training.  Yet even songs of the birds follow simple and theoretical rules, whether they know it or not.

            The concepts of music theory seem foreign and intimidating, not because they are difficult to understand, but because they are not known.  Music is fundamentally built upon relationships which are numeric in nature.  Six minus two is equal to four. Like wise, seven minus three is also equal to four.  Thus if we say that the distance four is in some way a relationship that exits between different numbers, a relationship that we could call brotherhood, then we can say that the brother of two is six, and that the brother of three is ....   

            If the number ‘seven’ popped into your head, congratulations,  you just built a chord.  To bring this abstraction to music, we will use the layout of the piano.

            The piano is the most logically and simply laid out instrument in the world.  It is every concept of music theory, every note in western music, laid out in an evenly spaced order, from low to high.  There are 88 different keys on a piano, which can make it seem quite intimidating.  If I had never seen a piano, it would look like box with 88 different random buttons to press, the outcome of which was unknown.  Yet if I started to try different keys at random, I would soon discover that they have some kind of basic order.  The further the keys are to the left, the lower they sound;  the further the keys are to the right, the higher they sound.  Upon further exploration, if I were to start at the very bottom of the piano, all the way to the left, and begin to play every white key all the way to the top, I will only have played 7 different notes.  While this seems to defy logic as all keys on a piano produce a different sound, especially if played by someone who has never touched a piano, the reality is that there are patterns within the keys that repeat themselves, and thus allow for this to be true.  Before this can be fully explained, I need to define some terms.

            When I use the word ‘note’ I must distinguish it from pitch.  Pitch is relative to the frequency of a sound.  If the pitch goes up, than it is the frequency of that pitch that is getting higher and higher.  If by chance I play a note on the piano named A, I will find that it produces a pitch of 440 Hz, which translates into a frequency of 440 vibrations per second.  There is no other key on the piano that produces that same pitch.  Yet there is another key named A’, seven keys above the one that I just played.  If I play A’ and measure its frequency, I find it to be 880Hz, exactly double that of the original note.  The human brain registers this similarity, and the result is that the two pitches sound remarkably similar.  They sound like the same note.  Music theory defines this relationship between two different pitches that are the same note, as an octave.

            Thus it is possible that if I play every white key on the piano from bottom to top, I am only playing seven different notes over a great range of pitches.  I might also notice that for all of the white keys of the piano, if they are played in order, there is some kind of repetition that occurs every seven keys.  It is this repetition, this sequence of seven different notes on the piano before I reach the octave of the note I started on, that is called a scale.

            Suppose that we were to play the below piano.  If we start on note C in figure 2.1 and play the preceding 7 keys, we will have arrived on note C’. This C’ is an octave above C.

            What makes the piano ‘the most logically and simply laid out instrument in the world,’ is in its consistency.  If we look at both notes C and C’ and split the keyboard apart to the left of each note ...

... we find that visually, the two breaks are identical. The pattern of black keys is maintained, as is the number of white keys.  Each section has seven white keys which make up one scale, called a major scale. What makes the scale major rather than minor, diminished, or even blues, is beyond the scope of this paper.  What I can say is that it has to do with the number of black keys between the white keys.   Luckily though, a mbira only has seven different notes, so in effect there are no black keys to worry about.

            The notes on a piano are conventionally named with letters, such as those we have used so far.  As I have mentioned however, music is built upon relationships that can be easily described by numbers.  Therefore, for the rest of this paper, I will talk of notes and more accurately, keys, in terms of numbers rather than letters. Additionally, as will soon be seen, letters can not be applied to mbira without a great deal of confusion, for different mbiras are built out of different scales.  The pitches of the same keys on two different mbiras are not always the same.

            So instead of calling the note and key C, “C”,  we will simply call it key 1, as it is at the bottom of the major scale.  Going up from key 1,  we count to seven, at which point we return to 1.  This new convention is shown in figure 2.3.


            As was said in the beginning of this chapter, there exists a relationship between keys built on a distance of four called a chord.  Thus if we look on the piano above, and press the first key, key 1, and then we count up four keys, and press key 5, we will playing a chord that is built on key 1.  Since we know that notes repeat at the octave, we can also press the key 7, then count up four, and press key 4’.  We have constructed a chord that is built on key 7.  Note, however, that at the beginning of the chapter, it was said that 7 and 3 were ‘brothers.’  This seems to go against the chord that we just built.  The difference is that the chord we just built has key 7 as its base, called the root of the chord.  It was from key 7 that we counted four to find key 4’.  If we play the keys, 3 and 7, the root of the chord is key 3, for key 7 is four keys above 3.  3&7  and  7&4 are different chords, even though they share a key.   We can play key 4 (an octave below key 4’) and then key 7, and the chord is still a ‘7 chord’, but it is said to be inverted, because the root, key 7, is not at the bottom. Like wise, if we play key 1 and then key 4, we have built an inverted ‘4 chord’.

            Chords are the foundations of harmony.  Harmony is a word that simply refers to the relationship between keys and notes.  As you have probably inferred from the word itself, you can not have harmony if you are only playing one key.  Two or more keys played together create a harmony.  The distance of four is the harmony of the chords that we have just learned about.  Four is the relationship between the keys. While some forms of music such as jazz have harmonies built upon as many as 7 different notes, the harmonies of mbira is constructed only out of two notes, the root and the fifth.  The chord can be in root position, when the root of the chord is at the bottom, or they can be inverted.  This kind of two part harmony built on fifths is called dyadic harmony. (Goddard)

            I need to introduce a few more concepts before getting into the structure of mbira music.  The first is that of rhythm.  If I am  playing, notes, keys, or chords, or am just scratching my head at a steady pace, then the relationship of each instance that I do what ever I am doing, over time, is the defined as rhythm.  Harmony is built on a relationship of frequency.  Rhythm is built on a relationship of time.

            We can divide time into equal pulses.  Say, for example, we use one second as one pulse.  If we look at a watch, the second hand clicks at every pulse, or beat.  The majority of western music is built on a beat of four.  That is, the variations in the events, scratches, or whatever, will make some kind of repetition every four pulses.  This set of four beats that repeat is usually called a measure; in mbira music, however it is referred to as a phrase.  In chapter 4, the rhythms found in mbira music will be represented in a visual format.  To make this concept of visualizing rhythm familiar, the following diagram is a representation of a steady rhythm, repeating ever four beats, or equal divisions of time.

            We already know a scale as being a linear progression of seven notes which can then repeat itself at the next octave.  If, as in the case with the piano, we lay out a few of these scales in a linear fashion, we have a large palate of notes from which to choose. Those that we choose, when played one after the other, will create a melody.  A melody can be defined as the movement of pitch over time.  The melody forms a rhythm as well as creates a movement of pitch, called a tonal movement.   For example, if I start at the top of a scale, at key 7,  and then go down to key 1 and beyond to the lower octaves, I have created a descending melody, with a descending tonal movement.  Similarly, if I start on key 1 and play up the scale till I reach key 7, I have created an ascending melody, with ascending tonal movement.  Melodies are not constricted, however, to only linear ascending or descending order.  They can move like rivers, or they can be as random as noise.  The song ‘happy birthday’ is a melody, and so is song ‘happy birthday’ played backwards.

            If we have two different melodies being played at the same time, one being played on a saxophone and the other on a trumpet, and we were to freeze time as the two musicians were playing and write down the note coming from each instrument at that moment in time...  the two notes that we wrote down would have some kind of harmonic relationship.  If the sax player at a given moment, is playing a 1 key, and the trumpet player is playing a 5 key, then at that moment they are creating a diatonic harmony.  

            Look at the following example:

Sax player’s keys :         5 6 7 7   5 6 7 7   5 6 7 1’  5 6 7 1’

Trumpet Player’s keys :             1 2 3 3   1 2 3 3   1 2 3 4   1 2 3 4

            The harmony throughout the entire song is diatonic.  There is a distance of 4 between all of the keys.  This is true even of the last chord.  Looking at the numbers only, it seems that there is only a distance of three between the keys 1’ and 4.  Yet the 1’ key can be treated as if it were an 8.  A 1’ key is one key past 7.  If I were to rewrite the same example and replace all 1’ keys with 1 keys (a key an octave lower), then the final chord would read 1 & 4.  Now there really is a distance of only three between the keys, yet it is still a chord with a root of 4.  It is only that the chord is inverted.  This relationship of 3 is the easiest way to spot inverted chords.

            All the chords in the above example are grouped into sets of four.  As mentioned already, the majority of western music is usually built on patterns that repeat every four beats.  In the example above, the repetition can be seen in the movement of the chords over time.  Each set of chords that makes this repetition is called a phrase.  Note also that the first phrase is identical to the 2nd phrase, and that the 3rd and 4th phrases are only slightly altered from the 1st and 2nd.

            The concepts of notes, keys, scale, rhythm, melody, and harmony, are the foundations of music theory.  They are tools necessary in the exploration of the game of mbira.

Chapter III

Learning the Game of Mbira

            I was once told the story of a young boy who had become bed ridden by a strange illness. Even after speaking to numerous doctors, his family could still not uncover the source of his ailments.  Finally, they decided to consult the spirits to see if they could be of help.  They hired a professional spirit medium[2] to try and contact the sprit world.  When the ceremony was finished, the spirits had told the boy’s father to buy an mbira for the boy to play.  A day later, the father returned with an mbira his brother had made.  It was given to the boy, and within minutes, music was pouring from the keys of his mbira.  His illness soon disappeared.  The following day the boy was possessed with the Spirit of Monomotapa, one of the first rulers of the Shona empire.

            There is a great variety in the ways in which the music comes to the people of Zimbabwe.  Some, it seems, are been born solely for the purpose of playing mbira.  Others, finding themselves interested in the music, consciously decide to learn the game. 

            Traditionally, mbira was taught by indirect means.  Young boys would grow up hearing the music in ceremonies, in the fields, and at home.  They would learn to sing the songs that accompany the mbira, and then would try and remember the melodies coming from the instrument itself.  When the mbira was left unattended, they would sneak in, and play the song from memory, trying different keys until they discovered the ones that sounded right. (Mungate, 11/27)

            In more recent times, mbira has been taught by professional players who either teach for money, or for trade of goods or services.  As such, the teaching is more direct.  They break the songs down, key by key, chord by chord, and phrase by phrase, playing each note slowly until the student can repeat the pattern.  The student may only learn a single phrase in a day and then later build the phrases together to complete the song, or he may learn as many songs as he can remember. 

            I had two months in Zimbabwe, in which I played mbira.  As such, I wanted to learn as much as I could about mbira in the time allowed.  Therefore, in each lesson I wanted to be able to gain and retain as much information about the instrument as I could in the time allowed.   I also did not want to forget all that I had learnt after returning to the United States.  Because of this, I had to develop some system of writing that allowed me to record the songs and patterns that I learnt, a system of notation.

            In western music, there is a general consistency among instruments.  If I have a piano with which I am well aquatinted, and I go to a friends’ house who also has a piano, it is very likely that if I play a song that I just wrote on my piano, it will sound the same on my friends’.  All pianos are tuned to the same pitches.  If I play key 1 on one piano, and key 1 on my friends’ they will be the same pitch  The same goes for all other western instruments, from violins to tubas.  It is this consistency which allows us to have a system of notation in western music, called staff notation, that is based on the pitch of the key being played.  If we see the pitch written, then we know which key will produce that pitch.

            As I have mentioned in the previous chapter, the mbira is not such a consistent instrument.  Every Mbira dza vadzimu has a similar layout to the one shown in figure 1.1.  Some are missing one or two keys, or some may have one or two extra.  Yet the arrangement of these keys is always in the same shape.  There are always three manuals, forming a sort of ‘v’.  The inconsistency comes in the differences in pitch of the same key on different mbiras.   I once saw a set of three mbiras that were each tuned an octave apart.  If I were to were to play the same key on each mbira, I would find that each key I played produced a different pitches.  These mbiras were however, designed to be played together.  Though the pitches were different, the notes were still the same, as they were an octave apart.  Other mbiras come in entirely different tunings: the relationship between the pitches across the keyboard, from key to key.   Some are built on major scales, some are built on minor scales.  Because of this, a song I write on one mbira will not necessarily sound the same if I play it on another mbira.  However, though there are differences in the tunings of different mbiras, the song played on them will always sound similar, because the harmonic relationships between the keys is maintained, even though the pitches are not.  In other words, if I play key 1 and then key 5 on one mbira, to discover that the two keys produce a dyadic chord, and then I play key 1 and key 5 on another mbira...  I will have made a dyadic chord on that mbira as well, even though the pitches of the two chords from the two different mbiras may be different. 

            What this means is that if a song is learnt on one mbira, by the order in which the keys are played, the song can be played on any other mbira, regardless of the mbiras’ tuning.  The song might sound different, but it will still sound good.  Because of this, the system of notation has to be based on the order of the keys played rather than the pitches that they produce. 

            To develop such a system, every key on the mbira must have a number.  Rather than numbering each key singularly however, as there can be as many as 25 different keys on a given mbira, it is much easier to group the keys together.  As there are three different manuals on the mbira, we can group the keys according to the manual they are on.

            In Figure 3.1, there are three rows of keys, numbered from the center of the mbira, outwards.  The keys are numbered in this order because the further the keys are from the center of the instrument, the higher their pitches are.  This works well, yet the system of notation still has to distinguish between the three rows.  The keyboard is first split between the left set of manuals the right manual, as they are played by different hands.  The keys played by the right hand will be written on the top line of our notation and the keys played by the left hand will be written on the bottom line.  Though there are two manuals on the left side, there is only one thumb responsible for playing them, and therefore only one key can be played at a time.  Therefor both can be written on the same line of notation, provided that each manual is somehow distinguished.  This will be done by writing the numbers of lower manual using a boldfontIn practice, I typically circle these numbers from the bottom left manual when writing by hand.

            Figure 3.2 is a notation of a single phrase of Kariga Mombe, one of the oldest and most widely known shona songs.

            As the top row of the notation represents the right side of the mbira, the first 3 of that row refers to the 3rd key from the center on the right manual.  The bold 8 on the bottom row refers to the left most key on the bottom left manual.  The following 4, is a key in the upper right manual.   The bars at both ends of the numbers designate the section as being one phrase a song.

            Songs in western music always have a start, some kind of macro structure such as verses or bridges, and an ending.  From pop songs to symphonies, there is always some kind of linear progression that carries the song.  Mbira music is different in that each song has neither a beginning nor an end.  The songs are cyclical in nature.  The end is the beginning, and the beginning occurs at no particular point.  Thus, if I learn a song that is built on the pattern :  1 2 3 3   5 4 3 3,

it is possible that you know a song built on the pattern:  3 3 5 4   3 3 1 2.

They are the same song.  The song is a sequence of notes that repeats infinitely.  If you start the sequence from ‘3 3 5 4,’  all that I have to do is wait until you play ‘1 2’  before I come into the song from my ‘beginning.’   Once we have ‘locked’ our parts, the song carries on as one. 

            In term of a more macroscopic structure, mbira songs are also not linear.  To better explain this, first look at the structure of folk songs found in the United States and other parts of the world. A Folk song usually begins with an instrumental introduction.  Then, the singing begins.  The musician might sing a verse or two, then go to a chorus, and then go to a bridge.  Eventually, there is some sort of ending.  Similarly, symphonies have even more complex macroscopic structures.   They are often broken down into different movements, some of which wander into different keys, using different scales, and time signatures[3]...    So when I say that mbira music is not linear, I am saying this because the harmonic movement of a song is the same for the entire song.  It is the harmonic movement that defines a song.   The music is a state of mind that the player enters, like a trance, or a hypnosis.  And the further into the ‘song’ the player gets, the deeper the trance becomes.

            Though the patterns that make up mbira songs remain the same through the entire experience, there is a more macroscopic structure to the music, this ‘deepening of the trance.’   It was broken down into four stages by Dumisani Maraire[4].

Stage 1 (Preparation): -  Here, the main harmonies of the piece are played together as chords, to give the player a ‘feel’ for the piece.  This is treated as a warming up stage.

Stage 2 (Basic pattern) : -  Now he breaks into the piece’s basic pattern, repeating it over and over again, listening to the many sounds it gives him.

Stage 3 (Climax) : -  Building on the various sounds that he hears from the basic pattern, he develops the piece by adding ideas of his own, subtly varying the melody, harmony and rhythm as well as responding vocally to the voices coming from the interaction of all these new inputs. The piece builds to a climax of expression and intensity where the basic melody as well as the elements of development and variation which he has introduced, interlock in an intricate polyphony[5].

Stage 4 (Basic Pattern) : -  After all that, never mind how long it takes, the mbira player must come to the end of the piece.  He begins to eliminate elements from the complex texture he built up during the climax, until he is left once again with the basic pattern.  This happens gradually since a player who hurries this process is being “rude to the piece.”  The player can end the piece as soon as he wants now, and usually he will just let the basic pattern disintegrate so that the voices merely fade out.

            Because the chords of an mbira song stay the same throughout, this ‘climax’ of the song come from other types of variation in the music.

            A chord is a relationship between keys.  If two keys make a chord then the chord is formed by their relationship within the scale that they are notes of.  When you go up an octave from any given note, you find the same note.  Therefore if you take one note of a chord and play it 5 octaves above the original note, you will still be playing the same chord, but you would have changed its harmonic structure.

            In Mbira music the songs are varied by playing different notes that still fit the same chord at a given time in the music.  These changes are called variations of a song.  A song, like Kariga Mombe, usually has some pattern forms the standard for the song.  It represents its foundation and it is from this foundation, this base pattern, that other variations are made.  In shona, this base part is called the kushaura.   Figure 3.3 is the kushaura of Kariga Mombe.

            As can be seen in figure 3.3, the song is made up of four different phrases, each of which appear to have 6 beats, or pulses.  In actuality, there are twelve pulses in each phrase as we will soon see.  Every written note has a space after it, which in this case is a pulse in which no note is played.   Thus mbira songs can be divided into 48 pulses, as there are 4 phrases with 12 pulses in each. 

            The reason that I use the word pulse rather than beat is that in mbira music, just as there is no beginning or end to a song, so there is no main beat.  The dancers may stomp at once pulse, while they might clap at another.  And the mbira music might have its own ‘beat.’   I mentioned that the songs can have the freedom to begin at any point in the cycle.  It is this freedom between the parts that gives another way for the music to be varied.  By changing the accents of the music, or how hard one plays certain notes within the cycle, an mbira player can create of feeling of a beat, which can move across the pattern, even though the pattern remains the same.  He can shift the feel of the music, so that he gives the implication of starting at one point in the pattern, perhaps by accenting the beat, and then can move that accent to another point in the cycle.

            When the variations are played, they are usually of rhythmic as well as harmonic nature.  However, before I can get into this we need to look more into the structure of the mbira keyboard.

            The piano which we have thus far been using to take a look at music theory, chord structure...  everything...  will soon very clearly seem like the ‘most logical instrument in the world’. 

            The Mbira is often referred to as a thumb piano.  This may seem to be a fit description, yet I have the bias that it greatly undermines the instrument.  After studying mbira for 2 months in Zimbabwe, I will return to happily play my school’s Steinway thumb mbira back home. 

            The piano is so logical because it is completely linear.  The further to the right on plays, the higher the pitches get.  I already mentioned the that mbira is different because the lowest notes on each manual are located at the center of the keyboard.  It took me some time to get used to this difference, as I grew up playing a piano, but it was not long before I simply learned to switch the direction of my left hand.  The differences and complexities of the instrument go much further.  I think that the easiest and most painless way is to jump right into it.

            Where as on a piano, you play a scale left to right across the entire keyboard, on the mbira you must weave a complex pattern if you are play the scale in order.  If you are intent on really learning the mbira, this is a pattern well worth learning, but for the purpose of this chapter it is fine to just see and recognize its existence.  The figure 3.4 is a diagram of a piano and the path in which one must play the keys to play an ascending scale.  Figure 3.5 is a diagram of the mbira, and a path, or better, a thread, that weaves across the instrument.  It is the movement of this thread that plays the 1234567 1’2’3’4’ ... scale on the mbira.  It is this pattern that is the linear scale in the game of mbira.

            If you start on the lowest key of the mbira, and play up the scale to the highest key, this is the order  in which you must play the keys.  This is the logic of the thumb piano.  When the solid line jumps across the keyboard for the last time, where it branches out into the dotted lines, it is one of three paths that cam be taken at this point, the others being the two dotted lines.  This occurs because some of the keys on the instrument are doubled, in that they have the same pitch.

            While this linearization of a non linear instrument might to do little more than create confusion, or a fear of the mbira, it is useful in further analyzing the instrument, not through the path of a scale on the instrument, but through the patterns that emerge from this path.  I have played mbira for two months, and the path presented in figure 3.5 still bends my mind.  Yet over time, I have learned to use pieces of it to guide my playing.   The patterns that will emerge from it, will emerge if we replace the line and dots with numbers, the numbers of the scale.   Thus the first key, where the path starts, will be key 1, and all others along the path will be numbered accordingly.  When the scale reaches key 7, it will begin again with key 1.   Figure 3.6 is a mapping of all these notes of the scale, called scale degrees[6], across the entire keyboard[7].

Figure 3.6 :   Scale Degrees on the Mbira Dza Vadzimu

            In figure 3.6, we can count up the scale degrees if we trace the same path across the keyboard as in figure 3.5.  This new diagram makes it no less confusing, or perhaps even more since the ‘path’ is now gone.  But simplification comes if every key is matched will all others that are of the same scale degree, a mapping of the octaves across the mbira. 

            I will do this mapping only on the left side of the keyboard, as trying to connect keys across the entire keyboard, between right and left manuals, would be as cryptic as tracing the scale.  The two left manuals, while not sharing the same layout in terms of the order of the scale degrees in them, do put the octaves in close range of each other.  There is a closer logic to the arrangement of the keys.  Looking again at figure 3.6, notice that both the rightmost keys on the left manuals are of scale degree 1.  The leftmost keys on both manuals are of scale degree 2.  As might be expected, from there, it gets a bit more confusing.  The octave of scale degree 3 on the lower left manual, for example, is the first key on the right side of the keyboard.  But this, luckily, is the most confusing one.

            While figure 3.7 might not seem significantly less confusing, it is far easier to memorize octave pairs than it is to figure them out from the order presented in figure 3.5.  The octaves of the mbira form a kind of visual relationship.  For example, if you start on octave pair 7, and go down to 6, 5, and then 4,  notice that there is a pattern in the lines drawn between the keys.  By pattern, I do not mean something that necessarily repeats, but rather a visual form that can be memorized.  Look at key pairs 7, and the connection between them.  Then look at key pair 6.  The connection is the same for both.  They keys are straight down from each other.  Now look at pairs 5 and 4.  They form a cross; they are what one might call reversed pairs.  Thus if one were to play down the scale, from 7 to 4, and also want to include the octaves, he could think in terms of a visual pattern governed by the concepts :  straight, straight, crossed, crossed.

            I have never used the words straight or crossed when playing mbira.  Rather, I have linked the keys by their visual[8] relationship to one another.  There are of course no lines across the octaves on the mbira, but the relationships between the keys form lines. Therefore when I use the words straight and crossed, I use them to explain the concept of linking keys, rather than a rule by which one must do it.

            The octave is the most fundamental relationship between keys on the mbira.  The patterns from the octaves create a frame work out of which all other harmonies can be built.  As mentioned, mbira music is built on dyadic harmony, a harmony of roots and fifths.  Therefore, the seven octave pairs[9] are the roots of all the chords on the mbira.  The fifths of these chords can be found if we look at the figure 3.6.  For example, the fifths of chord 1 are any given key that is of scale degree 5, and all of its octave pairs.  The fifths of the 4 chord are all keys that are of scale degree 1.  Thus, each octave on the mbira creates a base, unto which, keys pairs of that octave’s fifth can be attached.  A chord built on key 1 has all keys of scale degree 1 as the roots of the chord.  These roots are octave pairs as drawn in figure 3.7.  The fifths are all octave pairs of scale degree 5.  Thus the mental process of learning dyadic chords on the mbira is done by attaching octave pairs together that are four notes apart.  A mapping of all of these chords, across the mbira, can be found in Appendix A.

            Because the system of notation used in this paper is based on the physical order of the keys rather than their scale degrees, it becomes difficult to analyze the music harmonically from the notation.    The figure 3.8 is a chart which shows the octave pairs and their fifths, written out in Nor notation, rather than the numbers corresponding to each keys scale degree.   Because the chart is based on key position, the logic of the chords is lost in the cryptic layout of the mbira.   Therefore, it is a tool by which one can analyze the music when written out in Nor notation.

Figure 3.8         Chords on the mbira dza vadzimu written in Nor notation

Scale Degree of Root

Roots on Left Manuals

Fifths on Left Manuals

Roots on Right Manual

Fifths on Right



1 , 1 , 6

5 , 2

2 , 9



2 , 8 , 7

6 , 4

3 , 10




7 , 5

4 , 1[10]



4 , 3

1 , 1 , 6


9 , 2


5 , 2

2 , 8 , 7


10 , 3


6 , 4



4 , 1


7 , 5

4 , 3



            Yet, the mbira is not all that cryptic. Both the bottom left and right manuals are arranged in a linear fashion, with the exception of one key in each.  This can be seen firstly in the roots of the left manual, whose scale degrees all correspond exactly to their positions on the keyboard.  On the right manual the correlation is nearly perfect, except that there is an extra key that does not belong by normal logic, key 1 on the right manual.  After it, however, all the keys on the right manual are in order.  Key 2, is scale degree 1, key 3 is scale degree 2, etc...    Thus they are all shifted, but still subscribe to a linear logic.  As I mentioned, the chords on mbira are built by connecting octave pairs.  This can be seen the chart. Each box represents an octave grouping on the particular side of the keyboard.  Therefore the grouping that makes up the roots of scale degree 1 on the left manual, also makes up the fifth of scale degree 4.  The octave pairs are reused.   The most common variation on the song Kariga Mombe is built on octave pairs.  In figure 3.9, it is written using Nor notation.  

            This is a variation of the kushaura part of Kariga Mombe.  These variations are called kutsinhira.  I will talk of the relationship between kushaura and kusinira parts in mbira music a great deal more in chapter 4.  For now, notice the differences between the two parts in terms of the rhythmic structure.  In the kusinhira part, all the keys played by the right hand have been shifted over one pulse, resulting in the alternation between right and left hands.

            The bottom row of the first phrase has the following keys :  1 , 2 , 5 , 7 , 2 , 5.    Referring to the figure 3.8, we can begin to analyze these keys to figure out what the chords are of the first phrase.  All keys,  1 , 2 , 5 , 7 , 2 , 5 are on the left manual, as they are from the bottom row of in the notation.

            To start, we look up the first key, key 1.   In the chart we find that key 1 is either the root of a 1 chord, or it is the fifth of a 4 chord.   Keeping that in mind, we look up key 2 .. .  key 2 is either the root of a 5 chord or the fifth of a 1 chord.   Thus, the first two keys can be notes in the following chords :

            Key 1 : chords 1 or 4.

            Key 2 : chords 5 or 1.

            Both keys are notes of a 1 chord.   We can say that the first chord of Kariga Mombe is a 1 chord.

            Notice, as I mentioned, that there are 12 pulses in each phrase of the song.  Chords usually change every 4 pulses in mbira music, so this is the guideline by which I can assume that these are the only two notes of the first chord.  Key 1 falls on pulse 1, and key 2 falls on pulse 3.  Key 5 falls on pulse 5, so it is ‘out of range’ of the first chord.

            Moving on to the next chord, we have the keys 5 and 7.  If we look in the chart, we see that these keys are octave pairs.  Therefore we can assume, for now, that they are the roots of a 7 chord, as can be seen in the chart.  The same goes for keys 2 and 5, which we can assume are the roots of a 5 chord.

From the analysis just done on the keys of the left manual we got the following chords : 1 , 7 , 5.

            You probably noticed that I said that we ‘can assume, for now’ that this is true.  The problem is that our analysis of the song is only based on one note for the 2nd and 3rd chords.  In the keys of the first chord, we found a root and fifth and in music based on dyadic harmony, this is the entire chord.  In the other chords, however, the keys are octave pairs, thus, without the fifths of these key, we can not be sure if they are indeed the roots of the chords or if they are the fifths of some other chord.  Therefore, we must look at the top line of the song, look at the keys played by the right hand.

            We can skip the first chord as we know its fifth, and start on chord two.  The keys from the top row for the 2nd and 3rd chords are : k,k, 3 , 3.

            k  means that both keys 1 and 4 are played together on the right manual.  They are often played together as they are octave pairs.   Using this symbol, therefore, is much easier than writing another row in the notation for this 1 pair.

            According to the chart, keys 1 and 4 are either the roots of a 3 chord, or the fifths of a 6 chord.   In our analysis of this chord from the left manual, we assumed that it was a 7 chord.  Something is wrong.  If we again look up keys 5 and 7 in the chart, we see that they are indeed the roots of a 7 chord, but also the 5th’s of a 3 chord.  Thus chord two is a 3 chord.  More specifically, it is an inverted 3 chord because the fifth is being played by the left hand.  This is usually referred to as a fifth being played in the bass.  If we analyze the 3rd chord, we find that key 3 in the right hand is indeed the 5th of a 5 chord.  Thus our new chord progression for the first phrase of Kariga Mombe is : 1 , i3 , 5.  The ‘i’ designates the inversion.

            If we analyzed the entire song in such a manner, we would find these to be the chords of Kariga Mombe:    1 , i3 , 5                1 , i3 , 6           1 , 4 , 6                        2 , 4 , 6 

            Such an analysis can be carried out for any song in order to learn the chords of the song.  From these chords, one can build the song simply by picking keys from the chart that make up a given chord.  In other words, with the chords we found, we can reverse the process, perhaps to test a variation that someone wrote, to see if the keys of the variation are correct against the chords of the song.  This is a useful process, as it allows for the writing of completely new parts from the chord progression.  I therefore came up with a way of notetating the chords of mbira songs.  In the figure at right, which has probably been intruding your thoughts as you read the above, I have mapped out the chords of Kariga Mombe. 

            The song moves in time from left to right.  As there are 48 pulses in a song, and as chords usually span four pulses, there are 12 different divisions of time in the diagram.  Each phrase along this time line, each made of 3 chords, is distinguished with a vertical bar.  On the left side of the diagram, there are the seven scale degrees, the seven chords.  Each square in the diagram is the chord at that moment in the song.  Inverted chords are designated by a hollow square.  The circle represents the fifth of the inverted chord.  As in Kariga Mombe, it is this fifth that is ‘in the bass.’  When a fifth is in the bass, it becomes part of the melody that is played on the left manuals.  The circle helps to trace the melody as it sounds in the song.  Harmonic mappings of other songs can be found in Appendix B.


            An understanding of the way the keyboard works, an understanding of the theory just presented, gives us the tools to analyze, build, and create the logic that is at the core of improvisation.

Chapter IV

Checkmate : The Art of Improvisation on the Mbira

            A game of chess is a form of improvisation.  It is a game governed by specific rules on the movements allowed for certain pieces.   Every knight, for example, can only move in the form of an L.   If his horse ran off, and he had to make his way on foot, the rule governing his movement would take the following form :   You can walk 2 paces in any direction, but then you must walk 1 pace in a direction perpendicular to that in which you just traveled.   At right is one possible shape that describes the knights movement based on this rule. It is a movement that is visual in nature, a set of relationships among the squares of the board that give a structure to each pieces’ allowable movement.  After the knight has moved, the other player can choose to move any of his pieces,  but hopefully with attention to the logic and objectives of the game. 

            The game of chess is an improvisation of movement that occurs within a given framework of structure.  Each piece can be compared to a different chord on the mbira.  The rules governing each chess pieces’ movement are just like the relationships between the keys within a given chord.  In the improvised game of chess, each move takes a visual form like the shaped in figure 4.1.  In the game of mbira, the shapes are similar, though they are spread across rows of keys rather than a two dimensional playing field of squares.  Thus, figure 4.2 could be one move in the game of mbira...

and figure 4.3, the next.

            These ‘moves’ are chords.  Move 1 is chord 4, and move 2 is chord 1.  The moves are played across the entire keyboard.  The left thumb is restricted to play only one note at a time, but it can still build chords by connecting keys in rapid succession.  The figures above are example of the left hand playing in a triplet pattern, a pattern that I will describe later in this chapter.   As there are two fingers available on the left side of the keyboard, two notes of a chord can be played at once. 

            In the rawest form of improvisation, as in the most drunken game of chess, an mbira player can move from chord to chord at random.  There are no rules yet written for mbira, governing the movement of chords.  There are only rules for the selection of keys for a given chord. 

            On the mbira I have experimented with improvisation within the context of various style of music.  From jazz to reggae, western styles can be applied to the mbira by using the structure of the instrument to guide the movement of chords that are characteristic of the given style of music.  The possibilities are endless for this kind of exploration on the mbira.  I have heard one of Bachs’ preludes rendered on the instrument.  I have also heard happy birthday, though I have yet to hear it being played backwards.  For the purpose of this paper, however, I will focus on improvisation over traditional mbira songs, such as Kariga Mombe.

            Since a song like Kariga Mombe has a set chord progression, improvisation can be done by selecting different keys to construct the chord at a particular point in the song.  This process might not seem like it gives a great deal of freedom to the music, as the chords in mbira music are fairly simple.  They are constructed only out of the roots and fifths of a given key and therefore there are only 2 different notes to choose from for a given chord: the note of the keys that are the fifths and the note of the keys that are the roots.  If we look at a chord, however, say chord 1, we find that there is a great deal of variation allowed.

            There are a maximum of 4 pulses over which keys of chord 1 can be played, as chords usually change every 4 pulses.  As we are selecting keys to play at each pulse, I will, for now, call each pulse a slot.  Then, for chord 1 we have, at most, four slots that need to be filled with keys.  This is assuming that only the left hand is playing.  One or two fingers of the right hand could play in each slot as well, for a total of 8 or 12 ‘slots’ for a 4 pulse chord.  For simplicity, however, I am going to remain with only a look at what the left hand is doing at this point of improvisation over chord 1.

            If we look at the keys of chord 1 in figure 3.8, we find that there are 5 keys on the left side of the keyboard that are either the roots or fifths of the chord.  Thus for each slot we can choose at random, any of these 5 keys to fill our chord.  In addition to playing one of these five keys, we can also choose to not play any key,  to leave the slot blank.  Thus, there are a total of 6 possibilities at each of the 4 slots.

            The number of variations of chord 1 can be calculated by some simple math : 6 to the power of 4, which equals 1,296.  There are 1,296 different ways that chord 1 can be played by the left thumb over 4 pulses of a given song.  A progression of two chords, a progression 8 pulses long, can be played in over 1.5 million different ways.     With a computer powerful enough, one could calculate every variation ever written for a given song, as well as every variation ever to be written.  Within the chord progression of a given song, there is clearly an vast amount of freedom over which a player can improvise. 

            Along with harmonic improvisation over a given chord, the improvisation of the rhythms that the keys of each chord are played in gives another form of freedom to  music.  A player can indeed play keys on all four pulses of a chord,  but mbira music is typically created using more intricate rhythmic patterns. I have already mentioned the kushaura and kutsinhira parts of mbira songs, but I have yet to explain what the relationship between the two really is.

            In traditional mbira music, when two or more players play together they each play a variation that is both harmonically and rhythmically different from the other one.  The two examples that we have seen so far are the kushaura and kutsinhira parts of Kariga Mombe.  They appear again, on the next page, written out in Nor notation.

            Compare the two parts above.  The kushaura part is fairly simple.  Both left and right hands play together through the entire variation.  All keys in the left hand, expect one, are played on the upper left manual.  There is little movement done across the entire keyboard by either hand.  The kutshinira is a bit more complex.  It is a based on the kushaura part, as it shares many of the same keys, but its rhythm is altered.  The right hand plays one pulse after (behind) the left.  Additionally, both right and left hands explore other keys in each chord. 

            If we map out the rhythms of each hand we can see a clear comparison of the rhythms of the two songs.

            This simple rhythmic analysis between right and left hands can be done easily from any song written out in Nor notation.  In the above case, the top line has simply been shifted to the right by one pulse.  This kind of pattern is called an alternating pattern.  There are many more rhythmic patterns typical to both traditional and improvised mbira music.  These rhythms can be found in appendix C.  One of these rhythms is the triplet pattern mentioned earlier in this chapter.  The triplet pattern is useful for building chords in the left hand, as it allows three keys to be played together as depicted in figures 4.2 and 4.3.  Figure 4.7 is a rhythmic analysis of the triplet pattern.

            The set of three pulses played by the left hand allows for the construction of three key (voice) chords..  The following is an example of a variation I composed using the system of analysis presented in chapter 3 and the rhythmic pattern in figure 4.7.

            The above variation could be played as the kutshinira part of Kariga Mombe.  Traditionally, this would likely be the case, as it is a more rhythmically and harmonically complex variation than the kushaursa part as written in figure 4.4.  However the relationship between kushaura and kutshinira parts is not necessarily one of complexity.  It has to do with the relationship and function of the two parts within the song.

            The kushaura part is considered the ‘base’ of the song.  It has the function of provide a rhythmic and harmonic base over which exploration can made by the other player.  Therefore, the rhythm is usually steady and the harmonic progression of the song is maintained.  In the game of mbira improvisation, this forms a base ‘around’ which the improvisation can occur.  I use the word ‘around’, for the kutshinira part has freedom to break away from the chords, as well as the rhythmic structure of the song. 

            As described by Jonathan, the kutsinhira part has the duty to ‘explore the missing dimensions.’   The ‘dimensions’ are the missing keys of a given chord that were not played by the kushaura part.  Typically, the kutshinira part falls one pulse behind the kushaura.  Thus if one player was playing the variation of Kariga Mombe shown in figure, 4.8, the other player, playing kutshinira, could play the exact same variation except delay it by one pulse.  Figure 4.9 is a rhythmic analysis between these two parts.  The top rhythm is that of the kushaura part, and the bottom one is the kutsinhira part.

            The two are identical except for the shift in the rhythm by one pulse to the right.  Aside from creating a beautiful rhythmic texture, this shift, in the world of improvisation, allows the player who is playing kutshinira to hear the variations of the kushura part one pulse ahead of his own.  From his perspective, he can hear into the future, and react accordingly.  This reaction is the filling of the ‘missing dimensions’.  

            Two mbira players are improvising over Kariga Mombe.  Player 1 is playing kushaura, and player 2 is playing kutshinira.  Player 2 is therefore playing one pulse behind player 1.  Suppose that at a particular moment in their improvisation over Kariga Mombe, the approaching chord is chord 1.  Player 1 is the first to begin to play keys of this chord, as he is one pulse ahead of player 2.  When he plays the chord, and constructs it out of the keys that he chooses, player 2 has one pulse in which he can react to each of the keys played by player 1.  He can choose to play the same key, or he can fill the chord with other keys not played by player 1.  This is the fundamental relationship between kushaura and kutshinira.  It was described to me as the following:

            “When playing kutshinira, it is like you are chasing the other player up a ladder.  You are always one step behind him...  you know how he going about the game.” (Mungate, 11/20)

            Ultimately, however, the kutsinhira is not restricted to remaining one pulse behind the kushaura in the world of improvisation.  This is what I meant when I used the word ‘around’.  I have played kushaura and seen the kutsinhira part jump ahead of my own.  I have seen it remain on a single chord through an entire cycle of the song, playing the keys of the chord as if they were drums.  I have seen it follow behind with a reggae beat, altering the entire feel of the traditional song I was playing. 

            Ultimately, the game of improvisation has no rules.  It has structures which help to establish a relationship between two players.  But once this relationship is formed, it is the relationship itself that guides the music, rather than any structure.  In the end, improvisation is about the sounds heard inside, and the exploration of these sounds.  The game is about the path that these sounds must take to be produced on the keyboard, a path greatly distorted by the instruments complexity.  

            Look at the following pattern for mbira, written out in Nor notation.

1 2 3 4

It is clearly an easy pattern to learn.  One simply plays up the keyboard from key 1 to key 4.   However, if we compare the pattern of the keys being played with the melody they produce, they do not match.

                  order of the keys                        melody                . 

            At left is the input, at right, the output.   It is the instrument itself that is creating a complexity in the music.

The keys 1 , 2 , 3 , 4,  produce the pitches 1 , 5 , 4 , 6.  Therefore the complexity inherent within the instrument adds to the complexity of the music, even when a pattern as simple as that described above is played.  When someone is playing the visual patterns of the songs, the songs as they are written in Nor notation, the instrument gives to the music in that it increases the complexity of the melody created.

            Suppose, however, we reverse the situation and wanted to play the pitches 1 , 2 , 3 , 4.  This is a very simple melody, as simple as the pattern of keys played in the previous example.  If we look at the keys which must be played to produce this pattern, written out in Nor notation, we find the following progression.


            To play the first 4 notes of the scale, one must play at least one key on each of the three manuals on the mbira.  In this example, when we go from melody to pattern, the complexity of the instrument takes from the process.  The simple becomes complex.  To create the melodies heard in his head, the improvising mbira player has to filter these sounds through the complexity of the instrument.  It makes the simple complex, and it makes the complex reside in a whole other realm.  This is the art of improvisation on the mbira.

            All of the pieces that make up improvisation can be analyzed from a theoretical framework, as has been the process of this paper thus far.  This analysis helps to create an understanding of the music.  It breaks down what is going on as the music is created.  But the key to improvisation is not analysis.  Behind the entire process, behind all the patterns, is the music itself.  The music does not come from the logic.  The music comes from creativity, and creativity needs not to be analyzed.  As a good friend once told me, “to master these keys, you must do so not on paper, but through the songs.  Whenever I touch the mbira, I will be learning.”  (Mungate, 11/20)

            All the structures, patterns, connections, and rules I have presented thus far are not those of my creation.  Rather, they have existed for centuries; they are in the songs themselves.

            When a Zimbabwean is going about the game, the result of what he is doing is somehow governed by the rules presented.  But the player himself has never considered these rules.  He simply knows them.  He was born listening to the melodies, to the patterns, to the relationships.  They are as familiar to him as his own family.  They are the inherent order of the mbira. 

            “When I play, I do not think of the notes.  My fingers know where they are.  I hear the melodies in my head, and then they are there.”  (Mungate, 11/24)

            For a man who is a master of mbira, there is no division between kinesthetic, visual, and auditory senses.  The three melt into one process, a process governed by his understanding of the instrument itself.  His creativity flows through the logic of the mbira to emerge singularly, as music.  He is a master of the game.  


            This paper presents a great deal of analytic thought.  It traces my own process of discovery over the two months that I spent, exploring the mbira while in Zimbabwe.  I would like to finish this analysis with one that I made not on the instrument, but on the process of exploring all of the above.  The following is an except from a journal I kept during this two month period.

            “My journey into the mbira began with the implication of my previous knowledge of piano onto the instrument.  It started with an exploration of my knowledge itself, and a transformation of my ability onto another medium.  In the beginning I started to figure out the instrument, to learn how to relate to it physically, to let go, and to go beyond the limitation of playing with 3 fingers instead of 10.

            In the past few days, I have realized that the process has now changed.  Every time that I play, I get closer to the instrument, more familiar with its design.  But where the envelope is being pushed is not in my knowledge of mbira, but in my knowledge and ability within music itself.  The mbira presents a complexity like one I have not previously known.  It is that game of chess moving at a fantastic speed...  one in which each move, each of the thousand possible positions of the approaching chord, has to be calculated in a fraction of a second.   Because I see what is possible, and I understand what is involved theoretically in that which I see,  I can also see the complexity and ingenuity of the players I have encountered.  They are Einstein’s of the mbira.  And I know what is going on in their heads.  But I know it in theory;  I know it on paper.  To be able to play it is an entirely new universe, one that I can enter, but so barely...

            We all have our limits; we can not all be Einstein’s.  If we were computers, and all that mattered was raw processing power...   But, whatever.  Life is good, and I’m groovin in Zimbabwe.” 


Primary Sources

1.      Mungate, Jhonathan, informal interviews conducted from 11/12 - 11/28
1202 Zuruvi Cressent, Harare, Box 5941
Tel: 073-3232

2.      Hall, William, informal interview, 11/20
Tel: 301 519

3.      Chimedza, Albert, informal interview, 12/02
Tel: 498 309

4.      Goddard, Keith, informal interview, 12/03

Secondary Sources

1.       Berliner, Paul F, “The Soul of Mbira”. University of Chicago Press. Chicage, 1981.

2.      Dumisani Abraham Maraire, Booklet accompanying LP record The Mbira Music of Rhodesia, Seattle, 1971, pp. 4-5

Appendix A

The Seven Chords of the Mbira Dza Vadzimu

            The following are mappings of the seven dyadic chords on the mbira dza vadzimu.  The keys are numbered in order, as they would appear in Nor notation, except for the keys making up the chords.  These keys are either numbered with a 1 or 5, designating root or fifth.

Appendix B

Harmonic Analysis of Traditional Mbira Songs



Appendix C


The following are some of the more common rhythms that can be found in mbira music.


[1]Paul Berliners book, The Soul of Mbira, provides a comprehensive look at all of these elements of mbira music.

[2]Spirit Medium : a person possessed with a spirit.

[3]The number of beats per phrase.

[4]Dumisani Abraham Maraire, Booklet accompanying LP record The Mbira Music of Rhodesia, Seattle, 1971, pp. 4-5.

[5] Polophony: the playing of two or more notes at one time.

[6] Ex : Key 1, note 1...  can be called scale degree 1, or the first scale degree.

[7]               Notice that the majority of the bottom left manual is in perfect order.  Thus, if trying to learn mbira for composition or improvisation, this row of keys forms quite a nice base from which to work.  The problem I find is that songs are, in the begining, always taught on the upper left manual, which is the most cryptic of all. Thus the patterns are learnt on this manual, but the notes and there relationship from one another are not understood.

[8]I often use diffenerces in the appearance of key on my mbira such as spots on the metal or rust, to help to keep track of which key is which.  This makes a challange when I play on someone else’s mbira.

[9]There are more than seven pairs of octaves, as can be seen from figure 70, yet because they are only made of seven different scale degrees, it is simply to refer to only seven ‘pairs.’

[10]This note is the octave pair to 3 on the bottom left manual.

Comments are closed.