The Orsa style in perspective

It is customary to speak of "dialects" in instrumental Swedish traditional music. This means that there are local stylistic features which are just as different as for the spoken dialects. Among the traditional music dialects the Orsa style holds a somewhat special position and is one of the most cherished dialects. It turns out that very many fiddlers outside Orsa and Dalarna specialise in Orsa tunes. But how much do we really know about the Orsa style?

To understand what makes the Orsa style special we must view it in a wider context.

click for a closer view

One thing which is striking when we take a closer look is that today's Orsa style is nearly exclusively based on the playing of one fiddler, Gössa Anders. This is hardly surprising considering his talent as a fiddler, his large repertoire, his reliability, his authority, etc. However, to be able to understand a local style one has to delve deeper. A dialect should not be an individual playing style, but rather a combination of many players' interpretation of it. When listening to older players we see that the differences between individuals in the same village or area are often greater than the difference today between dialects. At least this is the case when comparing areas with roughly the same proportions of various polska types.

This record has come about in order to shed light on the music of Orsa in this way, and to give it a wider range of expression. I recommend that all those who want to play Orsa tunes study and compare the styles of these fiddlers in detail: rhythm, tonality, ornamentation, tempo, agogics (rhythmical shifts), bowing, tone - simply everything which together defines a style.

In 1955, an interview was made with Olof Andersson, one of the men behind the seminal and monumental collection Svenska låtar (SvL). Unfortunately the sound quality of the recording is very bad, probably because the tape has lost nearly all treble. But what can be heard is interesting enough, as the interview mostly concerns rhythm in tunes from Dalarna. Olof Andersson begins his account with Ore, then continues west to Orsa and Älvdalen, and ends with Lima and Malung in western Dalarna. During the conversation he describes similarities and differences in this large area, which he seemingly perceives as one large dialectal area.

In Ore, with Timas Hans, he mostly mentions that quaver pairs are syncopated, which is also found in other parts of Dalarna and also in other regions. When he gets to Orsa he quotes himself from SvL I, page 54:

At certain points, at points where he [Gössa Anders] played, and also Blecko Anders, his teacher in Orsa... there suddenly comes out a series of powerful and heavy strokes of the bow, which lend particular emphasis to some certain place in a part, and thereby so to speak want to overthrow the barriers formed by tempo, bars and rhythm, which has a strange and captivating effect. An example of this is for instance polska 104b" [as played by Gössa Anders. In SvL I no.95 as played by Blecko Anders is used as an example.]

N.B. SvL I no. 104a and b are the same tune as track 4 on this record and track 24 on the previous record in this series (Fiddle Tunes from Orsa I, Hurv KRCD-24), where Gössa Anders plays together with his daughter Anna. How Gössa Anders plays differs considerably from the transcriptions.

One of the interviewers comments that in the polska SvL I no. 111 as played by Gössa Anders, which is also an example of the above, there is a problem in the transcription in the third and ninth bars of the second part. Here Nils Andersson has written two quavers on the first beat and a crotchet triplet on the second and third. Olof Andersson answered:

Well, it was played by Gössa Anders when Nils Andersson visited him the first time, maybe in 1907 [---]. Then when I visited Gössa Anders and checked all the tunes, I realised that they sometimes changed their way of playing considerably [---]. Here [---] I noted that Nils Andersson had transcribed these emphasised places which broke the pattern like this: [plays a few bars]. That's how he played for Nils Andersson, but this is how he played [for me]: [plays again]. So this triplet came at the beginning of the bar instead of at the end. [---] I want to add that you could say that these rhythmical peculiarities only occur in Dalarna. I haven't found much of them, well some from Uppland, something similar...[---] But these rhythmical shifts they sort of increase, so that they are more and more prominent the further west you go. Now we're only in Orsa, but now we'll go on to the music of Älvdalen."

N.B. SvL I no.111 is track 12 on this record and track 8 on Fiddle Tunes from Orsa I. When Olof Andersson talks about this rhythmical tendencies increasing as you go westwards he means within Dalarna. If you go down to Värmland these rhythmical tencencies decrease. This can be because the playing is more and more ornamented by triplets. Even in the dialects from Dalarna which we are concerned with here triplet passages are played with a straight rhythm, even in polskas which otherwise have a shifted rhythm.

Olof Andersson explains that the irregularities he noted in the music of Ore and Orsa are also found in Älvdalen "and nearly even more". At least this is true for the older players, according to O.A. "It is surprising at first", he says, "but soon you find that it's a style". In the rest of the interview it comes out that Nils Andersson has been greatly assisted in understanding the style of Dalarna by the fiddler Lars Åhs from Älvdalen, who could write music. Nils Andersson has also been influenced by him in his way of transcribing tunes from Dalarna. O.A. takes the polska no.420 (as played by Lars Åhs) in SvL II as an example of the same rhythm found in Orsa playing.

N.B. For those who do not have access to SvL, the crucial point here is two bars consisting of a crotchet plus two quavers plus a crotchet. During the tune it is noted that these bars were played so that all the notes in the bar were of roughly the same length, so what you hear is a crotchet quadruplet.

The polska no. 428 as played by Lars Åhs is also used as an example of this quadruplet rhythm (four notes of equal length played over three beats, i.e. a whole bar of a polska).

Then Olof Andersson continues with the music of western Dalarna:

In Lima and in Transtrand the rhythm was sometimes even more indefinite, especially among older players. They could prove to be a hard task for the transcriber.

Olof Andersson uses SvL III no. 732 as played by Valla Erik Samuelsson from Lima as an example. The interview continues:

[---] this possibility to shorten the first crotchet of a bar of a polska... yes, that's typical of the style of western Dalarna. In this way it was halved to a quaver, immediately followed by the other two beats. In this way you only get 5/8 time. Einar Övergaard, born in Elverum on the Norwegian side, half Norwegian and half Swedish, who transcribed folk tunes himself in these parts, obviously had severe problems getting the rhythm right [in the transcriptions]. He started his collection trips in western Dalarna already in 1896 [should be 1897], which is a few years before Nils Andersson. In his manuscripts which are kept in the dialect archive in Uppsala, he usually transcribes polskas as a quaver followed by two crotchets and in the index he uses 2?/4 time. He divides the polskas into two categories: the usual simple type he calls polska, and transcribes them in 3/4 time, and the other, which is more like Norwegian music, he calls 'springare' [springlek], and transcribes them in 2?/4 time [---]. We have transcribed a number of such springare from our Swedish folk musicians.

N.B. What he wants to demonstrate with SvL III no.732 is just what he earlier described as characteristic for the Orsa style: "there suddenly comes out a series of powerful and heavy strokes of the bow, which lend particular emphasis to some certain place in a part, and thereby so to speak want to overthrow the barriers formed by tempo, bars and rhythm". In the transcriptions he has tried to describe this using crotchet triplets, sometimes at the beginning of the bar, sometimes at the end.

Olof Andersson goes on to Malung and plays SvL III no.823 as played by Troskari Erik Persson, one of the fiddlers whom Einar Övergaard also visited and collected tunes from. However, he gets into a bit of trouble, since he happens to choose a springlek with many quaver triplets, which means a tune where a straight rhythm is introduced into an otherwise syncopated one. He changes his mind and plays a tune which is unknown to me. Finally Olof Andersson discusses Einar Övergaard's and Katarinus Elling's transcriptions from Elverum in Norway, where there are rhythmical shifts similar to those found in polskas from Dalarna.

Conclusion. What O.A. discusses is a type of polskas with what is today usually called "short first beat". These polskas are found in Dalarna from Orsa to western Dalarna, more and more as you go further west. As the "short first beats" increase it can be noted that the semiquaver ornamentation decreases. I find that this is something which is not only found in Dalarna. For instance, I have heard some older nyckelharpa players from Uppland playing with distinct "short first beats" in some of their quaver polskas. However, there are polskas where this rhythmical shift can not be used at all, especially those with a semiquaver subdivision. This means that areas where semiquaver polskas are predominant have less of this rhythmical shift, but there it can also be found in certain quaver polskas. Something else which should be noted is that the shifted rhythm does not seem to be as consequent as Övergaard's transcriptions might suggest, not even in western Dalarna. Often the "short first beats" are played in every second bar, in this case even bars: two, four, six and eight. This rhythmical phrasing corresponds well to how bakmes polska is danced. However, this rule has numerous exceptions, just consider how common it is to find parts with an odd number of bars. Gössa Anders is a good example of a player who often uses a shifted rhythm in every second bar, or a pattern where the stress on the first or second beat changes every second bar. Track 2 here is an example of this. In the tune on track 3 he uses another rhythmical pattern: two bars with stressed and long first beat, followed by two bars with a stressed second beat where the first beat is shortened ad lib. Polskas such as these go very well with the bakmes dance. The polskas in tracks 10 and 11 are obviously of a slightly different kind. Gössa Anders introduces a few of these phrasings here too, but the overall rhythm is more even or straight, and it can also be noted that this kind of polska can be played at a slightly higher tempo. These polskas are similar to what is called "slängpolska" in many parts of central Sweden (not to be confused with the slängpolska of eastern and southern Sweden) and the dance is a precursor of the hambo. The polska dance collected as "Orsa polska" corresponds quite well to "slängpolska".

The traditional music regions mentioned here are the larger musical area of which the Orsa dialect is a part. Probably it should be enlarged a bit more, mostly northwards and a bit into Norway, but it is hard to draw exact boundaries. Primarily Särna in the northernmost part of Dalarna should be included, and probably parts of western Härjedalen. Typical of this musical area is not only rhythmic peculiarities, but also certain characteristics of the melodies, especially an element of "pastoral tune" modes. Even if there are polskas with "short first beat" further south, these tunes have a different tonality than those from Dalarna. So why do we speak of "dialects" in traditional music? This concept was not invented by the players themselves, but rather by scholars studying the music. To me the difference between old and modern playing is greater than between the various "dialects". Gössa Anders from Orsa and Spak Olof from Särna have more in common than they have with the modern playing style, and it is virtually impossible to find a detail in the playing of an old player and show that it is unique to his "dialect". We can only go by tendencies. Certain traits are more frequent in one area than in another, and in this way we can find where an unknown player or tune is from. But often it seems like we have learnt to recognise some of the influential players in traditional music. Hjort Anders, for instance, has an unmistakable style which we have learnt to recognise as "Bingsjö style", but a listener in the past might who heard him might have thought he was from somewhere in Hälsingland. What makes Hjort Anders' style come from Dalarna is the element of "pastoral tune" tonality, not the rhythm.