Problems of transcribing -97
Ilpo Saastamoinen (22.10.1997):
Problems of transcribing Kola Sámi music.
The Kola Sámi Musical Tradition -project by Tromsø University is coming to the end after four years’ field- and transcription work. Instead of the presupposed 200-300 recordings the result was over 600 tape recordings (DAT) of leudds, luvvts and other song types - approximately 18 hours of vocal music from 1994-97.Only the Scolt Sámi part of the material is huge - about 500 pages of transcribed music (as manuscripts - including song texts) from three Scolt Sámi women living in Upper Tuloma (Verhne-Tulomskij) and Tuloma in the western part of Kola peninsula (Russia), near the Finnish border. Some leudds were also recorded in Joona. The collected Kildin Sámi material from Lovozero, Revda, Loparskaja and Teriberka is not as extensive, but in certain aspects it includes some very archaic elements. There were some other people with ability to sing luvvts - according to rumours for example in Gremiha - who couldn’t be recorded by us within the frames of the project. Therefore we were able to record only one Ter Sámi singer. In this case there are for the times to come only older Ter Sámi recordings chiefly made by Estonians during 70’s and 80’s.
The Kola Sámi music is very little studied. Besides only one book (by Irina Travina) there are approximately 30 articles - most of them very short ones written in Russian and sometimes concerning not so much music itself. The basic problem is, that there are no native ethnomusicologist who could speak Scolt or Kildin Sámi. In the project group there were - besides of Norwegian (understanding North Sámi) and Finnish partners - one Kildin Sámi and one Ter Sámi (speaking Russian), two Estonians (Russian, Finnish, English and understanding partly Eastern Sámi languages), three Scolt Sámi (speaking Finnish,) and one Finnish (the only one being able to write Scolt Sámi language but not understanding Russian). The informants themselves in general couldn’t write their own native language.
The difficulties in transcribing the text :
The final musical transcription is in this case impossible without having the song text. After getting most of the Scolt Sámi material (and a part of Kildin Sámi material) I have made a list of general problems, which have arisen during the transcription and translation work. From the data material I keep in mind specially the improvised epic songs, leu’dds and luvv’ts , because they consist the most difficult part of the material from the point of view of transcription.
1A) The transcribers don’t have cassette players good enough!
1B) The singer herself doesn’t recognise (afterwards from the tape), what the words were in the original moment of the performance .
2) In the song text there are Russian loan words, whose meaning is not known by the Finnish Scolts.
3) There are mistakes concerning the names of the people mentioned in the songs.
4) The family relationships are sung wrongly or not mentioned at all, which makes it more difficult to understand the logic of an epic personal song.
5) The lack of key person names, which leads to difficulties in understanding who is saying something to whom.
6) The linguistic time forms are unclear.
7) The lack of basic linguistic words (predicate, subject) in the sentences.
8) There are a lot of ’meaningless’ syllables (or even words) in the middle or in the end of other words. Sometimes this kind of endings can change a ’fact’ sentence to a question (’go’ -ending).
9) The dialectal differences of an oral culture in personal or place names can create an unbelievable amount of different versions in the pronunciation:
A) The names are sung as nick names or with diminutive forms in different situations.
B) An original Scolt Sámi name has a Russian form (or only written with Cyrillic alphabet)
C) ...Kildin Sámi version (written with Cyrillic alphabet)
D) ...Finnish Scolt Sámi version (written in literary Scolt Saami form)
E) ...Finnish (translated) version
F) ...written English version (translated or not)
G) The names (or words in general) how they are heard by a Scolt, Finnish Scolt, Kildin, Estonian etc., for example the existence or non-existence of the letter ’r’ in the word ’Nuortti-’.
H) The lack of teeth, when an old singer should sing the letter ’r’....
10) The pronunciation when singing is different from the spoken language.
11) The symbolic ways of expression:
A) A realistic deed in a song is sometimes also a representation of an abstract feeling.
B) The - possibly secret - text is there only for ”showing the melody”.
12) The changes in the language since the dividing of the Scolt Sámi people during World War II by the Finnish-Russian border line.
13) The incompetence of the Scolt Sámi people to write their language. On the Russian side the problem is even deeper, because there has been only the Kildin or Ter Sámi way (with Cyrillic letters) to write Scolt Sámi.
14) The singer has chosen a ’wrong’ melody and cannot remember the words.
15) The unconsciousness in dividing the words into syllables. It happens often that the singer places the last letter of the former syllable at the beginning of the next syllable. Who is right? The same happens with the words, when the first letter of the next word is tied to the end of the former word.
16) The difficulties in the terminology. How to translate the different boot- or reindeer terms into English? How to translate the ’akka’ -ending (the ’hag’) in the local place names?
17) The lack of suitable special fonts in the computer concerning the Cyrillic way of writing Kildin Sámi language.
18) In two cases the singer him/herself is died before the text transcriptions have been checked in the co-operative session.
The problems in transcribing the rhythm :
1) The lack of the steady ’poetical’ meter in the text. The text doesn’t give in this case any regular advice in defining the musical structures. Usually it seems, that the words in general begin on the same place where also the main beat of the musical measure seems to stay. The music and the words are in this way in a certain connection with each other.
2) The musical and the textual structures are not identical. This is the case especially with improvisational performances, where those structures are only partly overlapping with each other. The complexity of music consists of the different but simultaneous levels of textually, melodically and rhythmically individual structures. There are many similar examples in the Northern Sámi yoiks.
3) The variation technique of improvisations makes the phrases dissimilar - especially concerning the length of the phrases. Every phrase has a new rhythmical structure.
4) There is a lot of ’swaying’ between 5/8- and 3/4 -meters. In some cases the ’objective’ meter is not existent at all. The both possibilities can be in ’fifty-fifty’ balance with each other but nevertheless occur in ’chaotic’ disorder. We should estimate this as one conscious method of improvising, as a game between probability and disorder, rule and exception, order and chaos, following and immediately escaping any rules, imitating and creating, giving birth and killing...
The length of the tones are ’fuzzy’. This can happen only in an oral culture without written notes or any systematic music theory. The same concerns breathing pauses which sometimes can be a part of the musical structure.
5) The meter can seem different depending on the level of figuring. On the macro level the music seems to be in 3/4 meter but - being listened with the half speed - turns out to be in 5/8 meter. Somehow this reminds us about fractal structures working as a source of new melodies.
A western ear catches only the audible macro level. Therefore most of the Sámi music in general sounds like being ’rhythmless’. The normal western musical ear cannot remark the changing of 2/8’s and 3/8’s - especially if it is irregular.
6) The biggest problem with the mixed rhythms is to configure the most reasonable ’right’ combinations of 2/8’s and 3/8’s. This is connected with the text analysis, because the unstressed syllables lie usually on the off-beats. This is almost a rule concerning the meaningless syllables or short words like ’a’ or ’go’. An ’A’ can even lie on the off-beat - just before the bar line - beginning a new phrase. If we put it after the bar line - like some one-syllable words in many transcriptions of Estonian runo-songs - the result may seem quite complicated from the musical point of view.
7) The lengths of decorative tones or the long tones with a large vibrato can easily confuse the rhythmical steadiness. We have to keep in our minds if we are trying to document melodic lines and structures of the music or trying to document the personal vocal technique of an individual singer. In many cases the latter is possible only with the help an audible form of a recorded tape. The western way of writing notes doesn’t fit to the ethnic music with a high amount of personal intonation style.
8) The so called speech song style with long glissandos is a part of Eastern Sámi vocal tradition especially connected with spoken tales including occasional cuts of singing. These performances do not easily surrender to the transcription.
9) The changes in tempo should be treat as a relative phenomenon. The effort trying to mark the absolute durations leads to a chaotic view of the music. We cannot make any comparisons of any phrases, because we are in tremendous difficulties in recognising any adequate motives or other forms.
Melody: How to define the correct pitch and phrase length?
1) The normal western ear is acquainted with totally other types of melody, which consist of diatonic scales including lead- and basic tones. With these we are accustomed with fixed amount of bars, which are grouped in pairs (or ’triplets’). In the Eastern Sámi music we’ll meet endlessly changing lengths of short motives and longer phrases.
2) There are no harmonic I/V -relationship in the melodies which consist of pentatonic scales but as well only of three or four tones. This means difficulties in choosing ’the right key’ from two or even three equal possibilities.
3) As a result of the two first reasons there could be found melodies, whose beginning and end are ’fuzzy’. We cannot decide the exact frames of a phrase. It means, that for example the bar lines can be put as well one beat back as forward from the first chosen point. It is a very frustrating experience to the transcriber, who would like to treat the material from the ’objective’ point of view. The idea of one absolute reality turns out to be pure romanticism. In this case it is best to trust to the text with its natural stresses.
4) The ’fretlessly’ ascending pitch is a typical feature all over the Sámi vocal music. The way to avoid this problem is treated already in the connection with rhythmical problems (N:o 9).
I strongly criticise the idea of writing down the absolute pitch of the melody in the cases where the key of the melody keeps gliding up ’fretless’ from the top of the song to the very end of the musical performance.
The regular ascension of the pitch is actually the same as the steady acceleration of the tempo: they don’t have any real information after they are remarked. We just tell - outside the note sheets - how much in a certain period this kind of changes occur, which are the minimum and maximum limits of the change in the pitch or tempo.
5) The situation with sudden ’jumps’ in the pitch is a little bit different. Normally these happen downwards and demand a totally new key. There could lie some relevant information in the ways how big those jumps are and when do they occur, what kind of probability lies in their occurrence.
In order to make comparisons between the musical phrases etc. we are nevertheless allowed to keep the music in one key after finding some possibly interesting structures in the periods and widths of irregular jumps. If a song is cut into many fragments in different keys, they should treated just like separate songs in the chain of different keys. One possibility is to write the song with a system consisting in two rows; the first one in a compressed form - in one key - and the second one with absolute pitches.
6) The possible existence of the micro intervals in the scales. These are important in two ways:
A) They can be fixed: They are easily recognisable when the general pitch is not ascending ’fretlessly’. They could be marked with extra symbols pointing the quarter tone pitches etc.
B) They are the structural element of the fretlessly ascending pitch. In this case the place of the micro intervals can be anywhere inside of the melody. In practice they are not possible to define. It is best to forget them, because even the quarter note pitches are as misleading as to keep everything in one and the same key.
7) The pitches of the strongly vibrating tones are sometimes confusing because they often consist of falsetto voices or are otherwise unclear in the intonation.
8) Sometimes one meets the mystery of occasional ’two voice’ sounds. You cannot hear two different pitches at the same time but you’ll find two different versions of the ’right’ pitch. This case has nothing to do with micro intervals or two voice singing technique but the result of acoustic conditions (microphones etc.).
9) Inside the spoken stories the storyteller may sing just a pair sentences and continues the speech immediately. Sometimes the singer may speak a few words in the middle of the singing. The latter is a very common way. In these cases there are strong glissandos and a lot of pitchless tones in the melody. It is remarkable, that the recorded discussions or interviews include sometimes spoken cuts, which sound like being sung.
10) In some cases there are totally atonal singing, which has adopted from the tradition almost all except the style of singing ’in tune’.
11) Heterophonic singing with two participants trying to sing in unison is very problematic. They just don’t know, how the melody should be sung, because everybody has her own words and melodic interpretation from the ’same’ song. The result is automatically chaotic. Normally it’s no use to try to transcribe it. The personal interpretations can be documented better, if the recordings are made from separate personal solo performances. The duo performance can undoubtedly be a very strong experience to the audience, but it is too heavy a challenge to a transcriber. In 1993 we recorded in Mordovia seven women, who sang funeral laments at the same time and everybody had her own melody, words, tempo, scale, rhythm and key...
The results and conclusions.
1) The transcription is always a personal choice. Therefore it is always subjective in pointing certain features and leaving some others untouched, to the background.
2) Any longing for some basic ideal root forms in the performances is at least doubtful.
In our hands we have only some reproductions of a general idea, which itself is only a virtual abstraction, not existent in this reality. What is real, is only some frames in the game between certain rules and exceptions, which themselves are not music at all.
The idea about two identical versions of the same song is even theoretically impossible, because the audience, the situation and the surroundings are changed with every performance, even from the moment to the moment. This concerns as well the documented audible forms like tapes or any other kind of recordings as the transcriptions.
This ’fuzziness’ doesn’t concern only the music, but it is well seen in the ways to treat words, names etc. in the spoken language - as we have found out earlier. This instinct, drive of personal variation is a basic feature of the individual life - the only life worth of living.
In this context the ethnomusicologist’s point of view remains only a narrow glimpse of something, which itself is a tremendous huge package of information being able to give thousands of possibilities for unique recoding.
3) I’m afraid that - concerning the Eastern Sámi music in general - the scientifically documented material will be the only tradition carrier in the future. Therefore we have to accept this narrow attitude and try to get out of it as much relevant information as possible.
An oral culture presumes the rules, which are easy enough to learn by heart. Therefore the simplicity of a structural, rhythmical or melodic rule is of the greatest importance. One important aim of analysis is to find the simplest possible (symbiotic?) formula of/for both text and music. The one cannot be ruled by another.
4) The result of an analysis should be able to give basic advice how to reconstruct an improvised song - at least partly in the traditional way. The main goal shouldn’t be in showing the complicity of the music but in finding the simplest possible way to understand the common structure of text and music.
5) The rules have to be simple, because the singer must get rid of any intellectual level of consciousness when performing. Instead it demands an ability to loose the rules from the conscious mind just like in the process how to learn to swim.
6) The written transcription is not the right way to offer methods for vocal intonation.
7) The basic aim in the long perspective is to lighten the question, how to teach a new generation to improvise so, that the local tradition is recogniceable from the performance. Simultaneously it means, that one has to recover the positively understanding attitude towards the importance of the improvisation in every living culture as the archaic traditional way of the self expression and of strengthening one’s self identity.
About Kola Sámi musical transcriptions.
In 1987 I have written an article concerning transcriptions of Scolt Sámi leudds (”Kolttasaamelainen leudd, Etnomusikologian vuosikirja 1987, ss. 66-95). The article introduces a possibility of understanding the time values of the music as combinations of two’s and three’s. This way of thinking leads to a new point of view, where the music is seen as a game between chaos and order. The musical structures seem to be created by chance, but as the result we can find the total balance between rules and exeptions. Especially this is valid in the improvisational singing.
From this point of view I try to analyze a transcription made my Irina Travina (”Mierre jodtjem”; N:o 11 in Travina’s book ’Saamskie narodnie pesni, Moskva 1987).
I’ll introduce some critics concerning:
1) The idea of writing the absolute pitches instead of keeping everything in the same key - in the case, where the pitch keeps gliding upwards fretlessly, from the top to the end of music.
2) The relationship between the text and the music. The main goal shouldn’t be in showing the complicity of the music but in finding the simpliest possible way to understand the common structure of the text and the music.
It means seeking after simplicity instead of complicity as a background of creating music.
As a comparision I introduce my own transcription of the same luvvjt-song analyzing the structure as a game between rule and exception. (Mierre jodtjem -luvvjt from Kola)
.....
Päiväkirja:
3.10.97 - Tein valmiiksi esitelmän Kuolan musiikin transkriptio-ongelmista.
Ko. esitelmä oli tehty Jyväskylässä tapahtunutta seminaaria varten. >
K 16.-19.10.97 ESEM-kongressi Jyväskylässä.
K: 17.10. klo 16 oma esitelmä J:kylässä ESEM-kongressissa
= 'Problems of transcribing Kola Sámi music'.
> Rüütel, Leisiö ym. meillä vieraina
Improvisointi on:
SÄÄNNÖN - POIKKEUKSEN
JÄRJESTYKSEN - KAAOKSEN
0-INFORMAATION - INFORMAATION
TRADITION - LUOVUUDEN
VANHAN - UUDEN
RYHMÄTUNNON - ITSETUNNON
epäsäännöllistä vuorottelua - siten, että kuuntelijan ennakkoarvaukset ovat mahdollisimman tasapainossa oikeiden ja väärien arvausten suhteen.
JOHTOPÄÄTÖKSIÄ (28.03.06):
1) Teksti- ja musiikkiaiheet elävät kukin itsenäisinä palasina omaa elämäänsä. Anfissa Gerassimovan improvisointi pohjautuu suurelta osin sille, että
A) Saman lauseen sijainti musiikissa eri esityskerroilla vaihtelee
B) Lauseen kertautuminen ei ole sidoksissa musiikillisen fraasin kertautumiseen. Päinvastoin - kertaukset tuottavat uusia kokonaisrakenteita. Esim. Teksti A saattaa esiintyä melodia-aiheen A tai B kanssa. Toisaalta Tekstit A ja B voivat esiintyä vain melodiatyypin A kanssa. C) Yhteenveto: Tekstin ja melodian rakenteiden päällekkäisyys luo jatkumon, jossa mikään ei toistu täsmälleen samana. Sääntönä voisi olla pikemminkin: Jos tekstin rakenne vaihtelee (eri esityskerroilla), niin musiikin rakenne pyrkii pysymään samana. Tai jos tekstin rakenne pysyy samana niin melodian rakenne pyrkii vaihteluun.. Luonnollisesti on mahdollista, jopa tavallisinta, että molemmat rakenteet vaihtelevat.
2) Motiivien variaatioissa sävelkorkeudet saattavat muutella keskinäisiä paikkojaan siten, että variaatiot on kuitenkin katsottava pääpiirteissään identtisiksi. Jos meillä on motiivissa kaksi säveltä järjestyksessä d - e, niin variaatiossa järjestys voi olla e - d, tai sitten vain jompikumpi sävel esiintyy. Ehtona on tällöin, että ympärillä olevat muut sävelet tai fraasin rytminen kuvio säilyttävät samaksi aiheeksi tunnistettavuuden. Käytännössä tämä tarkoittaa myös sitä, että sävelkulun yleinen suunta saattaa olla riittävä ehto identtisyyden havaitsemiseksi. So. ei ole tärkeää se, paljonko sävelkulku hyppää ylöspäin, vaan se, että se yleensä hyppää ylöspäin oikeassa kohdassa. Kun melodian kaari on tärkeämpi kuin intervallit sävelhahmon tunnistamisessa, niin silloin samasta laulusta saattaa löytyä jopa duuri- ja molliversio. Tämän tyyppisiä tuloksia syntyy erityisesti silloin, kun 4-5 -sävelikön alkusävel sijoitetaan eri variaatioiden nuotinnoksessa perussävelen, terssin tai kvintin kohdalle.
3) Ilmeisesti tekstin muistumattomuus mieleen on yksi improvisoinnin tärkeimpiä tekijöitä. Muutoin A-osalle tarkoitettu teksti esitetäänkin vasta B-osassa, jos tekstiä ei heti muisteta. Mietiskelyn aikainen fraasi lauletaan tällöin merkityksettömillä tavuilla tai edellisen lauseen kertauksena.
4.12.1994
- Jokin musiikkiseminaari Jyväskylässä tai sitten muistiinpanoja Tallinnan seminaarista.
Joka tapauksessa ennen 27.03.1995, jolloin "Seminar on (Sami) Identity", Helsinki - ks. Vihko XXXB, s.16
1) Ilman analyysia transkriptiosta ei ole apua.
2) Analyysi tarvitsee filosofian: Mitä minä etsin tästä nuotinnuksesta?
3) Eräs ensi askeleista on: Onko tämä toistuvaa vai onko tämä improvisoitu?