Karelian Kantele Tradition

Research

Background

The five-string kantele, hollowed out from a solid piece of wood, is regarded as the oldest model of Finnish kantele instruments. The exact age is under ongoing evaluation and debates, estimates varying from 1,000 to 3,000 years. The reconstructions by Vladimir Povetkin from the pieces of instruments found in archaeological excavations in Old Novgorod have given rise to the theory that Baltic Psaltery has evolved from a lyre with a hand-hole during the 11th century. [Povetkin, Vladimir 1992: Musical Finds from Novgorod. The Archaeology of Novgorod, Russia. Ed. Mark A. Brisbane. Lincoln, UK: Society for Medieval Archaeology.] However, there are also many arguments given the assumption that the kantele is originating from much earlier period than that of Novgorod’s findings.

The recognized runo-singers were often also known as skillful kanteleplayers

It is clear that the historical research of an instrument should not be separated from the culture in which the instrument has been used. In Finland and Karelia (meant here as: Viena Karelia, Northern Karelia in present-day Finland, Ladoga Karelia, Karelian Isthmus and Olonets Karelia) kantele music with runolaulu (Kalevala metre runo-song), laments and incantations, was an inseparable part of the ancient aural music culture, which was found as a still living tradition in isolated Karelian villages at the end of the 19th century and even at the beginning of the 20th century. The recognized runo-singers were often also known as skillful kanteleplayers, and although runo-song and kantelemusic are mostly recorded as independent performances, there are several mentions of them being performed together, as well. With Heikki Laitinen’s words; the runo-song and the kantelemusic of that time had the same aesthetics – kantele music had the power of the song, and in the melodies of runo-songs the kantele music was clearly audible. [Kantele, ed. Risto Blomster, 2010. SKS.]

All the way to the 17th century the runo-song, was the only actual form of song used in the Finnish-speaking area. As Matti Kuusi has written, it is paradoxical that there are so few early, and so voluminous amount of late written documents about the ancient runo-song, originally preserved only in the memory of performers. The earliest recordings are from the 17th century, but the vast collection work happened during the 19th century and at the beginning of the 20th century. During 1908-48 the Finnish Literature Society published a series of books called Suomen Kansan Vanhat Runot (SKVR), containing Kalevala metre poetry in 33 volumes with over 85,000 poems and 1,270,000 lines. (Volume 34 was published in 1997. Nowadays the poems are available also online: http://dbgw.finlit.fi/skvr/)

After over 150 years and countless amount of research on the Kalevala metre runo-song with the mythical, heroic and shamanistic features, the scholars seem to be unanimous about the observations that the history of the tradition and the mythic themes are stretching back thousands of years. Like the rock carvings on the eastern shores of Lake Onega and southwesterly shores of the White Sea, the prehistoric rock paintings found in the area of present Finland, dated to ca. 5000-1500 BC, are connected to the motifs used in runo-songs. Antti Lahelma writes in his dissertation:

The art can be confidently associated with shamanism of the kind still practiced by the Saami of Northern Fennoscandia in the historical period. Evidence of similar shamanistic practices, concepts and cosmology are also found in traditional Finnish-Karelian epic poetry.” [Lahelma, Antti 2008: A Touch of Red. Archaeological and Ethnographic Approaches to Interpreting Finnish Rock Paintings. The Finnish Antiquarian Society. Helsinki.]

Matti Kuusi emphasizes the significance of Baltic influence in the birth of Kalevala metre proper. Parts of his thorough studies of the history of the runo-song are published in English in an anthology, entitled Finnish Folk Poetry, Epic (1977). Also a great source of information for English readers is Anna-Leena Siikala’s and Sinikka Vakimo’s edition: Songs Beyond the Kalevala. Transformations of Oral Poetry (1994. SKS/The Finnish Literature Society).

One of the most intriguing epic songs is the mythical song about the origin of the kantele

One of the most intriguing epic songs is the mythical song about the origin of the kantele and the supernatural power of the music played by its creator. Heikki Laitinen has pointed out how the Kantele-songs, used sometimes also as incantations, were among the first epic songs to be collected and published, and on the Finnish-Karelian-Estonian area they were also more widespread and survived longer than any other motif. In the SKVR there are slightly over 400 Kantele-songs collected from Northern Ostrobothnia, Lappland, Middle-Finland, Kainuu, Savonia, Karelia and Ingria. According to Matti Kuusi, the Kantele-song belongs to the same layer with other ritual songs like The Creation of the World, The Oak, The Origin of Fire, The Origin of the Bear, The Origin of Iron, the Primeval Boat, and they can be traced back two or three thousand years. The following version of the Kantele-song is scanned from the book Finnish Folk Poetry – Epic, ed. by Matti Kuusi, English translation by Keith Bosley and Michael Branch, Finnish Literature Society, 1977. The singer is unknown, and the text is copied from the manuscript of an unknown collector by K. Ganander, ca 1760. (By clicking the images you will see them larger.)Kantele-runo_Ontrei Vanninen

Kantele-runo_Ontrei Vanninen_001

Kantele-runo_Ontrei Vanninen_002Matti Kuusi concludes that the Kantele-song, culminating in Väinämöinen’s supernatural kantele playing, is an ingenious accomplishment from the first millennium AD. He also assumes that Kalevala metre gained its strongest position and its finest forms after it was adopted in Ancient Karelia in the 9th century through the colonists moving into western and northwestern shores of Lake Ladoga from Häme and Western Finland. In her dissertation, Ancient Karelia, Archaeological studies (1997), Pirjo Uino summerizes:

The hypothesis of colonists moving into Karelia from Häme and Western Finland during the Merovingian Period and the Viking Age is supported by the available archaeological material (antiquities representing the western religious tradition, western artefacts), a stratum of West Finnish place-names and linguistic data. The population which formed when the western colonists merged with the “Proto-Karelian” autochthonous population is here given the term “Pre-Karelian”. Behind the colonization from Western Finland is an economic boom which began in the Lake Ladoga region around the end of the Merovingian Period. (p.204) — During the Crusade Period, the identity of the population strengthened as it came to be regarded as “Karelian” both by themselves and the world around them (p.130).

Uino also points out how the people on the area had contacts to many directions over the course of history. The archaeological findings from the period AD 400-800 show contacts in West Finland, the East Baltic regions, Scandinavia and Eastern Europe. Uino concludes that the rise of the Ladoga region to become an important market place must have started around the end of the 8th century, because the number of metal artefacts increased during that time.

The beginning of the period 800-1100 (of this period Uino uses the term “Karelian Viking Age”) is signified by the emergence of 9th century Scandinavian artefact forms. The cemeteries of the period are linked with the West Finnish tradition but there is also multinational character of the material, like Scandinavian bracelet, Baltic weapons, Fenno-Russian axe. As there were people moving from Western Finland and Häme towards Karelia, the Slavs’ expansion extended to areas south of Lake Ladoga, and Scandinavians settled along the rivers in connections with their journeys towards the south and the southeast.

During the Crusade Period (AD 1100-1300) the settlement expanded from the water routes to the lakes and smaller rivers of the inland, but not to the upper areas further away. The islands of Sortavala in Lake Ladoga, as well as the regions of Antrea and Jääski on the upper reaches of the River Vuoksi, were settled, and the Viipuri region appeared as a new, emerging centre. Most of the fortification finds are from the Crusade Period and the Early Middle Ages. The most remarkable ones, the Old Fortress of Käkisalmi, Tiurinlinna in Räisälä and the Castle of Viipuri, were also settled centres of trade and crafts. According to Pirjo Uino, the richly furnished log-framework graves were those of Karelian allies in loose confederation of peoples led by Novgorod. These Karelians had a trader organization, and their contacts extended over Russia, the Baltics and Northern Fennoscandia.

About Karelian kanteleplayers at the beginning of the 20th century

With the support of a grant from the Kone Foundation, in 2011 the team Anna-Liisa Tenhunen (Ph.D.), Rauno Nieminen (D.Mus.) and Arja Kastinen (D.Mus.) begun a research project on the early 20th century Karelian kantele players and their music. Anna-Liisa collected the details about the players and their lives. Arja wrote new musical notations from the manuscripts and researched museum kanteles. Rauno took photos and made precise measurements of selected instruments (23), and also replicas of some of them. The results are published (in Finnish) as a book KIZAVIRZI karjalaisesta kanteleperinteestä 1900-luvun alussa (Temps Oy, 2013).

Kizavirzi_etukansi_72dpiOne Hundred Years of Beauty Sleep

During the summers of 1916 and 1917 a young scholar Armas Otto Väisänen made history by collecting 134 tunes from 23 old tradition kanteleplayers at the remote villages of Ladoga and Border Karelia. The ancient tradition was fading away, and as a skillful musician Väisänen was very well aware of the differences between the contemporary music of his time and these silently breathing strands of ancient music culture. Later he met and recorded some more Karelian kanteleplayers, supplemented the collection with earlier transcriptions from other collectors, and the book Suomen Kansan Sävelmiä VKantele- ja jouhikkosävelmiä was published in 1928 by the Finnish Literature Society containing 232 tunes from 50 kanteleplayers. As everyone else, Väisänen must have been thinking that he was preserving some knowledge of the tradition that had no future in practice any more. It would be nice to know what he would had thought if someone had told him that on the 21st century his book would be a treasure chest for modern musicians.

At the beginning of the 20th century it was not the kantele as an instrument that was vanishing, but the ancient tradition. New, bigger instruments were developed to serve the new music better, and there were new kanteleplayers around the country. The remote Karelian kanteleplayers were living in the middle of transition time; they had learned the old plucking technique as a child, most of them knew how to sing runo-songs, but they also knew new type of songs and played them on their kanteles. In addition to the old, 5 to 21-string kanteles, hollowed out from a solid piece of wood, there were also kanteles with over 10-strings that were made of several pieces of wood, like miniature copies of larger box kanteles. Some players had learned the new playing technique, as well, generally used in bigger kanteles, to play the new songs or dance tunes. But even if some used new, big instruments, they still performed the old music with the old plucking technique.

There is something very remarkable in the observation that although elsewhere another old, but totally different playing technique, the strumming technique, was very well known and commonly used by kanteleplayers, from this particular Karelian area not a single such player was recorded. With the strumming technique the music becomes automatically chord-based, whereas the old plucking technique connected to the runo-songs is mainly based on intervals like pure fourths and fifths, and thus gives the sense of more than a thousand year old roots of music. For some reason at this area it survived the longest. It is true that the remote location of many Karelian villages made the new influences to emerge the area slowly and late. It is also known that under the Eastern Orthodox Church, to which most of the Karelian people belonged, it was easier to keep the old traditions (like runo-song and laments) alive longer than under the Evangelical Lutheran religion, which controlled the most of Finland. But in addition to that, I would like to ask the same question as Matti Kuusi (he was referring to the Kalevala language): what exactly did this tradition signify to its users?

Interpreting the notes

At the beginning of the 20th century when Väisänen met the Karelian kanteleplayers, there were almost no 5-string kanteles left; the instruments that old players used had usually 9–14 strings. The most of the tunes that Väisänen recorded were no more than a couple of decades old, some only a couple of years old. Dances like Ristikondra, and Maanitus (Brišatka, Ripatška, Riivattu etc.), for example, were eastern, Russian dances and not very old at that time. Also dances like Hoilola and Kiberä had arrived from the east along the strand of Ladoga. In 1914 Simo Jaatinen from Rautalahti, Sortavala, informed that such western dances like Markkoa and Laputusta were quite common in the 1870’s, but at that time were already mostly forgotten, and that he couldn’t play those new dances like Ristikondra and Riivattu. At the same time Andrei Vilga from Suistamo informed that Maanitus-dance was vanishing but the Ristakondra was still quite common. (Pulkkinen 1946). In addition to dance tunes there were also new songs, runo-songs, and improvisation-based church bell imitations that were collected from the kanteleplayers, but only stories and short references about meditative improvisations lasting for hours, which Väisänen called the quiet enthusiasm.

It seems that however you do it, the result will always be somehow wrong

Although the actual tunes were not very old, several features connected the music to the earlier history: the playing technique, the aesthetics, the way the players used their instruments tuning with octaves, fifts and fourths, and built the music on constant and minimalistic variation of short themes. Making music notations from this kind of music is challenging. It seems that however you do it, the result will always be somehow wrong. The original music was based on oral tradition. Even at the beginning of the 20th century most of the old kanteleplayers couldn’t read or write. They didn’t have any idea about music notation or about the structures on which notation is based. It feels contradictory that we are now trying to learn and play this music via a medium that was so unfamiliar to the original music and musicians, and worse, without being able to hear examples; there are no living tradition bearers to listen to, nor many decent recordings from the old players. Without the understanding of the tradition, its techniques and conventions, the music is interpreted according to modern musical conventions and rules, and the result will be very far from what was intended to achieve.

The Väisänen’s book was written for the musicologists, and the music notations were published according to the music theory and the scientific thought of the time. In addition to the tunes there is a lot of detailed information about the old plucking technique and the fingering solutions and tunings used by different players. All tunes are given with the tonic on G4. In other words, whatever the original pitch and scale, the published notation is always written in G. If the original pitch of the kantele is known, it is stated, together with the player’s personal information. This solution was in a sense brilliant, and in its own way distanced the music from the western music notation. The kantele can be tuned in whichever pitch suits the instrument and the player, who reads the G4 in the notation as the tonic of the scale the kantele is tuned in (= transposes the music). For the old kanteleplayers it was perfectly normal to play the same tunes with differently tuned instruments. Their instruments were usually home-made, so each was unique, and as far as we know the kanteleplayers of the old tradition usually played alone rather than with others. Therefore the pitch varied in each instrument according to its structure, to the player’s preference, and to the prevailing circumstances. By documenting all tunes as if they were in G, they became relatively easy to read and compare, and could be fitted on one stave, and usually without ledger lines. There can be a problem, though, if the user of the book ignores the detailed foreword where all the guidelines are given as how to interpret the music notation. Furthermore, if people try to tune their kanteles, many of which nowadays are built to be tuned with a D4 tonic, to the printed notation’s G4, the result will be a bunch of broken strings. Based on these facts, I decided to write the music in most of the cases placing the tonic on D4.

How long each plucked string will continue vibrating depends on the plucking strength, the structure of the instrument, the material of the string and the playing technique

In the old plucking technique the strings are mainly plucked upwards, and they are only occasionally, if ever, damped. This results in sympathetic vibration, which means that also those strings that haven’t been played start to resonate. How long each plucked string will continue vibrating (and thus, how long it will be included in the music we hear) depends on the plucking strength, the structure of the instrument, the material of the string and the playing technique. Such things can not be noted in transcribed notation without it becoming so complicated it’s virtually unreadable and unplayable, and so just the bear bones of the melody are written down: which strings are plucked, that is, which notes are played in each moment.

In addition, all notes are written as if they were equal in volume, which they in practice are not. The music has many layers from which the musician can selectively highlight certain elements. The high notes played with the thumb stand naturally out, as do strong intervals like the fifths and the octaves, strong chords and notes hit with the fingernail. Besides those, one can purposely emphasize, for example, the medium-pitched tones or build melodic phrases on the low strings. Except for the fingernail strokes, none of those possible emphases is usually written down in the published notations. In some of the manuscripts there are ambiguous references suggesting that the notes emphasized by the kanteleplayer may have been marked with separate notes and longer durations. In his manuscripts Väisänen introduced sometimes the emphasized tones with red slurs. I ended up using short up ascending slurs and additional stems.

01_Alkulause_Arja_nuottien painotukset

How to interpret the written rhythms is also something worth considering. For example it is possible to hear from some of the old recordings how in some dance tunes the compound of two eighth notes is played so that the first is a little longer than the second one, to make a more suitable rhythm for the dance. Or sometimes if the stress is put on the second of the two then the first one becomes a little shorter. Quite well known and widely used practices among many music traditions. It should be clear that the written rhythms should not be interpreted metronome strict.

02_Alkulause_Arja_rytmien tulkinta

Time signatures and bar lines were unknown concepts for the old kanteleplayers. From the musician’s point of view in some tunes it often seems a better solution for the music to be written down as phrases, as in paradigmatic analysis. The length of the phrases can vary in this music, and if time signatures and the bar lines are left out and the music notation based on phrases instead, it becomes clearer and easier to read for the musician. Paradigmatic notation also helps to perceive the overall structure and the variation that constantly goes on. Most of the collected tunes are constructed from small motifs, which are variated by the player with different rhythms, different interval combinations and melodic variations. In the Väisänen’s book the variations played by the old kantele players are written after each tune as footnotes, which in many cases makes it quite laborious to find and play them, and in the worst case they are totally ignored. Thus the original idea of the music is missed – that it is a living thing, changing in time with the musician, familiar but still always fresh and new.

Here is one example of an old dance tune (by clicking the image you will see it larger). It was played by 68 year old Miinan Domi in Suistamo, Koitto, and was transcribed by A. O. Väisänen in 1917. The original transcription is kept at the Folklore Archives of the Finnish Literature Society in Helsinki. The new transcription is paradigmatic and without barlines.  An example of making variations from this tune can be heard at: http://youtu.be/FQD4GyYqsMg

06_Miinan Domi_Potvalssi otsikollaThe kantele was placed on the knees or on a table, orientated so that the highest string was nearest to the player. The playing technique was special. First, the strings were plucked upwards so that the ringing, either of the plucked string or the others, was hardly ever damped. That, together with the tuning system of (mostly) pure fifths and octaves, causes the high amount of sympathetic vibration. Secondly, the hands didn’t have separate roles of melody and accompaniment but their fingers worked together, in between one other – the right hand thumb playing the highest strings, the left hand fingers (index, middle and often also the ring finger) plucking the middle strings, and right hand (index and middle finger, sometimes also ring finger) playing the lowest strings. This leads to “music that flows like an endless stream” in which it is hard to distinguish between melody and accompaniment tones. Thirdly, when the hands are supported on the sides of the kantele, the approximate plucking point is about 1/3 of the whole length of the vibrating string, and plucking at that position causes almost the maximum amount of harmonics. Sometimes the beat would be emphasized by hitting the string with the fingernail, symbolized as a “horseshoe” above or below the note.

Quiet exaltation

In addition to entertainment, kantele was also used as a meditative instrument. At the beginning of the 20th century when A. O. Väisänen travelled through Karelia and made phonographic recordings and transcriptions of Karelian kantele music, there were not many instruments left and the players were mainly old men. For example, Jaakko Kulju was 81 years old when Väisänen met him in 1917, and this is what Väisänen wrote about this special meeting:
“…Many of us may have experiences of the ‘quiet exaltation’ of folksingers and musicians. I have noticed this from many kantele players. The players’ fingers touch the strings according to the tune, but his eyes do not follow the activity – instead they are dreamily vacant. As one old man from Suojärvi was playing his endless dance tune in a log cabin that was slowly getting dark, I took a photograph, exposing the frame for a rather long time, and marvelled at how he did not blink at all, nor did he pay any attention to my photography whatsoever. He had lapsed into his world of quiet playing. Gradually, as the same tune continued with constant variations, his body began to fall against the table, his eyelids closed, the old man played as if in his sleep. Although I was listening with the curious ears of an obdurate transcriber, I felt enchanted. After this experience it was easy to believe the old man Onoila when he told me of one man from Olonetsia-Karelia who could play so sadly that he made his listeners cry, and so joyfully that he made them dance…” [A.O.Väisänen: Laulu ja soitto kansanelämässä, Musiikkitieto 11/1943, published also in a book: “Hiljainen haltioituminen” (Quiet exaltation), edit. Erkki Pekkilä, SKS (The Finnish Literature Society) 1990.] Referring to Kulju, Väisänen later mentioned how impossible it is to write this music down as it is floating in a continuous stream of variations.

He replied that it was nothing, that he was simply playing his own power

Apart from runo-song melodies and some imitating church bells, the published transcriptions are mainly dance tunes. We have only written accounts of improvisation lasting for several hours. For example, in 1882 some Karelian explorers had such an experience in Olonets Karelia:
“…Having played a few dances, the man fell silent, claiming he knew no more. The kantele player was forgotten. There he sat alone in his corner, his kantele on his knees, staring in front of him. Gradually he began to play, softly, at times with a greater display of feeling and warmth, at others so that he was barely audible. On being told that this was precisely what we wished to hear, and being asked what he was playing, he replied that it was nothing, that he was simply playing his own power. We, therefore, left him in peace, each going about our business. But the kantele player, no longer asked any questions and being left to his own devices, played throughout the evening, and what a pleasure it was to listen to him…” (Relander, O.: “O. Hainari. Muistelmia” [O.Hainari. Memoirs]. Helsinki 1917)

Another interesting fragment appeared in the newspaper “Raja-Karjala” 15.7.1911
“…Sometimes you meet such kantele players, who play their ‘own power’, that is whatever comes to their mind. Such a player plays for all kinds of situations in life, he plays farewell tunes differently for different people, he describes how the ship is going at sea, he accompanies the stories he is telling, he describes his moods…”

Karelian kantele music was part of an ancient oral culture. The music was based on variation and improvisation. As most of the music outside the systematised Western music culture, it is problematic to interpret it merely on the basis of notation. Väisänen was clearly aware of these problems and he wrote accurate directions how these notations should be interpreted. For example, he gave variations in his footnotes, and he indicated that repetition marks with no number above them meant there could be an indefinite amount of repetitions. Sometimes the notations are incomplete, because he finished the transcriptions after returning home and the phonograph had not recorded the lowest and highest sounds at all. Nevertheless, the modern musician will find a mere skeleton when trying to play this music just from the music notation. A shadow of the original music can be found from copies of the phonographic recordings  kept in the Archives of the Finnish Literature Society in Helsinki. The meditative improvisations could last for hours. The player was playing his own feelings, movements of his soul, his own power, and ther is no way to bind it on a paper.

Read more About the kantele instruments

Sources:

Campbell, Murray & Greated Clive 1987. The Musician’s Guide to Acoustics. Schirmer Books. New York.

Karjalainen, Matti – Backman, Juha – Pölkki, Jyrki 1993. Analysis, Modeling, and Real-time Sound Synthesis of the Kantele, A Traditional Finnish String Instrument. Proc. Int. Conf. on Acoustics, Audio and Signal Processing (ICAASP 1993), vol. 1 Minneapolis, USA.

Kastinen, Arja 2000. Erään 15-kielisen kanteleen akustisesta tutkimuksesta. Sibelius-Akatemian kansanmusiikin osaston julkaisuja 5. Helsinki.

Kočkurkina, S. I.,1995. Muinaiskarjalan kaivaukset. Snellman-Instituutti. Arkistojulkaisu 3. Kuopio 1995.

Kuusi, Matti 1994. Questions of Kalevala Metre. What exactly did Kalevala Language signify to its users? Songs Beyond the Kalevala. Transformations of Oral Poetry. Studia Fennica Folkloristica 2. Ed. by Anna-Leena Siikala and Sinikka Vakimo. Finnish Literature Society. Tammer-Paino Oy, Tampere.

Kuusi, Matti 1977. Finnish Folk Poetry – Epic. An Anthology in Finnish and English. Ed. and translated by Matti Kuusi, Keith Bosley, Michael Branch. Finnish Literature Society. Helsinki.

Laitinen, Heikki 2010. Runolaulajien kantele. Kantele.  Ed. by Risto Blomster. SKS. Hämeenlinna 2010.

Siikala, Anna-Leena (ed.) & Vakimo, Sinikka (ed). 1994. Songs Beyond the Kalevala. Transformations of Oral Poetry. Studia Fennica Folkloristica 2. Finnish Literature Society. Tammer-Paino Oy, Tampere.

Peekna, Andres & Rossing, Thomas D, 2010. Psalteries and Zithers. The Science of String Instruments, ed. by Thomas D. Rossing. Springer. New York.

Pekkilä, Erkki (ed.) 1990. Hiljainen haltioituminen. A.O. Väisäsen tutkielmia kansanmusiikista.  SKS. Pieksämäki.

Povetkin, Vladimir 1992. Musical Finds from Novgorod. The Archaeology of Novgorod, Russia. Ed. Mark A. Brisbane. Lincoln, UK: Society for Medieval Archaeology.

Povetkin, Vladimir 2007. Musical Instruments. The Archaeology of Medieval Novgorod. Wood Use in Medieval Novgorod. Ed. by Mark Brisbane and Jon Hather. Oxbow Books, Oxford, UK.

Pulkkinen, Asko (ed.) 1946. Suomalaisia kansantanhuja. WSOY. Porvoo.

Relander, Oskar 1917.  O. A. Hainari. Muistelmia. Helsinki.

Uino, Pirjo 1997.  Ancient Karelia. Archaeological studies. Suomen Muinaismuistoyhdistyksen aikakauskirja 104. Helsinki.

Väisänen, A. O. 1928. Kantele- ja jouhikkosävelmiä. Suomen kansan sävelmiä. Viides jakso. SKS. Helsinki.

Väisänen, A. O. 1931. Kantele ja hyppivä puuhevonen. Kalevalaseuran vuosikirja 11. WSOY. Helsinki.