The Nyckelharpa

The most thorough documentation so far is Jan Ling's doctoral thesis Nyckelharpan, 1967 & 1979, (ISBN 91-518-1272) which also has a long summary in English.

The nyckelharpa can be traced back to the late Middle Ages. The earliest known depiction of the instrument is on the doorway of the church of Källunge in Gotland, which is dated to around 1350. During the 15th century nyckelharpas are depicted in church paintings, and it is interesting that the majority of these are from Uppland, which thus already then seems to have been the real home of the nyckelharpa. During the 16th century we find traces of it being spread from northern Sweden down to Denmark and Germany. During the 17th and 18th centuries we find a few instruments in Finland and Norway, but it disappears in the countries south of Sweden. Later the area is reduced again to Uppland, but on the way there the nyckelharpa was also played in other parts of Sweden. Even if we cannot prove that it was invented in Sweden, we can safely claim that the nyckelharpa is a genuine Swedish instrument, one of our two major contributions to the history of musical instruments - the other one being the large Swedish clavichords from the 18th and beginning of the 19th centuries (documented on Hurv KRCD-21).

Like the hurdy-gurdy, the nyckelharpa is a stringed instruments with a keyboard, and the note is produced by scraping the strings. The wheel of the hurdy-gurdy is really like an endless bow, and the idea behind this is that the drone should never stop when you have to change between up and down bows - like a bagpipe imitated by a stringed instrument. The step from the hurdy-gurdy to the nyckelharpa is short, but it is really quite surprising that the nyckelharpa - which originally was really like a fiddle with keys - was not invented before the hurdy-gurdy, which can be traced back to the 12th century. As the research stands now it seems as if the nyckelharpa got its keys from the hurdy-gurdy, while the shape of the body can be traced back to various other stringed instruments.

However, there is a Catalan miniature in a bible manuscript from around 1100, which is not mentioned in any litterature about the nyckelharpa, which depicts a fiddle player who holds his left hand over the instrument as if playing a hurdy-gurdy. This way of holding it requires a key mechanism, but it is doubtful whether it can be seen in the picture - there is something sticking out from the lower part of the neck, which could be interpreted as a key, but it could just as well be his thumb. This must be considered uncertain evidence, since the artist could have had limited knowledge. Otherwise this would be the oldest known depiction of a nyckelharpa - and the timing is right considering the history of the hurdy-gurdy. (The miniature can be seen as no. 24 in the picture appendix of Bachmann, Die Anfänge des Streichinstrumentenspiels, Leipzig, 1966.)

The simplest principle of a nyckelharpa is a row of keys (sing. nyckel, plur. nycklar or sing. knaver, plur. knavrar in Swedish), with tangents (sing. and plur. löv) set perpendicularly to the keys, which shorten the melody string, plus a number of drone strings. The distances between the points of contact between the melody string and the tangents determine the scale or scales. The scales on old nyckelharpas vary between different instruments, and all types of course are quite different from modern tempered tuning (piano tuning). This simple type is today called enkelharpa ('simple nyckelharpa'). One problem is that there is no such simple basic nyckelharpa preserved in its original state, because already at an early stage experiments began with more than one tangent on certain keys to produce fixed double stops. Such a connection is called a mixtur. The reason for this was that for certain notes of the scale one wanted chords which harmonised better with the melody than the drones did. Furthermore things are complicated by the fact that we are not sure that all "mixtures" were actually used on these instruments. They could have made experiments and mounted tangents to connect the strings, and then removed or turned away some or all of them to simplify the instrument again.

A modern term for these various experiments at making nyckelharpas with connected tangents is mixturharpa. This word was never used by the players themselves, and neither was the term "enkelharpa". Only when looking back at history and wanting to differentiate various instruments do we need these categories. The players themselves said "harpa", "nyckelharpa", "Schlüsselfiedel", "knaverharpa", "nyckelgiga", "nøglegiga", "nyckelfela", "nøglefeile" or something similar.

The earliest nyckelharpas did not have any sympathetic strings (resonance strings). The first preserved nyckelharpas with sympathetic strings are from 1777-1830, but there were probably earlier instruments which have been lost. The idea of equipping a stringed instruments with thin metal resonance strings is not Swedish, and hardly even European. It is not entirely clear where for example the viola family began to be equipped with this novelty, which was mentioned already in 1618-19 in Syntagma Musicum Vol.2, Organographia by Michael Praetorius, but one theory is that the idea came to Europe through trade contacts with India and the Muslim world, where there was already this kind of stringed instrument. These connections were opened wide when Queen Elizabeth the First of England granted rights to The East India Company in 1600. We can presume that not only silk and spices were transported on the ships, but also the odd musical instrument, and maybe even a "native" who could play it for a curious audience. Praetorius writes the following about a lyre viol (a special kind of viola da gamba which was played using various tunings): Now in England they have invented something new and strange, that under the usual 6 strings another 8 strings made of steel and twisted brass lie on a brass bridge (as the one used on a pandora), and these must be exactly tuned to the upper ones. If one of the upper gut strings is touched by a finger or a bow, the lower brass or steel strings resonate per consensum, and tremble and quaver so that hereby the sweetness of the harmony is increased and enlarged.

The most well-known and widespread bowed instrument with sympathetic strings in Europe is of course the viola d'amore, but there was also a violino d'amore, both in Sweden and on the continent. The Norwegian harding fiddle should probably be viewed in this context: a popular version of a "violino d'amore", which has acquired a distinctive national character comparable to the Swedish nyckelharpa, which has gone through a parallel development into a "nyckelharpa d'amore" through the addition of sympathetic strings. The nyckelharpa's cousin, the hurdy-gurdy, also had sympathetic strings fitted at around the same time. The use of sympathetic strings is interesting not only from a technical point of view, but also indicates that the folk musicians of old times followed the development down on the continent and kept up to date with it.

Towards the end of the 18th century the type called kontrabasharpa had developed. This word probably comes from the terminology of the players themselves, but we do not know when it began to be used. At least it was already in use in 1899, in Leffler's book Om nyckelharpospelet på Skansen ('About the Nyckelharpa playing in Skansen'), where it is called "the older type" and said to be "extinct", which was only partly true. Leffler includes what we would call "enkelharpa" today in his description of the instrument. According to him it only has three playing strings, tuned to a1, g (or a) and d1. It "usually" (today we would say "always") has one row of keys, for the outermost string, which is tuned highest. He also writes that on "some" nyckelharpas these keys have extra tangents further in, which touch the lowest string (the one closest to the player). As an example he takes a nyckelharpa from Roslagen, where keys number 1, 2, 3, 4, 5 and 7 have these extra tangents. Here we have a nyckelharpa with two melody strings surrounding a drone string, which is what we today define as a "kontrabasharpa". One characteristic of the kontrabasharpas is that there are good chromatic possibilities, and they can be played in a number of different major and minor keys through simple retunings. Later the kontrabasharpa was sometimes fitted with two drone strings and there could also be "mixtures" for one of these. On kontrabasharpas with two drone strings these were tuned to d1 and g, and later often to c1 and g. The latter tuning is maybe an adaptation to the more and more dominating key of C major in nyckelharpa tunes. It could also be an adaptation to the silverbasharpa, when this came, so rather than getting a new instrument, which required learning a slightly different way of playing, you modified the instrument you already had. Increasing the number of drones to two might seem to give the instrument more possibilities, but is really a voluntary restriction, since this makes it less suited for playing in different keys and the chromatic possibilities are not as useful.

The name "kontrabasharpa" has nothing to do with the double bass of an orchestra, and it is not a bass nyckelharpa, as one could be led to think. The name comes from the fact that it has a melody string on the other side of the "key box" (lek): contra the first melody string.

The enkelharpa is built on the principle of one row of keys and one melody string (with variations). The kontrabasharpa introduces two independent melody strings, however there is still only one row of keys. Since the drone (or drones) is placed between the melody strings we get the characteristic playing style and sound of the kontrabasharpa with a continuously sounding drone.

Many of the developments the nyckelharpa has gone through are determined by how the drone has been viewed in various musical periods, and also the reigning taste. On the one extreme we find pure drone instruments, which can hardly be played without a constant droning bass - and on the other extreme we find the modern chromatic nyckelharpa, which has completely rid itself of the drone, and thereby become something different than all the other nyckelharpas from over five hundred years. Somewhere in between these extremes the nyckelharpa called silverbasharpa can be found.

The term silverbasharpa also comes from the players' own terminology, and is also mentioned by Leffler in 1899, where it is called the "new type" and is said to be "nearly 100 years old", which is maybe a bit exaggerated. Here we also have a term which is not very precise. The name really only tells us that this is a nyckelharpa where some bass string is partly made from silver. Looking back on the history of musical instruments we find that metal wound strings for bowed instruments appear at the end of the 17th century, but unwound metal strings were at times used much earlier (e.g. for the violin) - and if they were used on one bowed instrument, it is likely that they were used on another one. The French led the development of metal wound strings. In Italy they were more "conservative" and kept using plain gut violin strings until the end of the 18th century, while the Germans during this time were somewhere in between, with plain gut strings and an open wound g string. Already at the beginning of the 18th century the French violins normally used plain gut strings for the e and a strings, an open wound d string and a close wound g string - which by the way is the normal string set used for the Norwegian harding fiddle to this day, except that the e string has been substituted with a metal one. The metals used for winding the strings were copper, silver or silver-plated copper.

The introduction of sympathetic strings shows us that the nyckelharpa players and other musicians in Sweden had their eyes on the continent in order to know what was going on and what was in vogue, and thus we can imagine that they were also aware of the new-fangled idea to wind strings with thin metal threads. The way to this could be through the higher classes and the contact between their musical culture and that of the "people" at meeting places such as factories in the countryside - the large number of employees there naturally resulted in a small "entertainment industry", where the most important ingredients of course were music and dancing, which just as today had to be according to the fashion of the time, since the dance culture was primarily supported by people in the prime of their youth. Lövsta bruk in Uppland is a well documented meeting place of this kind, where the most skilled nyckelharpa players were invited to play at the finer parties, while the less skilled players would provide the music for other occasions.

In their essay Nyckelharpans typologi, Gunnar Ahlbäck and Gunnar Fredelius write that in written records before 1967 the word "silverbasharpa" does not necessarily mean the instrument which Ling and the new typology use it for. Before that it could also mean "kontrabasharpas with double key rows", "mixturharpas", and even three-row chromatic nyckelharpas. An "enkelharpa" or a "mixturharpa" with a silver wound bass string could be called a "silverbasharpa" in this older terminology. Now it is not sure that the nyckelharpa players always used strings wound with pure silver thread when they started trying out metal wound bass strings. For reasons of economy or tone they might just as well have had strings wound with pure copper or silver-plated copper, and the winding could be either open or close. Some silverbasharpas maybe never even came close to any silver, while others maybe were "silver bass harpas" one day and "copper bass harpas" (if such a term existed) the next.

A modern definition of a silverbasharpa, which is rather accepted today, is a nyckelharpa with two melody strings and two drones. The biggest novelty was that the second melody string got its own row of keys with their own tangents. However a few "mixtures" were retained in order to make the instrument fit better for the most common key of C major, for example a connection of b1-d1 for the dominant of this key, the G major chord, which of course is also useful when playing in G major. This made it possible to play melodies on the second row, although to a limited extent, since the keys of the second row were not too many. Equally important was the fact that the second melody string and its row of keys could be used to embellish the melody with double stops as on a fiddle, without being confined by the fixed double stops. As we have seen, earlier instruments also had the possibility to play other harmonies than those formed by the melody and the open strings, but these were decided by the mixture tangents which always produced the same note when pressing a key. The only way to avoid these chords was to remove or turn away some tangents before playing. On the silverbasharpa, on the other hand, you can use these notes as you like, or not use them, in which case the second melody string becomes an extra drone. If you play the melody on the second melody string you get a drone above it, apart from the bass drones, which can be used ad lib.

To begin with there were up to 8 keys in the second row, perhaps a relic of the chromatic kontrabasharpas, but eventually it was noted that some of these were hardly ever used, and the number was reduced to those actually needed, typically 4-6. The first row of keys can be up to 21 in number. The melody strings were tuned a1 and c1, while the drones were tuned g and c. The silverbasharpa was so much adapted to C major that it is sometimes called a "C major harp". Apart from C major it was also possible to play tunes in F and G major, and also (more seldom), in a minor, the parallel of C major. As you got further away from C major you naturally had to be careful how you used the drones, or else you got a rather strange - or maybe interesting - sound in your tunes, something you hear now and then, not least on this recording of Sigurd Holmberg's playing. There is no record of retunings (scordatura) being used to adapt to other keys, but it is not farfetched to believe that it might have happened. There simply might be no records just because nobody asked questions about it while the old players were still alive, those who could tell how things were in days gone by, and could retell what they heard about even older players. At least some of the earliest silverbasharpas could very well be retuned in the same way as a kontrabasharpa.

Since the melody strings of the silverbasharpa are next to each other it was now possible to play melodies without a constant drone. From this there developed a characteristic way of playing, "dip bass", where the drones not only have a harmonic function, but also a rhythmical. This is not necessarily something completely new, since strictly speaking the old enkelharpas invite such a way of playing, which is slightly reminiscent of fiddles, tuned to the key, where the melody in older tunes often only uses the two upper strings, whereas the two lower ones are often used as drones. The silver wound lower drone is also very well suited to this style of playing by having more of a "ring" to it than an unwound gut string.

In one way the silverbasharpa represents a development, but at the same time it has certain limitations, which players of the kontrabasharpa often point out. The limitations are that it has less chromatic possibilities and is harder to play in different keys. At the same time, these limitations are there on purpose, and have been turned into different possibilities. Because the silverbasharpa is highly stylised, it is a strikingly elegant instrument, where every detail tells us of the intelligence and artistic sense of its originaters, whoever they were. The musician Jonas Skoglund and his contemporaries said that the first silverbasharpa was built by an organ builder for a sergeant Söderstedt. According to Ling's research, this organ builder was Pehr Olof Gullbergson, and the year was 1838. It could very well be so, but it seems more probable that the instrument was the result of a longer process, which could include metal wound bass strings on earlier types of the instrument, and a gradual trend towards modern tonality. The time might be correct, and it leads on to other questions. A new dance and music fashion started to spread at roughly the same time as the industrially manufactured accordions were introduced and gained ground, which seems to be more than a coincidence. At the beginning of the 19th century the waltz came in vogue and also more modern polska dances. Towards the end of that century other dances were introduced, such as the polka and schottische, and they have an even closer link to the melodeon, which is no wonder, since one-row melodeons were played at every other farm. A theory has been proposed that the dominant key of C for the nyckelharpa repertoire is due to the fact that they were often played together with C clarinets. This could be part of the truth, but it seems far more probable that the silverbasharpa is adapted to the repertoire of the one-row melodeons.

Something halfway between silverbasharpa and a kontrabasharpa is the so called kontrabasharpa med dubbellek, which is sometimes called österbyharpa, since it had its main-eng centre in Österby. It was probably invented so that players who were used to and liked the tone and possibilities of the kontrabasharpa wanted to keep up with the development and play the new repertoire without learning a new way of playing. It still has the low melody string on the other side of the lek (key box) and drone(s) in the middle. In addition it has an extra melody string beside the highest melody string, as on a silverbasharpa. It is played as an old-fashioned kontrabasharpa with a constant drone, which the design makes unavoidable. What is typical is however that you can use the melody string of the second row for ornaments in the form of double stops as on a silverbasharpa, and it has mixtures corresponding to those on a silverbasharpa.

Of old the melody string was made from treated silk, which was also common on fiddles and harding fiddles. On the kontrabasharpa the drones were mostly made from gut, but metal wound strings were surely used now and then. Around 1900 the silverbasharpa mostly had gut strings for the first three strings and the lowest drone string was silk wound with silver or silver-played copper. Naturally a high melody string made of silk could be used for the early silverbasharpas. Some time around the beginning of the 20th century some players started using a steel first melody string, perhaps influenced by the fiddlers. In classical music the metal e string on the violin made its breakthrough around the time of the First World War, even if experiments were made with both e and a strings made of metal at the end of the 19th century, but some say that this was an area where the folk musicians were at the leading edge of the development, perhaps because they needed more hardy materials which would keep their tuning better at dances when the place would be warm and damp. Many nyckelharpa and fiddle players, however, kept the old gut strings for tone reasons up until the middle of the 20th century. Gradually, while they finally switched to industrially produced strings, it became more and more common that all the strings were metal, or in the case of the thicker ones, metal wound gut. Today's "steel strings" with a metal wound metal core hardly existed before the 1950s.

One thing which should be remembered when it comes to tuning older stringed instruments is that the tuning standard has changed over the years - a1 = 440 Hz is a rather new standard, and is not even today the only one used for orchestras and pianos. In the old days there were even bigger differences in tuning, both local and national. Furthermore, folk musicians were less restricted by these standard tunings than orchestral musicians - they would tune to the pitch where the instrument sounded best and performed best - i.e. where the pressure on the instrument was just right, so high that there was a good margin to the point where the strings would break, while still getting as much volume and brilliance as possible. Gut strings can be tuned higher than silk and thus we can conclude that they tuned lower when using a silk melody string. As seen before the early silverbasharpas still had reminiscences from the kontrabasharpas. There are plenty of records of tuning to G even well into the 20th century , i.e. a fourth lower than the usual tuning today, and more typical of a kontrabasharpa.

Around the turn of the century 1900 there were both low pitch and high pitch tunings. For example clarinets from that time are often tuned to a high pitch, i.e. about a semitone higher than the "standard pitch" of that time, which could be for example a1 = 432 Hz. If a fiddle or a nyckelharpa was going to play together with such a C clarinet the tuning obviously had to be just as high. If it was a Bb clarinet it could be handy to tune down, and then the fiddle or nyckelharpa was turned into a Bb instrument. The folk musicians, who would always play by ear and were often "musically illiterate", hardly thought of it like this, but in this way they could play the tunes without transposing and with fingerings which were easy for both, which was the main thing. The clarinet players often used very old-fashioned instruments which were more like 18th century clarinets than modern ones. They could be boxwood clarinets with very few keys or ebony clarinets with the old German system - the Boehm clarinet, which made it possible to play easier in all keys, but also had a comparatively thin sound, was out of the picture. These old-time clarinets were all but equalised (evened out so that all the notes would have a similar timbre) and the scale steps could differ slightly. A scale could have the perfect combination of a natural third, pure fifths, etc. which made it well suited for playing together with a bowed instrument. However, if the tune was transposed and played starting on a different note, it might not sound too good, maybe even rather out of tune, apart from requiring a fingering which the fingers were not used to. For these reasons it was handier to retune the stringed instrument than to transpose on the wind instrument. Even the old melodeons could be either low or high pitch, which explains the fact that many of them seem to have very strange fundamental notes. When playing these together with a fiddle or a nyckelharpa the same basic rules applied as with the clarinet, except that they were even more restricted to one key in the case of one-row melodeons. Another situation where you had to adapt your pitch was when playing together with singing, and had to adapt to the pitch of the voice.

So when saying that "C major" was the predominant key on the silverbasharpa, this is relatively speaking, and really applies to the fingering key, not the absolute pitch. When playing with a melodeon tuned to G and high pitch (slightly lower than a modern A flat) you would tune down, and in this case it would not be a "C major harp" but an "A flat harp", which of course is just as nonexistent as a "kopparbasharpa". When reading Sigurd Holmberg's biography we see for example that he had two silverbasharpas, one tuned to G and one to D, which seem to be adaptations to two of the most common fiddle keys in Swedish folk music. Since typical A tunes can often easily be played in G too, Holmberg and Härdelin had a good bunch of tunes they could play together. The fact that he uses C tuning on our recordings might be because he played solo and that the nyckelharpa sounded best at that pitch, because the strings used were probably optimised to that key. So everything indicates that the silverbasharpas in the old days used to be tuned to different pitches depending on the situation.

The bows used for the older types of nyckelharpa were adapted to playing with a drone, with their curved form and loose hair. The hair was tightened by the thumb, either by pressing from below up towards the stick, or by placing the thumb between the hair and the stick so that the hair was tightened backwards towards the nut. The ring and little fingers or only the little finger would most often lie under or behind the stick, which resembles an old way of holding a fiddle bow called the "French way", where the thumb was also under the hair or the nut, and the little finger behind the stick. For the modern chromatic nyckelharpa this old bow and bowing style have been substituted by a bow more like a short violin bow, adapted to playing only one string at a time.

The silverbasharpa is the last of the "old nyckelharpas" and the last step of a development which went on for five hundred years or more. The chromatic nyckelharpa brings us into a new phase, and it is really just as much a new instrument as it is a development of the older nyckelharpas. The timbre, playing style, repertoire and design often makes it seem like a fiddle with keys - with the same possibility to play in all keys, melody playing on a single string, double stops as far as the fingers can reach, but at the same time without some of the most important means of expression of the violin: yet again it is confirmed that you cannot win anything without sacrificing something else. Personally, I find the older nyckelharpas - which do not try to be anything else than a nyckelharpa - much more fascinating instruments, with their combination of sophisticated technology, elegant design, the clear ring of the sympathetic strings and melodies in the brightest keys on the one hand, and on the other hand a primitive and almost coarse, medieval drone sound. This is the dualism which seems irresistible to me, and the more I listen to this happy and captivating music and watch the ingenious and beautiful instruments it is played on, the more enchanted I become.

To conclude it all, a figure from the past appears in my mind. He was an old musician, around 70, however he did not play the nyckelharpa, but the hurdy-gurdy. I came across him when he was entertaining people as a busker on one of the promenades on the right bank in Paris. He was not from Paris but from Picardie, but had come to the city to make a bit of extra money. He was dressed up with a strong element of bright blue, not fine cloth for a suit, but rather for working-clothes, but better cut, probably sewn by his wife, a bright silk scarf and of course a beret, all in all as French as he could be, but not the least artificial. (Behind him another person steps forward to my mind's eye: Sigurd Holmberg in a similar situation, dressed in the hat which all Swedish gentlemen wore at the time, when he had to go up to Rättvik for the summer to make his living with his nyckelharpa.) When the Frenchman had played a few tunes for his small audience, consisting of my girlfriend and me, he started a little lecture about his instrument which he was fiercely proud of. Then he ended the lecture with these words, which our old nyckelharpa players could also have agreed with: MY INSTRUMENT IS THE MOST FANTASTIC IN THE WORLD, BECAUSE IT IS NOT ONLY ONE INSTRUMENT - IT IS A WHOLE ORCHESTRA!

Anders Rosén,

September 2000

Some sources: