Northern Renaissance Instruments
6 Needham Avenue, Chorlton-cum-Hardy, Manchester M21 8AA, U.K., Yv
Phone & Fax. +44 (0) 161 881 8134 ; proprietor: Dr. Ephraim Segerman [USA]
e-mail: firstname.lastname@example.org ; on internet: http://www.nrinst.co.uk
TUNING AND STRINGING MEDIEVAL FIDDLES
The medieval fiddle was probably the most popular stringed instrument in Europe from when it started to be bowed in the 11th century to the middle of the 15th century, when growth in popularity of the lute overtook it. The loss of dominance then triggered many innovations that competed with the original fiddles. One increased the height of the bridge so that when the top was curved, one could play each of the strings individually (without the bow touching any other string), and thus it could take a part in playing polyphonic music (as the lute was then doing). Another innovation was to extend the flat fingerboard to cover much of the body, using it as a soundboard on which a low flat bridge could be placed (to play in the old style), while at the end of the fingerboard there was a step down to a curved second soundboard on which could be placed a high curved-top bridge (to play in the new style).
Another multipurpose instrument (called a 'vihuela'), invented then in Spain, was a combined lute and old-style fiddle. It had 11 strings in 5-course lute tuning, tied to a standard low lute bridge glued to the flat soundboard. Its body shape was that of an oval fiddle, but with a cut-out on each side at the waist, and it was constructed of thin pieces of wood glued together, like a lute was. The bodies and necks of previous fiddles were carved and scooped from a single solid piece of wood. Around 1500, two new families of instruments for playing part music were developed in Italy from the vihuela: The viols (with 5 or 6 strings tuned in 4ths and a 3rd) included the original size as a treble, double its size as a bass and intermediate sizes as an alto and tenor, and the Renaissance fiddles (with 3 or 4 strings tuned in 5ths) with the original size as bass, half the original size as treble and intermediate sizes as an alto and tenor. After the Renaissance fiddles appeared, they mostly replaced previous fiddles (two examples of late surviving medieval fiddles were found in Henry VIII's ship the Mary Rose). The rebec continued as a soloistic fiddle.
So when we discuss the medieval fiddle, we need to distinguish between the low-bridged instruments used from the 11th to early in the 16th century and the various modified versions used in the second half of the 15th century. And with the earlier type, we should distinguish between those with bodies not much wider than their bridges, such as the rebecs, where the end strings could be bowed individually and more incisively, and which had a more shrill sound, and the others. We will here be concerned only with the others.
Modern approaches to the medieval fiddle
Most modern musicians expect that the medieval fiddle was a single-line melodic instrument like a violin. A 5-string fiddle is often chosen because that number of strings appears most in the pictures, but 4-string fiddles were (and are) also popular. The tuning that appears obvious for a 5-string fiddle is that of a combined violin and viola, c,g,d',a',e". If the fiddle has only 4 strings, the viola choice (with the highest string omitted) is usually made. The low c is considered particularly useful since the
tenor part in the attractive polyphonic music, that often moves very slowly compared to the others in the music (and so appears to have a more instrumental character), can go that low. This tuning is easily achieved using more modern strings, but is very unlikely when using the low- and high-twist gut strings available in medieval times. Some other modern musicians opt for a tuning in alternating 4ths and 5ths such as d,g,d',g',d" or d,a,d',a',d" on a 5-string fiddle, with either the highest or lowest string omitted on a 4-string one. These tunings are possible with the strings available then, but the evidence on medieval tunings does not include them. We have made many sets of strings for customers using these popular modern tunings.
Historical reconstructions of military, craft or musical events are common and important in the tourism and entertainment industries. The audience wants to believe that what is offered is historically accurate, and would only be disappointed if presented with obvious anachronisms or blatant incompetence in the performance. In these circumstances, performers need to know little more about historical details than the average audience member, so delving deeply into the historical evidence is not necessary. If questioned on the historical accuracy of details of what they are doing, the standard response is that what is being offered is as accurate as can be managed practically, but whatever the details, much care is taken to get the spirit historically right. At best, that spirit is an interpretation of what the historical scholars have concluded, while at worst, it is an enactment of the popular stereotypes of what it was.
The problem with the performance of medieval music in general, and on the medieval fiddle in particular, is that the musicologists have been considering that these are mysteries, with no clear guidelines. They have been very involved with the early-music movement, and have been promoting modern performance of the music they study. Because of their overwhelming expectation that medieval music should sound attractively to early-music performers and audiences, they have been ignoring first-class historical scholarship which indicates that some aspects of original performance practices that do not conform with their expectations. Amongst these is the multiple-string bowing style of playing medieval fiddles implied by the evidence presented in the book The Origins of Bowing by Bachmann (1969). Early-music performers have mostly done whatever they felt was most comfortable and attractive, not taking seriously any contrary evidence.
Around 1300, Jerome of Moravia described three tunings for 5-string fiddles. Transposed to near modern pitch, they were 1: d’/g,g’,d”,d”, 2: a,d,d’,a’,d” and 3: a/a,e’,d”,d”, with a slash (/) after a 'bourdon' (Jerome did not mention that the first a in the third tuning was a bourdon, but this was added by his contemporary Pierre de Limoges, who owned the only complete surviving copy of the treatise). A 'bourdon' was a string that went normally over the ‘bass’ end of the bridge, but on the way to the side of the peg holder (where it went through a hole on its way to its peg), it went off to the side of the neck. The pictures often show the protruding thumb of the fingering hand between the bourdon and the neck. Options for what the thumb appears to have been doing were to ignore the bourdon, stop it in mid-air to get a higher pitch, dampen its sound or pluck it. Many fiddles, like in Jerome's 2nd 5-string tuning, most 4-string and almost all 3-string fiddles, didn’t have a bourdon (so all the strings went over the fingerboard).
A characteristic of all of Jerome's tunings is that some of the strings came in unison pairs (tunings 1 and 3) or octave pairs (tunings 1 and 2) or both. In 1497, at the other end of the medieval period, Tinctoris wrote that the 5 strings were tuned 'unevenly in fifths and unisons'. He wrote that this fiddle had a rounded bridge so that strings could be bowed individually. Even with this modern feature, his evidence agrees with that of Jerome on the point that the modern expectation of an even and 'logical' tuning has no historical support. Medieval fiddle tunings had their own 'logic' that is different from ours.
We do not have evidence indicating what that logic was, so it is appropriate to explore the possibilities. Reasons for having a unison pair of highest strings could have been: (a) the sound enhancement of an added string produced the balance that was preferred, (b) such strings are most likely to break, so a performance needs not to be interrupted if one broke, (c) the slight difference in tuning when both are fingered together was considered attractive, or (d) a choice between fingering the two strings together or separately was musically useful. Reasons for the lowest string having an octave companion could have been (a) the lowest string is the lowest in projection which is enhanced when playing the octave string with it, (b) there was octave ambiguity so that the listener would hear the pitch that was expected when both sounded, or (c) a choice between fingering the two strings together or separately was musically useful.
Jerome wrote that tuning 1 was for 'all the modes', while tuning 2 was for music of particularly wide range, mostly secular. He didn't mention what tuning 3 was for, but it appears to be one in which the tune rarely left the top pair of strings, and since this tuning had the smallest total range, it had the richest-sounding lowest strings. He gave no tunings for 3-string and 4-string fiddles.
Tinctoris gave a tuning for the 3-string fiddle, which was simply a tuning in two fifths. The only historical evidence on the tuning of 4-string fiddles is a the occasional 15th century picture that shows a string thickness sequence of thickest, thinner, somewhat thicker and thinnest, seeming to imply two octave pairs a fifth apart. Other likely possibilities for these instruments would involve dropping strings from Jerome's 5-string tunings.
If one wanted to explore the 5-string medieval fiddle in an historically informed way, it would seem appropriate to start with Jerome's tuning 2. The pictures show that the body was usually much wider than the strings where it was bowed, and the bridge height was considerably less than its width. This severely limits the angle range over the strings of the bow. So whether the bridge top is curved or flat, one would expect to almost always play more than one string at a time. There should be enough bowing angle range to choose whether or not to play the bourdon with the other strings. So that tuning can be considered to be that of a two-string fiddle (treating pairs, either in unison or octaves, as single strings) with the option of including (or not) the bourdon as a third drone string. The hurdy gurdy was invented as a medieval fiddle that had mechanical devices for bowing and fingering, and its original character of playing melody with drones has remained. Though this appears to be the basic style of playing the medieval fiddle, the fingering seen in many pictures is fairly complex, which implies that sometimes the pitches of the drones were changed by fingering.
There is some medieval evidence indicating that the number of hairs in a medieval bow was usually considerably less than half that of a modern bow. The resulting lower bow tension suggests that the string tension was also lower than on modern bowed instruments. Since the amount of sound output of a bowed instrument depends largely on string tension and bridge height (which was also lower), and is enhanced with the use of a soundpost and bass bar (which were Renaissance inventions), we should expect the loudness of medieval fiddles to be considerably less than of modern bowed instruments. The pictures show that bowing was usually in the second fifth (20-40%) of the distance from bridge to nut. Such bowing positions give a much less incisive sound than modern bowing positions.
On a bowed instrument, when a melody moves from string to string, for smoothness, we would expect the tension not to change very much from string to string. It is likely that unison pairs had the same tension. Guidance on the relative tensions of octave pairs can be gleaned from what musicians have found satisfactory on related instruments which have these octave pairs, which are the hurdy gurdies with octave chanterelles and the Welsh crwth. In these cases, the low octave string has a tension of about 60% of the tension of the high-octave string. On the hurdy gurdy, where the drones are played continuously, they have about 40% of the tension of the (high octave) melody strings. On the crwth, where the drones are played only when required, they have about 20% more tension than the melody strings. These relative tensions are probably rather arbitrary, depending on the relative resonances due to instrument design as well as individual taste concerning attractive balance.
Speculative string specifications
From the above considerations, it is possible to suggest speculative sets of strings for the different tunings of medieval fiddles mentioned. We assume that a typical or 'medium' set has the tension of the highest string (when stretched) of about 5 Kg at a string stop (vibrating length) of 36 cm and a pitch standard of a' = 440 Hz. For the Jerome 1 tuning, we assume equal tension for all strings except that the low octave g has 60% of the tension of the others. The bourdon d' is given full tension since it needs strength when plucked. For the Jerome 2 tuning, we make the same tension assumptions. For the Jerome 3 tuning, we assume equal tension for all strings. For the four-string tuning with two octave pairs, we assume that the low-octave strings are similarly of lower tension.
The resulting specifications are shown on the Table. The string types assumed here are those available at the time. If the thickest strings sound too dull for modern ears, and that is important for how the instrument is used, replacing the thicker high-twist ones with roped-gut (catline) strings (which cost more) will help considerably.
That Table also includes two popular modern stringings for medieval fiddles. The string tensions vary in the modern violin tradition, with the tension decreasing with lower pitch to provide quicker response in the bass and to suppress the boom of thicker metal-wound bass strings. The violin plus viola tuning assumes that the string length is 32 cm. The fourths and fifths tuning assumes a string length of 36 cm. It is possible to have all-gut stringing for these tunings by substituting catline strings (which are more expensive) for the metal-wound ones.
If the string stop is somewhat greater than that assumed here for the chosen tuning, the same sets can be used, but all of the string pitches need to be lowered (by a semitone for each 6% it is greater) to avoid the highest string breaking more often than is usual for gut-strung instruments. If the violin plus viola tuning is used with a longer string stop, longer strings will be needed, which cost more.