9. The Structure of Chanted Ipili Tindi

Transcription

1 9. The Structure of Chnted Ipili Tindi Frnces Ingemnn Introduction The Ipili people of the Porger nd Piyl vlleys of Ppu New Guine hve trditionl tles, clled tindi, which my be told either in ordinry spoken style or s long poetic chnted tles. Informtion bout the Ipili people nd tindi cn be found in the chpter 8 of this volume. Spoken versions of tindi re shorter nd less elborte thn the chnted versions. Chnted tindi re not memorized poems, but re recreted from trditionl elements in ech performnce. While nyone my tell spoken version, only skilled tellers re cpble of performing chnted versions. The fct tht gret skill is required for the performnce ws demonstrted by unsuccessful ttempts of two men who were not trditionl tellers. They were ble to imitte the chnt melody, but were excessively repetitive nd could not produce the rich poetic lnguge of skilled chnter. The chnters I recorded were ll men nd mostly middle-ged, three of whom performed more thn once. The observtions in this rticle re bsed on seven tindi chnted by four men from the Piyl in : Alu, Knend, Pinyti (two tles), nd Yndpke (three tles). As might be expected, there re minor differences in style mong the chnters. 1 The tindi were not performed in nturl storytelling settings, but under field conditions specificlly for the purpose of mking recording. The teller typiclly st cross-legged on the ground inside house nd rocked bck nd forth during the performnce without ttempting to mke eye contct with the listeners or ccompnying the chnting with gestures. In order to mke recordings tht would be cler enough to trnscribe lter, the number of people present ws limited, nd they were instructed not to spek during the recording. The recordings, therefore, do not provide informtion s to wht the norml interctions between the teller nd the udience might be. 1 One chnted tindi ws recorded in 2000 in the Porger region. Although there re mny similrities with the 1960s tindi, there re greter differences thn mong the 1960s versions. Since only one tle ws collected in 2000, it is not possible to know to wht extent the differences re due to the stylistic vrition of single chnter, regionl differences, or developments in chnting style since the 1960s. The 2000 chnter lerned to chnt tindi from his fther, who ws lso chnter. His skill is recognized mong the Ipili people by the fct tht he hs been ble to chrge dmission to his performnces. 197

2 Sung Tles from the Ppu New Guine Highlnds Tindi hve n internl structure which divides the long tle into segments nd lines determined lrgely by the chnt melody but lso by limited number of conventionl phrses t the end of segments nd lines. These structurl elements will be described in the following sections. Segments Segments re delineted by puses nd the chnt melody of the line immeditely preceding the puse, which differs from norml line by hving slightly different melody t the end nd greter-thn-norml lengthening of the finl syllble. All the chnters ended the segment with lengthened, semnticlly empty finl o (s discussed in chpter 1 under the rubric of vocbles ). There is lso usully finl phrse, mening tht the teller is repeting wht hs been pssed down, which is dded to n otherwise norml line. The most common such ending is lei tup y-o this is wht they sid. The length of segment is not fixed, with some chnters preferring shorter segments thn others. There were segments s short s three lines nd others well over hundred. Segments do not necessrily coincide with episodes in the story. Although n episode my end before puse, the puse my lso be inserted t suspenseful moment or elsewhere in the nrrtion. Repetition sets of the sort described by Borchrd nd Gibbs in chpter 8 often occur before puse nd re never interrupted by puse. Puses between segments lso vried in length. Some chnters took the opportunity to relx for moment, cler their throts, or engge in chitcht with listeners. At other times the puses were only long enough to tke breth. In generl, short puses followed short segments, nd longer puses followed longer segments. As the tle resumes following puse, there is frequently n introductory phrse such s de ote, which corresponds to something like the English well, or ondo tpe wit. Another common wy to resume the nrrtion is brief recpitultion or reference to wht hs hppened just before the puse. For exmple, the beginning of the second segment in the tle told by Pinyti is: une pene-ko ylu ndele-p thus did.3sg-def seems you.see-? It seems he did tht; do you see? The numbers in exmples refer to lines in the tindi. The trnscription used here is similr to the orthogrphy developed by Terrnce Borchrd. Minor differences re the use of single vowels for ll monosyllbic words (there is no vowel length distinction in Ipili) nd word divisions. Abbrevitions used in the interliner trnsltion re: 3sg (third-person singulr); ux (uxiliry); def (definite mrker); emph (emphtic); loc (loctive); T (poetic word not used in ordinry speech); V (vriety of);? (interrogtive mrker). No ttempt hs been mde in the trnsltion to identify individul inflectionl nd derivtionl morphemes.

3 9. The Structure of Chnted Ipili Tindi Lines Lines of tindi re mrked by the chnt melody. There re usully slight puses between lines, but occsionlly two or more lines my be chnted without such brek. Lines my optionlly be embellished with finl o, but it is not lengthened s much the o t the end of segments. Lines re vrible in length with no rhyming, llitertive, or metricl structure. The principl poetic device is the prllelism described by Borchrd nd Gibbs in chpter 8. Some chnters subdivided lines by lengthening n internl prt of the line, usully t syntctic constituent boundry. A line my equl sentence, but sentence my lso be spred out over severl lines. Although line my consist of wht could be norml spoken utternce, it usully is more elborte. Lines usully contin words which hve no relevnce to the story line. Interpreters omit them in trnsltion nd when sked, sy they hve no mening even though the words my hve literl menings elsewhere. For exmple, mindi one, hs no mening in the line from Yndpke (1964c): 38. ywo ypu lo mo nn mindi pe ote bking divide ux.-ing cross down one went.3sg emph Bking [food], he went round dividing it up. Mny lines end in formulic expressions to indicte tht the story comes from wht someone else hs sid. Exmples of this re le sid nd lm epe cme to sy. Another type of formulic line ending is one tht evokes involvement of the listener. Exmples of such expressions re ylu ndele-p it is, do you see/ know? 3 nd ndoko see. 4 The pssge below of some typicl tindi lines comes from tle performed by Alu in 1965 (n udio file of this exmple cn be found in online item 12). 5 Line 22 consists of cluse embedded in loctive phrse followed by formulic ending lm epe he cme to sy, indicting tht this is not something tht the teller knows from personl knowledge. The sentence is continued in line 23. Agin there is reference to this hving been told by someone else: le he sid. It should be noted tht in ordinry Ipili speech there is n ffix -pi dded to remote pst inflections to indicte tht the speker does not know of the event from personl knowledge. This ffix does not occur in the chnted tindi. 3 The verb nd, hving the bsic mening see, lso mens know. 4 The word ndoko lso mens where, but here seems just to be mens of involving the listener, like the colloquil English tg see. 5 Reders who listen to the online recording of these lines should note tht there is flse strt before line 32, in which the performer mistkenly repets prt of line

4 Sung Tles from the Ppu New Guine Highlnds 22. wmb yle pu yndene-si lm epe before like doing plnted.3sg-loc to.sy cme.3sg Where he plnted s he hd done before, it ws sid, 23. komb iyu kiy-lliy tup yo pene le komb up komb plurl putting did.3sg sid.3sg He plnted komb [ cbbge-like edible lefy plnt], it ws sid. 24. komb iyu kiyl tup yo pene le komb up komb plurl putting did.3sg sid.3sg He plnted komb, it ws sid. 25. wmb yle pu yndene-si lm epe before like doing plnted.3sg-loc to.sy cme.3sg Where he plnted s he hd done before, it ws sid, 26. tikili uli mole moke lm epe-o Tsweet.potto sweet.potto Vsweet.potto Vsweet.potto to.sy cme.3sg-o Mole nd moke sweet pottoes, it ws sid, [ T indictes tht this is poetic vocbulry not used in ordinry speech. V indictes vriety of plnt for which I do not know n English equivlent] 27. ppo uppo lm epe-o Tsweet.potto Vsweet.potto to.sy cme.3sg-o Uppo sweet pottoes, it ws sid, 28. utupne ote poli poko ne tete pim epene y le them emph Trt bush.rt tooth chew to.do cme.3sg ws sid.3sg bush rts cme to chew them, it ws sid. 29. poloko ne tete pim epene y le-o house.rt tooth chew do cme.3sg ws sid.3sg-o House rts cme to chew them, it ws sid. 30. de ote tu ote yoko ynduw-ko o de lo well emph enough emph cultivting I.plnted-def o well sying Sying, Well, I hve plnted enough. 31. kope yle nd mindi pukle piko o de lo Tbig like house I.will.do tht o - sying Sying, I will build big house. 32. nd mindi pukle ondo tpe le house I.will.do wit be sid.3sg He sid, I will build house, wit. Line 23 lso illustrtes nother chrcteristic of tindi lnguge: specilized vocbulry not used in ordinry speech. Kiy-lliy is used only in poetic lnguge nd hs the sme mening s the norml word komb, n edible lefy plnt in the cbbge fmily. Mny of these poetic words re well known to everyone, but some re obscure, nd Ipili spekers were sometimes unble to trnslte 200

5 9. The Structure of Chnted Ipili Tindi lines becuse they did not understnd them. At times, s in line 23, the use of the ordinry word in ddition to the poetic vocbulry ids in understnding the mening. Although the use of words indicting loction is common in ordinry Ipili speech, it is much more frequent in tindi. Iyu up in line 23 would probbly not be used this wy in n ordinry sentence, but here provides n embellishment. Line 24 illustrtes the repetition discussed by Borchrd nd Gibbs (chpter 8). In this line, only single word is chnged from line 23: kiyl. In this instnce the substituted word is derived from word used in the previous line, kiy, by dding the syllble l. 6 It lso bers some phonetic resemblnce to lliy, one of the words in the two-word expression in the previous line for which it substitutes. Lines form nother sentence. Line 25 is repetition of line 22. Lines 26 nd 27 gin use poetic vocbulry nd the formulic lm epe of line 25, but this time embellished with finl o. Lines 28 nd 29 form nother pir of prllel lines with prtil repetition. In line 28, ote, which ordinrily is n emphtic prticle, seems to be here simple embellishment. Line 30 hs embedded direct discourse, beginning with n introductory interjection de well nd ending with o de, which seems to be n expression to invite response from the listener. Line 31, becuse it is continuing the thought in line 30, omits the introductory expression in 30. It should be noted tht the piko of line 31, normlly trnslted s tht, hs the function here of mking the cluse of line 30 the bckground for line 31 nd could be trnslted since or nd so, giving the two lines the mening I hve plnted enough, nd so I will build big house. Line 32 contins the phrse ondo tpe, which ws trnslted wit, but occurs in tindi much more commonly thn in ordinry speech. Its function is not cler. Chnt The chnt melody is generl pitch contour for line. The pitch rnge vries from one chnter to nother nd is within the chnter s norml speking rnge. The chnt typiclly begins with rising pitch on the first syllble. Within the line there re normlly series of syllbles t more or less the sme pitch, but there my lso be rises nd flls internlly. Often lines re chnted in pirs with the 6 Sets of such poetic words re lso used in trditionl songs. See Ingemnn 1968 for further exmples of sets of these words. 201

6 Sung Tles from the Ppu New Guine Highlnds first of the lines rising t the end nd the second flling. These pirs of chnted melodies do not necessrily coincide with the repeted prllel lines of the text. Figures 1 to 6 represent pitch movements in lines from three different chnters. 7 Figures 1 nd 2 represent the pitch chnge in the first two lines of the smple pssge given bove. Note tht in the first line the pitch rises on the first syllble, drops in the middle portion to sustined mid level, drops towrds the end nd then rises. At the beginning of the second line the pitch contour is similr but t the end hs high pitch followed by low one w mb y l e p u y nd e n e s i l m e p e Figure 1. Alu (1965:line 22): wmb yle pu yndene-si lm epe where he plnted s he hd done before, it ws sid. An udio file of this exmple cn be found in online item mb o k y i u k i y l l i y t u p y o e n p l e Figure 2. Alu (1965:line 23): komb iyu kiy-lliy tup yo pene le he plnted komb, it ws sid. An udio file of this exmple cn be found in online item These figures were produced by the Prt coustic nlysis progrm. Minor fluctutions in pitch should be ignored since they represent influences of consonnt nd vowel rticultions. There my lso be other fluctutions ttributble to lexicl tone, which hs not been fully nlysed nd is not mrked in the trnscription. 202

7 9. The Structure of Chnted Ipili Tindi Figures 3 nd 4 show the chnt melody s produced by nother tindi teller. These re the first two lines in Borchrd nd Gibbs s figure 11 in chpter 8. Like the contour in figure 1, figure 3 displys rise to high pitch t the beginning, grdul descent, nd rise to higher level t the end. Figure 4 hs similr contour but the end does not hve rise w n t u p i y u y ng i k u m n e n e ng p i t u t o Figure 3. Knend (1965:line 827): wn tup iyu yngi kum nene-ng pitu-to girls sitting up on Mt. Kum. An udio file of this exmple cn be found in online item w n t u p k u l m e w ng e n e ng p i t u t o Figure 4. Knend (1965:line 828): wn tup kulme wngene-ng pitu-to girls sitting up on Mt. Kulme. An udio file of this exmple cn be found in online item 13. Figure 5 represents the pitch contour of third tindi chnter. The line is the first one in text 26 in Borchrd nd Gibbs (chpter 8). This chnter hs lower pitch rnge thn the previous two, but the contour is much the sme s in figure 2, with rise nd fll t the end. 203

8 Sung Tles from the Ppu New Guine Highlnds i y p i ng o l l p o l u w nd p i p Figure 5. Yndpke (1964b:line 24): yi pingo-l lpo luw ndpi-p both previously mentioned pig nd dog. An udio file of this exmple cn be found in online item 14. Figure 6 represents line t the end of segment s chnted by the sme mn whose lines re shown in figures 3 nd 4. The min prt of the line is similr to non-finl lines, but ends with the level finl phrse lei tup y nd very long o with rise t the very end. 180 o y ng ii l 160 o m 140 p n u e k nj i d u nd e y n l ei t u p y o Figure 6. Knend (1965:line 68): molo yngi ip nnji dundu keyne lei tup y-o he climbed up over long Nnji River, they sid. An udio file of this exmple cn be found in online item 15. Sets Sets of lines bsed on repetitions hve been discussed by Borchrd nd Gibbs (chpter 8). These sets re usully two or three lines in length, but there re some longer. There is one plce within the line where one or more words differ, but the end of the lines normlly remins the sme. Such sets re illustrted by lines 23 24, 26 27, nd in the pssge from Alu given bove. The word tht chnges my be norml vocbulry or poetic word used only in tles 204

9 9. The Structure of Chnted Ipili Tindi nd trditionl songs. Often the portion tht chnges is the nme of plce, cln, or person. Some of the specilized poetic vocbulry is well known nd understood by ll listeners. For exmple, lnem is frequently used in plce of yi pig, nd gulupi nd miyl re used in repetition sets in plce of n stone, moon. Other words or phrses were not understood by Ipili spekers who helped trnslte the stories. At times the tindi teller helps the listener by dding clssifier word such s si tree. Frequently in the first two lines of set, there is phonetic resemblnce between the words tht lter the line. The most common wy of forming the second of these words is by dding syllble. For exmple, Kim, the nme of min chrcter in tindi, becomes Kimpe in repetition set. Content Although tindi stories differ in detil, there re number of events nd descriptions which occur in more thn one. The min story line usully recounts how young mn develops into n dult, gets wife, nd eventully disppers into the clouds, which those helping trnslte the tindi explined ment going to heven. The young mn is usully living lone in n isolted re, sometimes with brother nd occsionlly n ging mother. Frequently, the young mn is described s being scrwny nd unttrctive. At some point in the story, he meets young womn up in the mountin forests. After spending some time with her, he turns into strong, good-looking mn nd receives from her items of personl dornment so tht he returns home well-developed nd well-outfitted mn. This prt of the story is reference to the time when boys t puberty would leve inhbited res to live in the mountin forests, undergo ritul clensing, perform mgic spells so tht their hir would grow long enough to form the lrge trditionl hir style, nd return to the inhbited re s dults. Tindi usully include pssges bout mking grden, going hunting, meeting young womn, pying compenstion to the bride s fmily, nd going to dnce. These events contin similr detils nd descriptions from one tindi to the next. The interweving of these conventionl elements nd descriptions fcilittes the recretion of the tle. Conclusion Melodic lines, segmenttion, conventionl phrses (prticulrly t the end of lines nd segments), repetition, nd poetic vocbulry re ll fetures of n Ipili tindi. Tindi tellers differ slightly in style, but they hve enough in common to mke the Ipili tindi well-defined genre. The use of conventionl themes, stock phrses, nd repetition fcilitte the recretion of tindi ech time it is performed. 205

10 Sung Tles from the Ppu New Guine Highlnds Acknowledgements I m deeply grteful to ll the Ipili spekers who t vrious times over period of more thn forty yers hve helped me to understnd their lnguge nd culture. In prticulr, I thnk the tindi tellers, whose performnces formed the bsis for this chpter, nd the interpreters who helped me in the lborious tsk of trnscribing nd understnding the tles. My work on the Ipili lnguge in the 1960s ws funded in prt by the Luthern Church Missouri Synod, the Wenner-Gren Foundtion, the Ntionl Science Foundtion, nd the University of Knss. I lso pprecite the help tht I received from the Ppu New Guine mission of the Luthern Church Missouri Synod in getting to nd from the Ipili re, s well s housing nd support while living there. Mny New Guinens re to be thnked for providing housing nd other ssistnce s I hiked round, recording lnguge smples s well s the tindi nlysed in this chpter. I hve gretly benefited over the yers from insights into the Ipili lnguge generously provided by Terry Borchrd. Also, I cnnot thnk him nd his wife, Jnet, enough for their hospitlity nd ssistnce with trvel rrngements in recent yers, without which I would not hve been ble to continue to work with Ipili spekers to unrvel the complexities of the tindi. References Alu Recording trnscribed nd trnslted by Frnces Ingemnn. Ingemnn, Frnces The Linguistic Structure of n Ipili-Piyl Song Type. In Proceedings: 8th Interntionl Congress of Anthropologicl nd Ethnologicl Sciences, Tokyo nd Kyoto, Vol. 2: Ethnology, Tokyo: Science Council of Jpn. Knend Recording trnscribed nd trnslted by Frnces Ingemnn. Pinyti Recording trnscribed nd trnslted by Frnces Ingemnn Recording trnscribed nd trnslted by Frnces Ingemnn. Yndpke Recording trnscribed nd trnslted by Frnces Ingemnn b. Recording trnscribed nd trnslted by Frnces Ingemnn c. Recording trnscribed nd trnslted by Frnces Ingemnn. 206