44 AFRICAN MUSIC SOCIETY JOURNAL

MBIRA MUSIC OF JEGE A. TAPERA

by

ANDREW TRACEY

Jege A. Tapera, when I knew him in 1961, was a clothing factory worker in Bula­
wayo, Southern Rhodesia, but, first and foremost, he held himself to be a murid^i n>em-
bira, an mbira player. The African inhabitants of the Bulawayo area belong mostly to
the Ndebele tribe who do not possess the mbira, but there is also a considerable number
of Shona-speaking people resident in Bulawayo itself. Among these, only a very few
players of the mbira are to be found, perhaps because Bulawayo is well outside Mashona-
land; only two were well-known—Matheu Zvimba, a Zezuru njari player from the
Zvimba reserve near Salisbury, and Tapera, who is of the Nohwe clan of the Zezuru
tribe of the Shona peoples. He was born in about 1905 in the Mangwendi reserve,
fifty miles east of Salisbury in the Mrewa district of Southern Rhodesia, of a Rozvi
father and a Zezuru mother.
When he was a young man of about twenty-five he made a journey from his home
in Mangwendi some 200 miles down to the town of Tete on the Zambezi river in Por­
tuguese East Africa, and there he heard being played the type of mbira known by the
local Sena/Nyungwe people as chisansi or kasansi (i.e. the diminutive form of the word
sansi. The sansi is one of the larger forms of traditional mbiras that are played by the
Sena/Nyungwe). From the evidence of the basic layout of its keyboard, this chisansi
is in fact fundamentally the same instrument as that known throughout a large part of
the Zambezi area of Northern Rhodesia and in Nyasaland as kalimba. It is also found in
the north eastern districts of Southern Rhodesia among those peoples who are adjacent
to the Sena, the Korekore, Budjga and Zezuru, and there it is known as mbira, and
sometimes by the Northern Rhodesian name of kalimba or the Sena name of chisansi
or kasansi. Hugh Tracey recorded several examples of it in this area in 1932-33.

A — 8-key Nyasaland kalimba-, B — Tapera’s 13-key “ kalimba” mbira; C — 25-key Sena-Nyungwe

kalimba. The keys are numbered in the order of their musical pitch from the bottom.

It is reasonably certain that the chisansi is an importation into Southern Rhodesia

of relatively recent occurrence, from the evidence of the variety of names, the distribution
of the instrument, and that which we have by comparison with other Shona-speaking
 MBIRA MUSIC OF JEGE A. TAPERA 45

groups and from the Zezuru of north eastern Southern Rhodesia themselves, that the
mbira dt^amid^imu and madebe d^aMhondoro are their only original mbiras. The chisansi
has not yet spread among the whole Zezuru tribe, being generally confined to the north
eastern side of Salisbury. However, its popularity is still growing: Hugh Tracey has
recently (1958) recorded it being played by a Karanga man and boy in Fort Victoria.
These were samples that could be seen from their manner of construction to have come
down from Northern Rhodesia or Nyasaland, and in this case were brought down by
migrant Nyasa workers. However, it is evident from transcriptions that the music played
on them, and their tuning, was distinctly Shona and not Northern Rhodesian. Unfor­
tunately there is no space in the present article for an analysis of these transcriptions.
The chisansi is certainly not the only mbira to have been introduced into Southern
Pvhodesia from the Zambezi valley. Hugh Tracey has discovered that the njari, now
widely established among all the Shona peoples (with the exception of the Ndau), and
generally considered to be the traditional mbira of the people, was in fact brought into
the Buhera district of Southern Rhodesia in about the middle of the eighteenth century
by the Njanja sub-tribe of the Shona. The two men responsible were Masama and Gotori,
sons of Muroro Gambiza, a half-caste Portuguese trader who had visited and settled in
the area. His two sons went down into the Zambezi valley, to the Nyungwe district of
Tete, and there they found, and learnt the art of, the njari, whose keyboard offered
certain distinct advantages over that of the original Shona mbira d^aMidcjmu. They
returned with it to Buhera. It is said that the clan name ‘Njanja’ is a nickname indicating
the peculiar sound of the njari. Thus the instrument is often known as the 1njari d^a-
Manjanja.
By 1900, the instrument had spread outwards about a hundred miles in all directions
from Buhera, having reached Mrewa, Rusapi, Bikita, Victoria, Chibi, Chilimanzi and
Salisbury. It did not, however, reach Mtoko in the north east until about 1935, and has
not yet arrived in other Shona areas such as the Darwin area further north, the country
east of the Sabi river, or the country south of the Lundi river and the line 20 degrees
south latitude.
Tapera, like others before him, bought his new instrument at Tete, and brought it
back to the Rhodesian highlands. This type of mbira was probably not completely
unknown to him, from the evidence of the several recorded examples of the Kalimba
in the north eastern quarter of Southern Rhodesia, but it was then, and still remains a
young man’s instrument, the njari and madebe d^aMhondoro being the ancient and estab­
lished instruments in his home district.
Before leaving Tete, Tapera was given the correct medicinal charms for successful
mbira playing, and he also learned two local tunes for the instrument (“Chikunda” Nos.
1 and 2). He did not learn the Sena names of these tunes, and their lyrics, being in a
foreign language, he still does not know.
Eventually he went to work in Salisbury for a number of years, subsequently going
to Bulawayo on the death of his wife in 1949. He has recently returned to live in Salisbury.
Thus most of his musical inspiration has been drawn from his home area of Mrewa,
tempered perhaps by other mbira players he has heard, and whom he frequently mimicks
humorously. The Valley Tonga of the Zambezi valley above Kariba, the Nsenga of
Petauke, Northern Rhodesia, and the Shona/Karanga of Fort Victoria, Southern
Rhodesia are his favourite victims for musical parody.
Tapera s instrument. (See illustration).
The primary keys of a kalimba-type mbira are those numbered one to eight in the
illustration, i.e. three consecutive high notes and a low note in each hand. Instruments
in this form are found in Nyasaland and several parts of Northern Rhodesia, and are
usually small, with fan-shaped bodies, and are played over a small gourd. Instruments
of varying degree of complexity are found in the lower Zambezi valley from the borders
46 AFRICAN MUSIC SOCIETY JOURNAL

Tapera’s “kalimba” mbira.

of Northern Rhodesia and throughout the north eastern corner of Southern Rhodesia,
the additional notes being in most cases an upward continuation of the scale, the keys
being shorter and placed in an upper rank or manual. From one to six keys may be added,
where the principal eight keys of the lower rank have not been increased. Many Sena/
Nyungwe examples of the instrument, however, use more than double the number of
keys in both ranks, the result being as illustrated in Fig. 1, one of the more complex
examples recorded by Flugh Tracey in 1932/33.
The construction of Tapera’s mbira follows the usual pattern of the Sena/Shona
board-resonated type of mbira, and not the Northern Rhodesian type, i.e. rectangular
in shape, a separate piece of wood being used for the backrest, and a thick soundboard
with straight sidewalls. The keys are large and rather more roughly made than the
Northern Rhodesian kalimba. A thin metal plate is secured between the keys and the
soundboard to which rattles are attached, in this case metal bottle tops, which vibrate
in sympathy with the keys. The whole instrument is resonated inside a large mpud^i
calabash, and is propped in bv means of a short stick between the straining bar and the
inside of the calabash.
There seem to be certain well-understood principles behind the layout of the keys
of Sena and Shona mbiras of all types. 1. Most notable is that of duplicated notes on
either side of the centre of the fingerboard. 2. Related to this is the practice of duplicating
the few top notes of one section of the layout with the bottom notes of the next higher
section. The two advantages of this principle are firstly that it allows a rhythmic rippling
or tremolo device, which is a notable feature of Sena/Shona mbira playing, and secondly
that it permits the duplicated notes to be played in harmony with any other note on the
fingerboard, left or right. 3. There are at least eight distinct types of mbira in use among
the Sena and Shona, the mbira d^aMid^imu, madebe d^aMhondoro, njari, njari burn, nyon-
ganyonga, nsansi, mbira d^awaNdau and kalimba {chisansi), and on all of these but the nsansi
and its related mbira d^awaNdau, the bass notes are central and all scales ascend in an
outward direction from the centre of the fingerboard. On those two the bass notes are
on the left, and the scale always ascends towards the right.
 MBIRA MUSIC OF JEGE A. TAPERA 47

4. Adjacent keys in each section of the layout almost invariably produce adjacent
notes of the scale. This may seem obvious, but the principle does not hold true of African
mbiras from many other areas. 5. Notes in one rank are frequently arranged directly
above their octave in a lower rank. The two keys may be played together to reinforce
the sound of the particular note, either with thumb and first finger together, or with a
sweep of the thumb from the upper onto the lower key. 6. The few outside keys on both
left and right sides of the layout are usually set so that they can be played upwards with
the tips of the index fingers. Alternatively, part of the upper right manual of some
instruments such as some kalimbas and mbira d^arvaNdau may be arranged for the right
index to pluck downwards. A reason for this Sena/Shona insistence on more than the
two thumbs alone to pluck the keys may perhaps be found in the great predominance
of triple times in their music. The fact that man has two hands, and that much African
music is in three time, undoubtedly accounts for much of the hemiola in African music
and mbira playing, but if one has three or more separate means of plucking the keys one
can more easily produce unreservedly triple music, as well as many more intricate counter­
plays of rhythm and harmony.
Tapera’s mbira has five extra keys in the upper rank, and of these two pairs are
duplicated, in typical Sena/Shona fashion.1 When he discovered that I too was capable
of making mbira keys, he asked me to make him an extra key (G'), the upper octave of
his keynote (G' Key 6). This I did for him, and it appears in the illustration as the fourth
key from the left in the upper rank. I also made, with his approval, a duplicate key for
his right outside key in the upper rank (F#1), which appears as the left outside key of
the instrument. He said that other players whom he knew used this top G' key, but that,
not being a maker, he had never been able to make one for himself. He began to work
out ways of using the new key as soon as it was made, but unfortunately most of the
transcriptions were done before the key was in operation.

The timing o f Tapera's mbira.

Progressive M ajor tempered
Key N ote trans­ intervals scale fo r comp­ Intervals
No. cribed as v.p.s. (cents) arison (cents) (cents)
(12) (GO (592) (1200) (1200)
(147)
11. F r 544 1053 1100
159
10. E' 496 894 900
208
9. D' 440 686 700
328
8. B' 364 358 400
181
7. A' 328 177 200
ill
6. G 296 1200 1200
147
5. F # 272 1053 1100
159
4. E 248 894 900
208
3. D 220 686 700
328
2. B 182 358 400
358
1. G, 148 0 0

1 Cf. the Nsenga kalimba described by Blacking on p.00, where the six extra keys of the upper rank each sound difftrent notes of the
scale.
48 AFRICAN MUSIC SOCIETY JOURNAL

It will be noticed that there is no equivalent of ‘fa’ in the scale, that is, the pitch
that would be represented by C. This omission is found in several Sena/Shona kalimbas
of the smaller type, and I suggest that this is not because the players are hexatonically
unaware of the existence of the note (as can be shown in the case of the mbira d^aivaNdau)
for the note is used regularly when singing (e.g. in “Kana ndoda”, Fig. 4). What is more
likely is that this note has no place in the harmonic sequences used on the instrument,
and as the kalimba is rarely used melodically but rather as a rhythmic and harmonic
background to the voice, the player has no need for it. Other types of mbira, such as the
njari and madebe d^aMhondoro, from the same areas in which the kalimba is found, show a
complete heptatonic scale in all octaves. This would suggest a different harmonic system
for both the kalimba and the njari, and from listening to many njari tunes, I would suggest
this to be so. N jari tunes are generally longer, with a much wider choice of harmonic
progression. Some, however, show unmistakeably the ‘kalimba’ type of harmonic
progression, as for instance, one that has been transcribed by Nettl: “Shona Karanga
music for sansa orchestra and voices”, published in Nettl’s Music in Primitive Culture
(Harvard 1956).2 It is probable that there are several types of harmonic progression on
the njari and related mbiras, and that the one most suited to the kalimba layout has been
appropriated. This however, needs extensive research into njari technique, which must
remain for the future.
The tuning of Tapera’s notes sounds, on a casual hearing, much like a European
major scale, but the above measurements show that this is not so. Tapera’s scale has two
notes (Keys 4 and 5, and their octaves) that are invariably tuned slightly sharp of the
average Shona mbira tuning, i.e. towards European temperament, which probably
accounted for the relative ease with which I could accept his scale when learning to play
his instrument. These two degrees, the sixth and seventh of the scale, sound severely
flattened on most Shona mbiras.
The table below shows an average Sena/Shona mbira tuning over one octave. This
scale cannot immediately be taken as ‘the Sena/Shona scale’, because many individual
musicians display certain constant tendencies to tune some notes differently, and the
various types of mbiras sometimes employ different tunings. Hugh Tracey first calculated
from measurement of a large number of Sena/Shona mbiras the average tuning of each
type of mbira, and from this calculation he obtained the average tuning for all types.
This overall average differs from any one of the instrumental averages by not more than
ten cents in nearly all cases, with a maximum, in one case, of twenty seven cents, or one
quarter of a tempered semitone. When one bears in mind that the individual variations

A verage Sena/Shorn Taperd’s scale

mbira tuning {cents) {cents)
1200 1200
170 147
1030 1053
178 159
852 894
168 208
684 686
168
516 — 328
173
343 358
181 181
162 177
1*1162 177
0 0

s This tune, although not named, 1 can almost certainly identify as “Ndinosara nani”, composed hy the rninsin ) Ve'awu C.hipitca fro
the Gutu district of Fort Victoria, in January, 1933.
 MBIRA MUSIC OF JEGE A. TAPERA 49

usually concern only one or two notes in the scale, and that the remainder conform
substantially with the average scale given here, this is indeed a remarkable degree of
unanimity in the concept of a ‘right’ scale. One hopes that this average may have some
validity, and that it may serve in future as the basis for a standardised Shona mbira scale.
Although Tapera’s overall tuning varied from time to time within the range of
about one semitone, the relative pitches of his scale remained constant, as I ascertained
from exact measurement on four separate occasions.3 He is always conscientious about
tuning his mbira before performing, and the accuracy of his ear is reflected in the accuracy
of his octaves and unisons.
A NOTE ON THE TRANSCRIPTIONS.
I have chosen to transcribe kalimba music as if in the key of G, with one sharp,
but two things must be remembered while playing or reading from the transcriptions.
One is that the actual pitch of Key 6 of Tapera’s mbira is very near to D (293 v.p.s. new
philharmonic pitch), so the music is written approximately a fourth higher than it
sounds. This was done because the key of G seemed the most suitable key in order to
fit the range of the kalimba into a treble stave. For future studies on any kalimba-type of
mbira, it would undoubtedly be useful to consider it as a ‘transposing instrument’,
for the sake of uniformity of fingering and ease of sightreading from written music.
Thus the G below two ledger lines would always be recognisable as the central bass
note of the instrument; a middle line B would always be the outside left key of the lower
rank, and so on. What this method might lose in technical accuracy would be more than
adequately made up in clarity of reading, comprehension and comparison with other
mbiras of the same type. Others may prefer to use the keys of D, C or B(? for the same
purpose, but I have found G to be the most suitable.
Secondly, the notes, written as they are on our stave, do not represent European
tempered intervals as would be produced by a piano playing this music, but represent the
exact notes given above in the table of Tapera’s tuning.
Key 6 is taken as being the ‘keynote’ for various good reasons:
1. Tapera affirms that this is the “great note” {intern burn) and that the chord 6-3 (G-D)
is the best-sounding chord on the instrument.
2. The note is played at the start of most tunes, and at the end of all.
3. It holds a central position in the harmonic progression, being always returned to
from the other two main chords.
4. Lastly, to my European ear, it sounds plainly like the keynote, although, of course,
it would not be wise to apply this criterion to all mbiras.
It will be seen that in most cases it is not indicated which hand plays which note.
This is because there are only two notes, E in the top space and the D below it, which
can be played by either hand. Where either of these occurs together with its lower octave,
it can be taken as being played by the right index. Where the note occurs alone, the hand
which plays it is indicated by an L or an R. Where it is necessary to make clear the
contrasting rhythms of the two hands, the one is marked by up-tails, and the other by
down tails, L or R being marked at the beginning of the stave.
All these tunes were taught me personally by Tapera, so I can confidently vouch for
the accuracy of the mbira parts, having played them all under his critical eye many times.
The voice parts are transcribed from tape recordings, and represent a condension of
what Tapera would sing with each tune. Falsetto yodelling notes are shown by a crossed
note.

3 Thu method used was comparison of each mbira note with a set of tuning forks calibrated every four vibrations per second betu'ecn
212 y.p.s. and 424 v.p.s.
50 AFRICAN MUSIC SOCIETY JOURNAL

The music.
The single unifying factor behind Tapera’s music, and behind the majority of tunes
I have heard played on this Sena/Shona type of mbira, is the harmonic progression
used. This is something more than a simple root-progression around a tonal centre,
as for instance that described by Blacking in duscussing Venda ocarina music (African
Music, Vol. II, No. 2), and it is strong enough to support a seemingly infinite variety of
tunes. It consists basically of three main ‘chords’ (1, 2 and 3) and a passing ‘chord’4

By a ‘chord’ here is not meant a triad, but a chord of two notes only. That is to say
that when two keys are played in unison, the two always represent, with few exceptions,
one of these chords, or their inversions in different octaves. However, when we consider
the notes that regularly precede or follow each of these chords, while the harmony still
remains on the chord in question, we can add extra notes to three of the chords, which
will make them into ‘understood’ triads. It must be remembered that at no time is a
full triad played on the instrument, although the two mbira notes, together with that
of the voice, may sometimes make one up, particularly when yodelling on the tonic chord
(Chord 1) takes place. Here are the basic chords with the extra notes that are allowed
near each of them printed in black. 5 and 6, parts of tonic chord 1, are the only other
simultaneous chords that are regularly played. Chords 1 to 4 are also played in all the
inversions possible on the instrument.

These chords may now be described as that of the tonic, the submediant minor
(relative minor), the dominant (without its third degree) and the mediant minor. The
order in which they occur is fixed, namely 1-2-1-3. The passing chord 4 may occur in
any, or all, or none of the spaces between the main chords, but more particularly after
each chord 1, i.e. 1-4-2-1-4-3. Some of the vocal parts, such as “Kana ndoda” and
“Amai achauya”, start on chord 2, or on an upbeat before it. This might perhaps demand
a rearranging of the barline for the vocal part. However the mbira generally has the
chords in this order, and starting in this place. The first chord of the above sequence is
also the one used to end a tune, i.e. after the dominant chord 3. So the sequence dominant
—tonic can be said to have cadential function. The voice part usually drops a whole
tone onto the tonic at the cadence, sometimes a third from the mediant.
Tapera’s voice is, by his own recognition, not as flexible as it once was, and he is
no longer capable of the extended vigorous yodelling that would otherwise form an
important part of most of these tunes. This Sena/Shona technique of yodelling is, I
believe, unique among Bantu-speaking peoples, having been only recorded elsewhere
in Africa among the Congo pygmies and the Khoisan peoples of the south. From what I
can hear of Tapera’s performance, from that of the two Fort Victoria performers on
the kalimba-type mbira, and from that of several n\ari players, the yodelling reinforces
the suggestion that the background of this music is chordal. The notes used in yodelling
are those of the chord being played on the mbira, together with the ‘understood’ notes
mentioned above which go with those chords. The characteristic yodelling words are
“Wo-ye i-ye i-ye” etc., where the ‘i’ represents the high falsetto tone. It usually starts on a
high pitch and descends stepwise, one up, one down, following the chords as they change
throughout the length of the yodelling phrase.
 MBIRA MUSIC OF JEGE A. TAPERA 51

The existence of music based on a fixed sequence of chords may be of great interest
when comparing traditional Shona music with that of the towns. If we were to replace
the relative minor chord with the subdominant major chord, we would have a progres­
sion whose similarity with that of everyday African town jazz could not be overlooked.
In order to find out how these chords are put into effect we can follow the instruc­
tion that Tapera gave me for what he considers to be his best tune, “Chemutengure”,
one of those numerous Shona mbira tunes for which every player has his own variation
of the standard. “Mbiriwiri”, “Magonde” and “Neura” are the names of other traditional
tunes that many Shona ri)ari players know. This tune seems to be of more recent origin,
and used mainly by players of the kalimba-type mbira. The first lesson was to learn the
basic chords in a simple rhythm. Tapera would always play with me, to keep the time.
The mbira I used was one that I had made myself, in order to be able to learn, and it had
been tuned by Tapera.

The last bar here is the first concession to cross rhythm between the hands. Here,
and often later on, I discovered for myself what Kubik has observed recently, that there
is a great difference between the player’s and hearer’s image of the music being performed.
Tapera was always able to play the individual part of either hand by itself, thus indicating
that he was conscious of each part as a tune in its own right. But in many of his accom­
paniments the tunes of the two hands overlap in pitch, and so the listener, naturally
associating together notes of similar pitch, or of a regular rhythm, constructs for himself
a rhythmic and melodic framework of which the player need not necessarily be aware.
A good example of this is “Chikunda” No. 2.
A simple matter of technique that had to be overcome from the start was learning
to pluck the right upper rank downwards with the right index, while at the same time
playing the lower octave with the right thumb. Another problem for a non-player of
the mbira was to move the thumbs sideways along the ranks of keys fast enough. Learning
this caused me several weeks of stiff hands.
The first lesson learnt, and the chords roughly mastered, most of my time was
spent in learning the great variety of rhythmic play to which the notes of the chords
can be put. Here are some of the rhythmic patterns for “Chemutengure” that can develop
from the chord sequence in Fig. 6 Pattern 2 is the first extension of Pattern 1.

m m f ¥ v ,y=?
The left hand plays a similar pattern to that of Pattern 1, in a regular 6/8, while the
right hand alternates from 6/8 to 3/4. The bass note, Gy, is used here, as it can be in
many tunes, regardless of its pitch, as a regular rhythmic thump on the third and sixth
beats of a 6/8 bar, which are the same beats that a single drum would emphasise if
accompanying this music. There are many possible hand claps, but if it were to be a
single clap falling every three beats, it would also fall on the third and sixth beats.
A deep note, usually Gr on the ka\imba, very often falls on the quaver immediately
52 AFRICAN MUSIC SOCIETY JOURNAL

preceding the on-beat in a number of Shona mbira tunes that I have heard, and it can be a
useful sign for orienting oneself to the rhythm. The “on-beat” I have assumed to be the
strong beat of the first basic accompaniment for any tune taught me by Tapera, such as
Pattern 1 here; the strong beat of most of the songs fits with this basic pattern.

Pattern 3 is an extension of Pattern 2. Both hands are now double-thumbing in 6/8

time, playing the same quaver-crotchet, cjuaver-crotchet pattern, the left hand staggered
one quaver after the right. This is a common African group drumming technique, but
when one player has to do it himself, it is quite invigorating.

p
r t i fL i e----- f—-'I f T ♦ * Y fr i
~)r~ *
d 1 V H J
Pattern 4 is in the same rhythm as pattern 3, the left hand now repeating on the
upper rank the right hand’s note, instead of its own note.

Pattern 5 is very similar in rhythm to Pattern 3, the right hand now moving ‘up-
down’ instead of ‘down-up’. Tapera generally has at least two distinct accompaniment
patterns for each of his tunes, and often these are described as ‘Big Tune’ and ‘Small
Tune’ (using the English words). Big refers to low notes and Small to high notes;
Big Tune generally means a melody which includes most of the lowest four or five notes
of the instrument. Small Tune means one on the upper manual, together with the left
hand keys 6, 7 and 8 of the lower manual. In this pattern (5) the sequence G,-D-B-E etc.
in the bass, when emphasised, Tapera would call ‘making the tune big’. The effect is
more marked in the next pattern, No. 6.

Here the listener tends to appreciate these two tunes playing against each other:

m , fitc.
b f p / _ l> \ rP •

while the player does not think mainly of these two tunes, but of the fascinating mechan­
ical-rhythmical effect of playing a staggered 6/8 beat between the two hands. The listener
can never be aware of how this accompaniment is created, unless he happens to be a
player of the instrument himself.
 MBIRA MUSIC OF JEGE A. TAPERA 53

Keeping the same double-thumbed pattern in the left hand, the right hand now plays
in 3/4 time.
All the previous patterns are full accompaniments for the voice. The next three,
Nos. 8, 9 and 10, are all examples o f ‘Small Tune’. They may be used behind the voice,
but are more often used as mbira solos in between parts of the song.

This is the main Small Tune, usually played during a quiet or lamentative section
of the music. It was in fact the first accompaniment that Tapera tried to teach me for
“Chemutengure”, but finding that I could not grasp it, he went back one stage further—
to the basic chords.

Pattern 9 is an extension of Pattern 8. The rhythm starts to develop with the repeated
bass notes E-E and D-D in bars two and four, and the added semiquaver in the treble.

Pattern 10 is a further extension of pattern 8. It shows another way of increasing

the interest of the Small Tune. This filling up of all available rhythmic space is what
Tapera calls the ‘njari technique’. The njari has many more duplicated keys than Taper’as
Tapera’s mbira, and is played with four fingers, as against Tapera’s three, so he double­
thumbs to get the same full effect.

i#b i A f r — i- i •
--- +---- •1 j A i
L—— •
T ■t
This is the introduction to “Chemutengure” that Tapera played on a test recording
that I made. Or rather, this is what it sounded like to me. When he came in on the song
one had to adjust one’s rhythmic bearings rapidly, and working back to the start one
could discover what the background rhythm in his mind actually was. Tapera often
starts his tunes with an introduction that appears to be in a quite different rhythm from
the main body of the song. Frequently this introduction is the main contrasting rhythm
pattern that he would get a second mbira, in this case mine, to play behind him when he
had eventually set up the main pattern.
54 AFRICAN MUSIC SOCIETY JOURNAL

This is the introduction as it is in fact felt, and the accompaniment pattern for the
first part of the tune as recorded.
For the ending of the recorded tune, Tapera played this energetic “Big Tune”
pattern, followed by this typical ending pattern.

I I
Having grasped these patterns, one has to learn to vary them constantly, fading
one into the other, sometimes working up from simplicity to complexity, sometimes the
reverse; sometimes loudly emphasising certain notes in a pattern, sometimes unexpectedly
leaving some out. The elements of humour, surprise, improvisation, reaction to the
audience are all part of the art. Most tunes start slowly, even hesitantly, without complica­
tion, and gradually build up to the full pattern. They end, conversely, with a gradual
reduction of the complications. This is also a typical feature of Shona njari playing.
To complete the music of “Chemutengure”, here are some of the vocal phrases
regularly used by Tapera in the song. Each phrase length of four bars, marked off by
double bars, corresponds with the four bar length of one accompaniment pattern.

V |'|j fLLI*|LU’ LU
M UKAbzl NKHAIRi KVVi R a
w&- PA

“Chemutengure”
“Chemutengure” was considered by Tapera to be his best tune; he also had the
greatest number of variants in its mbira part. It is a virtuoso piece—an exercise in
rhythms. Most of his other tunes have a more standardised accompaniment. One of the
first that I learnt was “Kana ndoda kuramba murume”, a sad song in 5/8 time. “If I
should want to refuse (divorce) my husband, I should be very worried. Have you eaten
your food with the mouse’s head?” (A mouse’s head is considered a delicacy).
 MBIRA MUSIC OF JEGE A. TAPERA 55

w r .
/n
! y>
>
— rH r 1 1 r mI*1#1_#
Ml
1 I . 1' i t :
7 •O06H
• 6.-R
-#
KA- NA NDOPA KU N E-- -0-
K6-T-jrA.

-i- t- tH ” 1—*I— *i— t—

I J n_tz__ cT
t *7 .. 0 r ; -.
Vf if-•—' • »— •—-—#--------- # 0 7 ------J - 1-------

-fcf1
UNOVJ RAM6A
S i® tilPPp
UNO\j RAmBA
5

0 -0 0 -0 0 -0 AlUftOKANBA Ml/MoRAMgA
MUftUME An/KA^il

Alternative patterns that can be used with “Kana ndoda” are: 1. Where the right
hand pattern is doubled, in imitation of the song; 2. Which is the same as the Small
Tune solo of “Chemutengure”, but adapted to 5/8 time; 3. The first basic accompaniment
that I was taught; 4. The pattern that Tapera used when he had the extra top G' key;
5. Clap patterns that are used with this tune are any, or all of these three.
(See transcription on P age 62).
The next tune that I learnt, because Tapera considered it easy to play, was “Kuyaura
kwasi nakatura”, another sad personal song, sung with a catch in the throat that makes
onlookers exclaim, “Anochema kwazvo”—“He is really crying, mourning”, which,
they say, is how old men ought to sing. Others sung like this are “Amai achauya”,
“Gumbukumbu” and “Ndonofira msango”. The words of “Kuyaura” were explained
to mean ‘the suffering of a man who has nothing, neither land, house nor cattle’. Tapera
also mourns the loss of his wife in some of the exclamations that fill many of the pauses in
the singing. These exclamations in most Shona mbira music, sometimes taking up more
56 AFRICAN MUSIC SOCIETY JOURNAL

time than the singing itself, are often an important and permanent part of a song, and
invariably performed with it. They are comments in the mood of the song itself, tradi­
tional exclamations, or else topical ad libs. Many Shona mbira songs, in fact, have no
sung words at all but the yodelling “woye iye iye”, interspersed with these spoken
exclamations. Not being a Shona linguist, I am not able to comment on Tapera’s numer­
ous exclamations.

“Kuyaura kwasi nakatura”

“Amai achauya” is a mourning song composed especially for his wife, Maggie,
whose loss must have affected him deeply. “Mother will come; will come and see; will
come and see this poor man, yuwi uye yuye iye iyuwe”.
 MBIRA MUSIC OF JEGE A . TAPERA 57

M ri i u —i |
•—•—* *■) >-------- \------------1------*—h• • * • j i —
KA w i\ RoTj&f AW wo (vfa wARAf k AwA ^ '6 6 won K MNoNtfWA wo-

1 and 2 are alternative accompaniment patterns; 3 is a short solo passage in between

parts of the song.
Shona mbira music in duple time is not common, and is more typical of the Eastern
Shona, that is, the Manyika of the Umtali district of Southern Rhodesia and eastwards
into Mozambique. “Butsu Mtandari” is a Manyika song. The clap that the Manyika
put to the song is written below.
“Butsu mtandari,
Wona inenge vombe.”
58 AFRICAN MUSIC SOCIETY JOURNAL

“Butsu mtandari”

“Ndonofira msango” is another song in duple time, from Tapera’s home in Mang­
wendi. It is a song for the Mhondoro spirits, originally played on the njari. It is performed
very quietly.
 MBIRA MUSIC OF JEGE A. TAPERA 59

V INTI*). N Js UJ
—
*_ •

It would be possible from the accentuation, and from observing Tapera’s movements
while playing, to bar this song thus:

but the song definitely fits the other barring, and, to judge from the beginning and end­
ing, the mbira part should also be barred the same way. The eight quavers of the bar are
divided into three, three and two, placed not on 1, 4 and 7, as in the common Latin
American rhythm, but on 3, 5 and 8.
Gumbukumbu is the name of a certain kind of small bird that frequents rivers and
marshy places, and the song “Gumbukumbu” is addressed to this bird as, one imagines,
the composer loiters alone and palely by the river. “Gumbukumbu, you understand;
you understand but you refuse to answer”.
60 AFRICAN MUSIC SOCIETY JOURNAL

“Gumbukumbu”

This and “Chikunda” No. 1 are the only tunes that I heard Tapera play in any other
mode than his usual one. Yet he still ends it on the tonic G-D chord. 1. appears to
be the chord sequence. 2. shows two regularly played variants for bar one of the
mbira pattern. 3. is the full form of the accompaniment using the right index.
“Chinotamba mudzia mwacho” was of interest as being the only song I learnt
that was in triple time and of eighteen quavers length. The others are all of twenty-four
quavers length. This is a children’s song from Mrewa, or it can be sung by a mother
to the child on her back. According to Hinde, in NADA 1932, this is a children’s game
of pick-a-back. “Little imp which is on the back, so not carry me. It dances round in its
own pool, the little chigwaya” (a kind of small fish).
 MBIRA MUSIC OF JEGE A. TAPERA 61

“Chinotamba mudzia mwacho”

The rhythm of the words “Karombo kare pamsana, zandibereke” is a good example
of how the Shona like to give a tricky rhythm to a sentence. .It is an important part of
their poetry. The barring may be in doubt here; the song starts every time on the fourth
beat of the 12/8 bar, as does the introduction, so the bar line, for the song at least, ought
perhaps to come in front of that beat. As for the mbira accompaniment, the present
barring indicates the phrasing, the 3/4 bar being ‘on its own’.
Finally, here are some extracts from the two Chikunda tunes that Tapera learnt as
a young man in Tete. ‘Chikunda’ is the name of the language of the Sena/Nyungwe
people of Tete. The first is in the relative minor mode; the legato passage at the end is
how Tapera would have me play against him on a second mbira.
62 AFRICAN MUSIC SOCIETY JOURNAL

“Chikunda No. 1”
{See P age 55).
 M B IR A M U S IC O F JE G E A . T A P E R A 63

The second is in Tapera’s usual mode. It is a display of virtuosity on the instrument.

You can only play it when your fingers are ‘clever’ he says. Its main interest is in the
changing directions of the melody; the basic tune itself varies intriguingly between up
and down, and, in the variations, the direction is constantly changing, with notes appear­
ing here and there unexpectedly, giving an effect of surprise which invariably pleases
an African audience. Note the use of dynamics for variation.
This is another tune with a different player- and listener-image. The player is
conscious of the contrast of the left and right hand rhythms:

“Chikunda No. 2”
The listener, however, associates together notes of similar pitch and constructs
his own rhythmic framework for the tune on the normal two against three pattern.