The Family of the MbiraTHE EVIDENCE OF THE TUNING PLANSA. TraceyInternational Library of African Music, RoodeportINTRODUCTIONAbout twelve major types of mbira areplayed at the present day by the peoples ofthe central and lower Zambezi valley and en-virons. I would like to present some evidenceto show that, in this large part of Africa atleast, all these types are related, and furtherthat several lines of descent can be postulatedwhich point back to one common progenitor.The area in question (see map) comprisesmost of Rhodesia, central Mozambique, south-ern and eastern Zambia, together with parts ofsouthern Malawi, southern Mocambique andnorthern Transvaal, or to put it more simply,much of the lower Zambezi basin, with a spillover towards the south.The mbiras found in this area exhibit agreat variety in appearance, manner of con-struction, tone quality and musical technique.Distribution maps could be drawn for eachof these factors which would not reveal an over-all relationship. However, if one focuses onone factor alone, namely the arrangement ofthe notes on the keyboard, or 'tuning plan', con-sistent relationships can be established, whichalso concord with African nomenclature. Afri-can musicians are completely decided on thenames of the various types of mbiras. What issignificant to them in this respect is not prim-arily the number of reeds, or keys, on the instru-ment, the method of sound production or anyother constructional feature, but the tuningplan. Most mbira tuning plans are more or lessirregular, in the sense that the notes of thescale are not necessarily positioned next toeach other, as they would be, for instance, ona piano or on most African xylophones. It isby tracing several significant 'irregular' tuningplans that we have been able to come tothese conclusions.Irregularities in tuning plans can be com-pared to non-functional components of materialobjects, such as decoration, which may betransmitted unchanged from one culture toanother, the original function or symbolismbecoming lost in the process. Tuning plans aregenerated by musical concepts or needs. Never-theless, one generally finds that an mbira tuningplan is adopted lock, stock and barrel, whetheror not its special characteristics fill the sameneed as in the music of the originating group.Some of the irregularities that we shall befollowing then become merely a vestigialhistorical trace that links a type of mbira withits antecedent, but does not necessarily haveany significance in the music of the peopleusing it.I should add that I have been considerablyassisted in my research by the ability to playmost of the mbira types under discussion, andthus to form an idea of what is actually of1musical or motor-pattern significance, and whatseems 'irregular' and should therefore beattributed to other sources, historical, socialor personal. Secondly, I am not concerned withestablishing a time scale, but rather a seriesof relationships among the mbiras of this area.On considering these relationships it becomesobvious that their development could onlyhave occurred in one direction, from simple tomore complex. One does not need to assumethis to start with, although it is also, of course,a perfectly sound ethnographic principle. It hasbeen seriously suggested to me by an ethno-musicologist, that the development took placein the other direction. On the basis of thisevidence this seems to me impossible; however,we are trying to retrieve a historical processfrom the evidence of present-day artifacts,which is always a risky business.Nevertheless what I have observed ofpresent-day African musicians is a generaltendency among leading musicians and instru-ment makers to add notes and increase the sizeof their mbiras. Many inventive variants ofthe basic tuning plans are produced in this way,a process similar to that which must have ledup to the currently existing mbira types.Similarly in the matter of the so-called'Africanisation' of foreign musical traits, Ihave noticed that many musicians, on beingexposed to a new foreign musical instrument,whether from another part of Africa, or West-ern, will immediately play it, applying motor,rhythmic or melodic patterns to it which theyalready know. They are keen to accept innova-tions in musical instruments, providing they arecapable of expressing their own musical langu-age. In common with musicians the world over,they are rarely interested in learning theoriginal, foreign musical idiom of the instru-ment. This is why the kalimba, for instance,from the north side of the Zambezi, and thekarimba, from the south side (roughly speak-ing), which are identical except in matter ofsize, can sound so completely different. Similar-ly, in organological terms, when a new typeof mbira is adopted, the significant adoption isthe tuning plan; other details of constructionwill usually continue to be based on those ofthe local mbira type, such as the shape andsize of body, profile of the reeds, attachmentof bridge, bar and backrest, method of effect-ing buzz-tone, and choice of resonating body.Diagram 1 presents in simplified form thefamily relationships of the mbiras under dis-cussion; their geographical distribution andsupplementary description is shown in themap and its subtitles. Diagrams 2-4 present ingraphic form the actual tuning plans on whichthis study is based. For present purposes wedo not need to follow the steps in detail; readerswho are interested may find it more fully des-cribed in Tracey (1972). Here it will be suffi-cient to point out the main points of similarityamong the members of the three lines orfamilies of mbiras.THE FAMILY OF THE KALIMBAThe present-day kalimba is played over awide area, mostly north of the Zambezi.Usually there are added notes, bent up to forman upper rank, or manual, in the manner ofthe Nsenga kalimba illustrated. On the southbank the kalimbalkarimba name is retained,but the instrument is considerably enlarged,giving a foretaste of the further enlargementsand transformations other south-bank mbirashave undergone. The original kalimba 'core'tuning plan is shown in black, to show how itremains embodied in the subsequent tuningplans of this family. The njari, already an in-dependent instrument from the kalimba, makesits additions in its own way, the most significantbeing the enlargement of the bass section forthe left thumb. The njari huru, or 'big njari',is just that Š the bass section is further en-larged, and the right hand rank is extended byfive more notes.The dipila is included here tentatively, as Ido not have conclusive evidence for its mem-bership in the kalimba family. However,allowing for the fact that it is pentatonic, itstuning plan is strikingly similar to that of thekalimba. The black notes here represent thebasic form of the instrument, as acknowledgedby dipila players; the white notes are optional.A note on the numbering system used ŠŁthe numbers 1 to 7 indicate the seven notes ofthe heptatonic scales used on most of thembiras in our area; 1 to 6, respectively, if thescale is hexatonic (for instance the mbira dzavaNdau, Diagram 4), and 1 to 5 for the onlypentatonic instrument in the area, the dipila.That is, the heavy black lines are an octaveapart, divided respectively into 5, 6 or 7 inter-* *Ł>Ł %A i-i ivals. This numbering system does not showtonal, intonational or musical importance ofnotes, but it is carried through all the diagramsto facilitate comparison between the differenttuning plans.THE FAMILY OF MBIRA DZAVADZIMUThe mbira dzavadzimu and mbila deza,probably the older members of this family,embody the kalimba core in the centre of theirkeyboard, completely intact but for the factthat it is left-right reversed. Again the blacknotes show the kalimba core. The major addi-tion in these two instruments is the new lowsection for the left hand, in octaves below the7-6-5-1 of the kalimba core. The note no. 4,whose irregular position was one of the sparksfor this investigation, has been inserted herein yet a different position than the two positionsfound in the karimba-njari family already dis-cussed. Note that in the original kalimba thenote 4 does not exist, so musicians in a hepta-tonic tradition, faced with a hexatonic instru-ment, had to find a new place for the extranote. In the hexatonic Ndau mbira, of course,this note has never been added.Taking the mbira dzavadzimu tuning plannow as our model, with its characteristicirregularity in the left hand, we can see thatthe hera and nyonganyonga both embody it asthe basis of their own enlarged keyboards, theonly difference being the shifting of note 3away from its original isolated position. Thetransposition two octaves down of the righthand rank 1, 2, 3, etc in the nyonganyongacan probably be accounted for by the charac-teristic method of tuning these last two mbiras,the hera and nyonganyonga: the overtones ofthe deep notes, rather than the fundamentals,are taken as the significant tuning notes. Over-tone and fundamental on an mbira reed arenaturally about two octaves apart, so in pointof fact this low right hand rank on the nyonga-nyonga sounds at the same pitch as that on itsbrother instrument, the hera.A further link between the members ofthis family is that, to a greater extent thaneither of the other families, they are used forritual purposes. The mbira dzavadzimu andhera, for instance, and to a lesser extent thenjari, are an integral part of Shona religion.THE FAMILY OF THE NDIMBAThis family differs from the others in thatthe reeds are largely arranged with the deepnotes on the left and high notes on the right.Thus the inherent irregularity of the kalimba,which has been our guide, is largely destroyed.My argument for its inclusion in the kalimbafamily rests mainly on small pieces of evidence.The kalimba and the ndimba types are playedin several of the same localities to the northof the Zambezi, with the indication that thereis a progression from the first to the second;the kalimba. is used by 'amateurs', the ndimbaby 'professionals'. Further, the pitches of thenotes used are the same in both instruments.It seems entirely probable that the ndimbaresults from a deliberate change-over from thekalimba, although one must allow the possi-bility of the reverse process. The evidence ofxylophone tuning plans, such as that of theLozi silimba, may have something to add here.It is somewhat similar to the ndimba.The hexatonic mbira dzavaNdau, in itssimplest form, that which underlies the tuningplans of all the others, is similar to the ndimba.The lower rank is actually identical, which isobscured by the present numbering system,based as it is only on the number of intervalsin the octave. The upper left rank is alsosimilar in function and placing. The varieties ofthe mbira dzavaNdau are many. Three of themost widespread are shown here, the highland,or Tomboji, and the Danda types, which extendboth ranks to the right, and add a high thirdrank on the left, and the Uteve and the similarMashanga type, which adds its third rankdownwards. The Ndau mbiras are the onlyones of my acquaintance with three ranks ofreeds.The Mana embudzi, or 'goat's teeth', seemsto be a heptatonic adaptation of the ndimbatype, with added notes to allow it to play inthe full-sounding Shona/Sena style, although Ido not have sufficient evidence to be decidedon this.CONCLUSIONIf the relationships shown by these tuningplans are accepted, they may help to throwmore light on the early relations between thepeoples of this large part of Africa, alreadyincreasingly illuminated by Rhodesian andother ethno-historians. The least they show isthat trading or cultural relations in this areamust have been relatively stable for a longperiod of time. There is no mbira type in thisarea which cannot be connected, with varyingdegrees of certitude, to one particular seminaltype, the eight-note kalimba, which seems tooccupy the position nearest the centre. Otherswill be able to offer dates with more exactitude,but on the evidence of its family, I do notsee how the kalimba can be less than about1 000 years old. It may indeed be the originalAfrican mbira.BIBLIOGRAPHYBLACKING, J. 1961 'Patterns of Nsenga kalimba music', African Music, 2, iv, 26-43.DAVIDSON, M. 1963 'A Lunda kalendi', African Music, 3, ii, 15-16.1964 'The music of a Lunda kalendi', ibid., 3, iii, 107-8.1970 'Some music for the Lala kankobele', ibid., 4, iv, 103-13.JONES, A. M. 1949 African Music in Northern Rhodesia and Some Other Places, Livingstone, Rhodes-Living-stone Museum, Occasional Paper No. 4.1950 'The kalimba of the Lala tribe, Northern Rhodesia', Africa, 20, 324-34.1962 'Venda note-names', African Music, 3, i, 49-53.1964 Africa and Indonesia, Leyden, E. J. Brill.KIRBY, P. R. 1953 The Musical Instruments of the Native Races of South Africa, Johannesburg, WitwatersrandUniv. Press.KUBIK, G. 1964, 1965 'Generic names for the mbira', African Music, 3, iii, 25-33; ibid., 3, iv, 72-3.MAUGH, C. G. 1969 The Journals of Carl Mauch, ed. E. E. Burke, Salisbury National Archives of Rhodesia.MENSAH, A. A. 1970 'The music of Zumaile village, Zambia', African Music, 4, iv, 96-102.NURSE, G. T. 1970 'Cewa concepts of musical instruments', African Music, 4, iv, 32-6.TRACEY, A. 1961 'The mbira music of Jege A Tapera', African Music, 2, iv, 44-63.1963 'Three tunes for mbira dza vadzirau', ibid., 3,ii, 23-6.1969 'The tuning of mbira reeds', ibid., 4, iii, 96-100.1970 'The matepe mbira music of Rhodesia', ibid., 4, iv, 37-61.1970 How to Play the Mbira (Dza Vadzimu), Roodepoort, International Library of African Music.1972 'The original African mbira?', African Music, 5, ii, 85-104.TRACEY, H. 1961 'A case for the name mbira', African Music, 2, iv, 17-25.1969 'The mbira class of instruments in Rhodesia (1932)', ibid., 4, iii, 78-95.THE DISTRIBUTION OF SOME OF THE DESCENDANTS OF THEKALIMBA,Ł Ł*ZAMBIA.Livingstone- J If^ ABHOUESIAi- >;fK)f ' .milssp 5,0 1OO ISO 2O0 25O 300 /I I _ I _T J____i J /y VKEY TO MAPA KALIMBA TYPENyungwe, Gcwa, Ngoni, Tumbuka, Nsenga, Swaka, Kaonde, etc. Also called kankobelo,kankowela et sim. (Bisa, Lala, Lenje), nsansi, sansi, (Cewa, Ngoni, Nyungwe). Possibly alsoTonga kankobela and Lunda kalendi.B NDIMBA TYPENsenga, also called ndandi (Lala), kangombio (Lozi), kathandi (Mbunda).C KARIMBANyungwe, Chikunda, Sena/Tonga, Korekore, Zezuru, Karanga. Also called kansansi, chisansi,nsansi, sansi, shanzhe ye psviro (Nyungwe).D NJARI(a) Valley type. Nyungwe, Sena/Tonga.(b) Highland type. Njanja (known as marimba), Karanga, Hera, Bocha, Garwe, Manyika,Zezuru, Nohwe, Shangwe, etc. Sometimes known as njari dza maNjanja, from the Njanja, itsintroducers. An mbira called deza, probably the njari type, has been recorded among theValley Tonga.E NJARI HURUChikundaF HERA(a) Korekore, Tavara, Nyungwe.(b) Matcpe (Sena/Tonga), madhebhe (Korekorc/Budya).G MANA EMBUDZISena/Tonga, Nyungwe, Sena. Also called mbira dza va Tonga, nsansi, sansi.H MBIRA DZA VADZIMU(a) Zezuru, formerly also played by the Karanga (mbira dze midzimu).(b) Mbira huru, matepe, very few survivors of the Manyika type.I NYONGANYONGABarwe, Gorongozi, Sena. Also called marimba.J MBIRA DZA VANDAU(a) Highland, or Tomboji. (b) Danda. (c) Utee. (d) Mashanga. (e) Hlengwe (called timbila).(f) Shangana (called mbira, marimba, timbila).K MBILA DEZALemba. Venda. Four types Š scale regular/irregular, bass left/right?L DIPILAPedi, Northern Sotho.This map should be taken only as indication of what type of mbira is most likely to befound in any area. There are few hard and fast boundary lines. Individual examples of mbirasarc often found far from the areas marked. The most distinct boundaries are those betweenlanguage groups, such as Shona to Sena (njari to mana embudzi, nyonganyonga) or Shona toNdau (njari, mbira dza vacizimu to mbira dza vaNdau). There is also a certain amount ofspeculation involved; where I have no personal experience of the boundaries I have assumed thatthey continue along the lines of the language division, as in Mitchell and Fortune's map "Africantribes and languages of the Federation of Rhodesia and Nyasaland", Director of FederalSurveys, Salisbury 1964.In particular I do not know the exact boundaries of the Venda and Pedi types (K, L),the southern boundary of the mbira dza vaNdau type (J), the western boundary of the hera (F),whether the valley type of njari is played by all the Nyungwe and Chikunda peoples, as I haveshown, and the mutual boundaries of the njari, mana embudzi and nyonganyonga (D a, G, I,)which are not very clear on the ground Š in this area it is quite common to find up to threedifferent types being played in the same place, often with a shared repertoire, although notnecessarily playing together.The karimba (C) turns up sporadically over a large area; I have only marked those whereit seems to be widely popular. The C at Bulawayo refers to the type of karimba made and taughtthere at the Kwanongoma College of African Music, which was originally based on a karimbafrom Mrewa district, 50 miles east of Salisbury.Finally, as my information from Zambia and Malawi is largely from secondary sourcesI have only put one symbol in at the approximate centre of each language area. The kalimbaarea to the north of the Zambezi is probably much more extensive than shown.>Ł Ł"'i Ł<Diagram 1TIE FAMILY OF THE KALIMBAKALIMBA* 4KABIHBAfv y(DIFILA)?V Ji early" typ/"""'"" )\bass or, R' ID1HDI type](MAKA EMBOTJZIJ/ mbira -dza vadzimu1 eaiiy type\bass on L(MBIIA DZA VANDAU\(KJAai)/IMBII.A DIZAVbass on L or Rl-MBIBA DZA VADZIMUJMBILA DEZA\1 bass on L or EV order regular /f HlIBIf )( HKKA "\ ( NTONGANTONGA IV yv J V/ Mashanga\ Utee Danda \\ XombojI Hlengvej\ Shangana /Diagram 2THE. FAMILY OF THE KAIJMBA Š KAR1MBA Š- NJARl Š NJARI HURU'Ise^ga fcallmba (Blacking),The basic kallmba core is shown by the bkek notes.jari, highland type. Mot a!I ate as full as this, especially inthe upper ranks. This diagram will also serve for tbeilky type., v/hich sometimes adds more notes downwardsT the L jndejr, aod L thumb lo^/er tarsk, and notes 2 and 3in the R lower thumb ranks bent up as in the ntsn hum.'iiiru. The notes whb grrows srs bent up sHghtb.'fs"!'>"- Ł'<'-, ",Ł>,!'. - MBILA DEZA Š HERAnV ^ '---G- -«The basic kalimba core. jfiLHra dm -vadtimn. The kalinibs core is shown in bkekMhlia de?.a5 older type (KIrby).'5! -Ha AC/A, left-handed3 iregutac scale type.!Ł bkck orrfgs show rhe -'ssembt&nce with the11 ;h 'sorrows s.?e b&at to form 'AH uppftt .tauk.Ot1" IM' d note is found occasionally-< ,i'-i instruments.Diagram 4THE FAMILY OF THE NDIMBA Š MANA EMBUDZI Š MBIRA%' O rO '£nr-fTDZA VANDAUKdmiM (Blacking),1- OLE - 4', 5' and 6' in the L hand are often spaced. Ł me nrte further to the left.The basic kalinsha core. AO "o-f, O io- ^- o-o »-oO tJiMbsrst d&a vsNdata, Torp'boii type.The bracketed notes may be omitted-ŠeŠ; y--t«orHt Lrec >ic known to outsiders as Te