THE DEVELOPMENT OF EUROPEAN BOWED INSTRUMENTS
up to the baroque: a closer look
[draft as of February 2002]
Ensembles of viols in Italy
The original viol, the 15th century Spanish vihuela, was a multipurpose instrument with a low flat bridge that was either plucked like a lute or bowed like a flat-bridged medieval fiddle (when it was held vertically with the neck upwards). It introduced the waist cut out in the body outline, which subsequently became standard in most Western bowed instruments. The waist cut allowed the bowing style of a narrow fiddle to be used on a wider instrument. In that bowing style, the end strings (those nearest to the soundboard edges) can be played melodically without any other strings sounding, and when bowed at a shallower angle, other strings can sound as well. With the 9-string 5-course lute tuning and the high octave of the 5th course at the end, the two end strings would be tuned a fifth apart, just like in the Moorish rebab (the rebab's influence is also apparent in the vertical playing position). With its long neck that has 10 frets, the vihuela's melodic range covered by the end strings was about one and a half octaves.
During the 15th century, a tradition grew of musicians playing on the same kind of stringed instrument working in pairs. It seems to have been led by lutenists playing popular polyphonic songs, with one (the tenorista) playing mostly the tenor and contratenor parts, and the other playing an improvised highly decorated version of the descant part. In 1487, Tinctoris was very impressed by a performance of two Flemings playing bowed violas, one taking the treble and the other the tenor1. It is not clear whether they were instruments tuned mostly in fifths (fiddles) or fourths (viols). In 1493, there was a report of a pair of Spaniards from Rome playing vihuelas (viols) almost as large as people2. These viols (i.e. with a string stop of about a metre) seem to have been double the size of the original ones, and played an octave lower. By 1502 Alfonso dEste played in a group of 6 viols, presumably including these two sizes plus intermediate ones (he ordered 5 viole da archo in 1499)3. In the next few decades, there was an explosion of the playing of vocal part music on viols and other instruments that then developed sets of different sizes, greatly helped by the newly available vocal part music printed by Petrucci.
When the Spaniards developed vihuelas that were specialised for bowing around 1500, they reduced the number of strings from 9 in five courses to 8 in four courses4, to 6 in three courses5, or to 6 single strings6. The Italians included these modifications when they adopted the viol7, and appear to have been the leaders in further modifications of the original Spanish viol design. They made the instrument louder by raising the height of the bridge, now always movable. This also allowed the player always to choose to bow either near the bridge or farther from it (the original vihuela design allowed melodic bowing only far from the bridge, at the waist cut-out). With the waist cut-out no more being essential, alternative outlines without it were explored around 1500, largely in Ferrara8.
A 1505 painting9 shows a viol with a bridge with shallow curvature parked under the strings behind a higher bridge (being used at the time) that had rather more curvature. This indicates that players exchanged bridges to play in different styles: With a higher bridge having more curvature, individual strings could be sounded separately when bowing near the bridge, and groups of strings could be sounded while bowed farther from the bridge. With a lower flatter bridge, like that of the lira da braccio, groups of strings would be sounded while bowed near the bridge, and more strings would be sounded when bowed farther from the bridge.
It is likely that the first viols had flat soundboards and cross bars glued to their undersides on each side of the soundholes, as on the other descendants of the 15th century vihuela, the 16th century vihuela and the guitar. Soon, it seems that an alternative arched construction of the soundboard without cross bars appeared, which was successful with the fiddle descendants of the original vihuela. By the 1540's a new construction combining arching and cross bars became fashionable. This construction included an added bar perpendicular to the cross bars that passed under the bass foot of the bridge. The bass bar corrected the imbalance of deficient output from the top string. Having cross bars allowed heavier string tensions (giving louder instruments) without encountering uncontrollable resonances. Later in the century, the fashion reverted to arched construction with no bars. This gives a sweeter sound, and it seems that the makers found a variety of design changes (such as soundboard thickening down the centre or in other places) that could inhibit those unwanted resonances at higher string tensions.
As with the medieval pictures of fiddles, most of the early 16th century pictures of viols show bowing several strings at a time, either by low bridge curvature or by the large distance of the bowing position from the bridge. This indicates that there was much continuation of the musical style of the original vihuela, with the tune (now always on top) usually accompanied by other notes in the harmony beneath. When playing part music, this requires knowing how the harmony is progressing. Such enhancement with harmony is more demanding than just playing the notes, and suggests that just playing the notes was confined then to players less advanced in musical accomplishment.
By Ganassis time (1542)10, single-note viol playing had advanced both in possible complexity and status, and he wrote a method for it. Nevertheless, he mentioned that some viol players imitated the lira da braccio by using a particularly long bow with slack hair tension, or by using a viol with less curvature on the bridge and fingerboard. The latter type of viol (with names sopran, falsetto, contralltto, tenor, basseto and bason) persisted throughout the 16th century and into the 17th, and usually went under the name lirone11. Ganassi mentioned lirone as one of the names for the viol (viole and violone were others), but he didnt specifically associate it with.the type of viol that, like the lira, always played more than one note at a time. Modern organologists have confused the lyra da gamba, a very different instrument, with the lirone.
With a curved bridge, the melody (with or without additional harmony notes beneath) could be played on strings other than the first, so to cover the melodic range, the high positions on the 10-fret neck of the original vihuela were no more needed. This was exploited by viol design changes in Italy that appear around 1510. The size of the body was increased and the neck correspondingly shortened (accommodating only 7 or 8 frets) to maintain the original playing characteristics. The motivation for this seems to have been to lower the pitches of the instrument's resonances by increasing the area of the soundboard and the volume of air enclosed in the body. This idea is supported by a further increase in the body depth (with an upper-bout back fold to bring the body depth at the joint with the neck back to what it would otherwise have been). Another modification that increased the volume of enclosed air was to put the neck block outside the main body by having the curve of the sides turn towards the neck, going around the neck block and flowing into the sides of the neck. Previously the sides approached the neck perpendicularly with the neck block inside, as with most fiddles.
The viol tunings given by all of the sources from Lanfranco (1533) to Marinati (1587)12, were basso: D,G,c,e,a,d, contralto & tenore: A,d,g,b,c,a and soprano: d,g,c,e,a,d, with the middle sizes possibly a tone lower and the basso possibly a tone higher, depending on difficulties in assembling a set of appropriate sizes to sound well together. According to Alfonso della Viola (c.1536) and Ganassi (1542), if the music was in flat keys, players would assume tunings a tone lower than normal for their sets, presumably to make the music and its improvised decoration fall more familiarly under the fingers.
Tunings given in subsequent sources fall into two groups. Ceretto (1601)13 and Mersennes (1636) Italian informant14, gave the same tunings as earlier, while others, Zacconi (1592)15, Banchieri (1609)16 and Cerone (1613)17, gave tunings all a fifth lower. Since the relative tunings of the strings of each viol were the same, we can represent all of the string pitches of each viol by the lowest and highest string pitches. Thus we can call the early and higher later set as the D-d,A-a,d-d set, and the lower later set as the GG-g,D-d,G-g set. Banchieri also mentioned a fourth viol, the violone in contrabasso tuned DD-d, so his extended set was DD-d,GG-g,D-d,G-g.
We want to know the sizes of these instruments, and we can relate string stops to the nominal tunings if we know the type of bass gut strings used and the pitch level (which could be a recognised standard or not). We can see how these relate for gut strings from Praetorius (1619)18, who provided string stops, nominal pitches and an absolute pitch standard. From his data we know that the highest working pitch frequency for gut (measured in Hz) multiplied by the string stop (measured in metres) was about 210 (the same for plucked and bowed instruments)19. This represents the judgement of musicians then as to how high their highest strings could be tuned without making the rate at which they broke unacceptably high.
For any string stop, the lowest acceptable pitch depends on how much inharmonicity there is in the sound it produces. Inharmonicity makes the higher harmonics in the strings sound out of tune with the fundamental and limits the number of harmonics heard. The limit of inharmonicity is when the sound is considered too dull and unfocussed to be musically useful. This limit can be determined historically from the lowest pitch on an instrument with maximum open-string range used at the time, and for which the sound of that lowest string needed to be full. In families of instruments, where the string tension tends to be proportional to string stop, this limit can be extrapolated to other string stops by using the physical theory of string inharmonicity20. That maximum range for viols was two octaves and a fourth on Praetoriuss viola bastarda, which had a string stop of 73 cm. When this inharmonicity limit is applied to other sizes, that range decreases with shorter string stops, and increases with longer string stops.
This all refers to a lowest string of the roped-gut type. With the two-octave open-string range of 6-string viols, this gives the leeway of a range of string stops for a given pitch, or a range of pitches for the string stop. Earlier than the last quarter of the 16th century, when these bass strings were not generally available at sensible cost, that range was much less, with almost no leeway.
With the above quantitative information, assuming that the lowest strings used did not have roped-gut construction, we can estimate that the original D-d,A-a,d-d set, with the basso having a string stop of about 100 cm and the soprano about 50 cm, had a pitch level of about a fourth below modern, a minor third below the usual Italian pitch standard corista. If viols wanted to play with other instruments, they either had to transpose up a minor third or tune up that amount while removing the first string, leaving only five strings (see below). Late in the 16th century the lowest strings could have been of roped gut, and the increased range would allow either a lower pitch level or a shorter string stop. We find that both occurred, with both the lower pitched and smaller sets conforming to the corista pitch standard. With roped-gut basses and corista pitch, we can calculate that for the smaller D-d,A-a,d-d set, the string stop of the basso could be from 67 to 82 cm, the middle sizes from 46 to 55 cm and the soprano from 37 to 41 cm. For the larger GG-g,D-d,G-g set, we can calculate that the string stop of the basso could be from 88 to 123 cm, the middle sizes from 67 to 82 cm and the soprano from 51 to 62 cm.
From the above, we can see that the GG-g,D-d,G-g set had essentially the same string stops as the original D-d,A-a,d-d set, and a good number of players apparently used the newly affordable and generally available roped-gut strings to be able to play with other instruments from the same music at the usual pitch standard but with the new low tunings. The other players kept the original D-d,A-a,d-d tunings but reduced the sizes of their viols by about 20%. They didn't use the original basso of about 100 cm size (except as a contrabasso addition to the set), used the original middle sizes of about 70-80 cm size as its basso, kept the original soprano of about 45-55 cm size as its tenore, and added a new viol of about 40 cm size as its soprano.
Banchieri called his new DD-d viol violone in contrabasso, his GG-g viol violone da gamba, his D-d viol viola mezzano da gamba, and his G-g viol quarta viola in soprano. This illustrates a change in terminology from the early use of violone as any viol. The new use of the term meant either a viol with lower nominal tuning than any in the original D-d,A-a,d-d set, or the original largest viol size abandoned in the new small set, with the DD-d viol an even larger version of it. The calculated string stop range for the violone in contrabasso is 111-164 cm. The evidence from Praetorius and the few surviving instruments is that it typically was about 125 cm. Both of these types of violones were extensively used as the basses of 17th and 18th century string ensembles.
When a contrabasso violone was specified in 17th century Italian baroque music, there was no ambiguity about a DD-d tuning, but there was ambiguity in the intended tuning if the specification was contrabasso without violone, since the contrabass of the small set of viols was the GG-g viol, while that of the large set was the DD-d viol. That ambiguity could have been deliberate if either would do. The range of the written music is not evidence of an intended choice since bass-instrument players then readily transposed phrases by an octave either for musical function or expression, or because it lay better on their instruments21. These instruments could be played high in their ranges as well as low, and Monteverdis specification of viola contrabasso for an alto part in Altri canti dAmor in Combattimento illustrates how this could be effective.
All the above refers to 6-string viols. Ganassi wrote (1542) that most players played on 5-string viols with D-a,G-d,d-a tunings, with the third between the third and fourth strings. The viols Virgiliano22 (c.1600) illustrated all had five strings. The calculated string stop ranges are 80-110, 64-82 and 46-55 cm respectively. The calculation assumes the normal pitch standard, and the resulting ranges include the usual large viol sizes. Being able to freely play with other types of instruments could have been a major motivation for having only five strings before roped-gut bass strings were generally used. It is quite possible that smaller 6-string viols that would be easier to play, and could tune to corista, were made for children (especially choirboys) to play on in this period.
Ensemble viols in Germany
In 16th century Germany, the term geige meant bowed instrument, modified (as in Italian) according to size, with the large ones (gross geigen) being viols (tuned in fourths with usually one third), and the small ones (clein geigen) being fiddles (usually but not always tuned in fifths). The example of gross geigen illustrated by Virdung (1511)23 shows an instrument very similar to the original Spanish vihuela in that it had a low flat glued lute-like bridge, nine strings (probably in five courses), and a body waist cut-out to allow favouring the end strings during bowing. It differed from the original vihuela in that the waist cut-out was greatly elongated (so that it allowed favouring the end strings when bowing at a range of distances from the bridge of from a sixth to a half of the bridge-to-nut distance), and a relatively shorter neck that allowed only seven frets. This design seems to have been more respectable than co-existing Italianate ones, that had a free higher bridge having some curvature, and single strings. A 5-string Italianate example is in the title-page border decoration on some copies of the original edition.
Tuning information on viol sets was given in the Munich 718 ms (1523/4)24. There were four sizes with the two middle ones having the same tunings. All had five strings tuned in fourths with a third between the 3rd and 4th string, as in Italian examples. The assumed tunings were that of an D-a,G-d,d-a set if the treble part didnt go too high, and a G-d,c-g,g-d set if it did.
Agricolas (1528)25 tuning information was on several sets of geigen tuned in fourths with one third. Their tunings allowed maximum transfer of previous training in reading music on a lute with a g top string. One was a G-g,c-g,f-c set of gross geigen with the bass having six strings and the others five. The third in all tunings was between f and a. There were no illustrations of this set. Then two sets of 4-string instruments were mentioned, one called gross and the other called cleine geigen. The tunings of both were given as G-a,c-d,g-a 4-string sets, with the third between f and a in the bass and middle sizes and f and a in the treble. It is likely that the tunings of the cleine version were an octave higher, with sizes like fiddles. A set of 4-string geigen was illustrated, and the designs were similar to that of Virdung with elongated waists and low glued bridges. With only four strings, it is possible to tie the strings to a glued bridge to have some effective curvature so the player can choose to play chords with either the 1st or 2nd string on top, allowing an octave of melody range with chords beneath. This style is implied by Judenkunigs statement (c.1518)26 that the geygen was very similar to the lira.
Gerle (1532)27 gave assumed 5-string tunings identical to those of the Munich 718 ms, which were of a D-a,G-d,d-a set if the treble part didnt go too high, as well as the tunings of a G-d,c-g,g-d set if it did. For the latter set of assumed tunings, Gerle suggested that an added 6th string tuned a tone below the 5th on the bass would be useful to play an F in the music. Such a bass tuning is inconvenient for fingering chords, and it is likely that individual notes were usually bowed on this viol. His illustration of a viol does not have an elongated waist or a glued bridge. The bridge is missing, possibly just because it was normal to remove the bridge when the viol was not being played, and possibly because bridges with different amounts of curvature could be inserted, depending on style of playing.
Some illustrations from around 153528 show viols with holes in the soundboard through which the treble foot of the bridge goes through, acting like a soundpost. This would inhibit the sound of the highest strings and enhance the sound of the lowest strings, unless there was a longitudinal bar under the centre of the soundboard, which could result in some distrubuted enhancement (by the soundboard rocking around the bar), as on the modern Greek or Cretan lyra.
Agricolas second edition (1545)29 includes the grossen Welschen geigen, where the Welschen means foreign and probably Italian. It was a F-b,c-d,g-a set with the bass having five strings and the others four, with thirds between the end strings and the next ones in the bass, and between the 3rd and 4th strings in the others. Alternative treble tunings mentioned were either for all four strings to be tuned a fourth higher or for a fifth d string added. This would require a smaller instrument. An alternative bass tuning was F,G,c,e,a. The drawing of four 4-string viols is reproduced from the first edition. The drawing should have some relevance, so it seems that it applies to the 4-string sizes, which would then have been played chordally. The frontispiece drawing of Fraw Musica playing the lute shows other instruments, including a 4-string viol with a more modern movable bridge that allowed a choice between playing in a chordal style (like the Italian lirone) and playing single strings.
The fingerboard charts of Agricolas grossen Welschen geigen indicate notes for tuning with the lute and the cittern. Since playing with other types of instruments was indicated, a pitch standard was likely to have been followed. The prevailing pitch standard was the chorthon of southern German lands (where all of the viol activity mentioned above occurred), which differed little from the Italian corista standard. The same relationship between pitches and sizes should then apply. The string-stop ranges for Agricolas 1545 F-b,c-d,g-a set would be 70-98, 51-82 and 37-55 cm. If the D-a,G-d,d-a set of Munich 718 and Gerle conformed to the pitch standard, their string-stop ranges would be 84-110, 64-82 and 46-55 cm for the three tunings. Similarly, the string-stop ranges of Agricolas (1528) 5- and 6-string G-g,c-g,f-c set would be 63, 51-62 and 40-46 cm, and of his 4-string G-a ,c-d,g-a set would be 64-110, 51-82 and 37-55 cm. In this period, there was apparently more variability in the sizes of German viols than in the Italian ones, and a tendency to be somewhat smaller.
We have no tuning information on German viols in the second half of the century, but inventories testify to their continued popularity. Some of the pictures show quite large 4-string bowed instruments playing in ensembles with winds and treble fiddles. To modern eyes, they look like viols. There had been no German 4-string bass viols mentioned since 1528 (Agricola), so it is probable that the intervals between strings were fifths, and these were bass fiddles, equivalent to the French basse de violon. The modern association of right-angled waist corners and C holes with viols and sharper waist corners and f holes with fiddles is a poor indicator of historical identity. It is likely that the Germans just called them by the conveniently ambiguous term grosse geigen.
By Praetoriuss time (1619), the acceptable lower limit to the open-string range had expanded with roped-gut strings, so 6-string tunings conforming to the pitch standard became easy if the standard was the usual one in France, England, Italy and south Germany, where the vast majority of viols were made. At that standard, the Italian small D-d,A-a,d-d set at about 80, 55 and 40 cm string stops, and the large GG-g ,D-d,G-g set at about 100, 80 and 60 cm string stops, were all near as high in their ranges as they could be. But the standard in north Germany, where Praetorius wrote, was a tone higher. Those instruments could not tune up to these pitches at this standard. The solution shown by Praetorius was to use the small set of viols and tune them at the bottoms of their ranges, getting as close to the tuning of the large set as possible. The result was a GG-g, D-d,A-a set.
Tunings labelled 3 and 4 for each size in his table of tunings were just reports of Agricolas 1528 tunings, and those labelled 5 were reports of Ganassis 3-string tunings in fifths. Tunings 1 and 2 were for the his sets. The bass of his set was called klein bass, and beside the basic GG-g tuning (tuning 2), three alternatives were given for his tuning 1. One alternative was a GG-g viol with the 4th string down a semitone (to E), another (for a smaller bass) was a AA-a viol with the 6th string up a semitone (to BBb), and the third (for a larger bass) was a FF-g viol all in fourths. The set of viols Praetorius illustrated had double purfling, a characteristic of English viols. At that time, English viols players and repertoire were considered to be leading Europe. The bass, with a string stop of 75 cm, was more than a fret-length shorter than the optimum of 81 cm (it could have been made as a solo bass or the second bass of a set), and was probably tuned to the alternative of an A-a viol with the sixth string at BBb. The string stop of the tenor was a bit small at 58 cm, and that of the treble was 40 cm.
Praetorius also mentioned a gross bass or violone. The tunings given were a 5-string EE-c viol all in fourths, a normal 6-string DD-d viol and a 6-string EE-f viol all in fourths. The example illustrated had a string stop of 103 cm, which fits either of the 6-string tunings. Finally, a 5-string gar gross bass or gross contrabass geige was listed and illustrated. Its tuning was DD,EE,AA,D,G with a string stop of 128 cm.
In 1628, D. Hitzler30 reported the bowed instruments used in ensembles in his area of southern Germany. Included were 6-string bass geigen with standard C-c and D-d viol tunings, and one of five strings with tuning C,E,A,d,g, which could be the Italian 5-string bass fiddle depicted by Praetorius, but restrung to be tuned like a viol.
Later in the same region, the A.S. ms (mid-17th century)31 reported two different sets, a preferred GG-g,D-d,f-c set with a 5-string treble and the others having six strings, and an alternative AA-a,G-g,d-a set, again with only the treble with five strings. The G-g viol of the alternative set was identified as a solo viol, and it probably played a decorated version of both tenor and alto parts. It could also have a 7th string tuned to c, which would also include a treble part. This instrument developed into the viola damore.
Later in the 17th century, Prinner32 mentioned a new 5-string violone tuning of FF,AA,D,F#,B (as well as the usual 6-string GG-g one). With the top string down to A, this 5-string tuning was mentioned in the Talbot ms (c.1694), and remained particularly popular around Vienna till well into the 19th century (it was the type of violone usually used when Mozart and Haydn was played there). A new 4-string version of the violone was also developed33. It was tuned an octave below the highest four strings of the ordinary one, with the 4th string either a note lower at EE (resulting in the tuning that later became standard on the double bass), or a note higher at GG.
In the 18th century, bass viols had adopted a wound 6th string and became smaller, and were losing ground to the cello as the small bass of string ensembles. Early in the century, 6-string violones still predominated as the large basses. The usual GG-g tuning was used on surviving original-size bass viols with a wound 6th and on full-size violones with all-gut stringing. Some large violones (probably originally contrabass types) had a low CC string (probably metal-wound). The use of reduced-range violones with 4 and 5 strings grew steadily, and these replaced the 6-string violones by the second half of the century. The term violone (or violon in German) was gradually replaced by contrabasso during the century.
Late in the 18th century, a 3-string violone (called contrabasso by then) became an alternative to the 4-string variety. It didnt become prominent until the 19th century, when the two leading virtuosos (Dragonetti and Bottesini) used it. Tunings usually were GG D A (called French, an octave below the top three strings of the cello) or AA D G (called Italian). It used the ordinary size instrument and octave transposition to play in a most constricted range. One advantage was to be able to dig into notes without touching other strings while bowing not necessarily close to the bridge. Another advantage was of not needing any metal-wound string, so it consistently provided the fundamental-rich foundation note for an orchestra that a thick all-gut string provides. It was common for orchestras to have both 3-string and 4-string instruments. The music of late romantic composers demanded lower notes, and that led to abandonment of 3-string double basses in the 20th century34.
Ensembles of viols in France
There is a record of a viol player in France in the first decade of the 16th century35, and we know that viols were popular enough for Gervais to publish a method for it around 1547 (now lost) which used tablature notation36. There are two 16th century surviving sources giving the tunings of French viols. Jambe de Fers (1556)37 set of 5-string viols had tunings E-c,B-g,e-c, all in fourths, and Mareschalls (1589)38 set was the same except that the dessus (treble) was a tone higher (f#-d).
At the expected ton de chapelle pitch standard, the calculated acceptable ranges of string stops of Jambe de Fers set are 74-94, 54-63 and 43-47 cm, and of Mareschalls treble it is 39-42 cm. Mersenne (1636) and Rousseau (1687)39 wrote that these early viols were larger than the viols of their own times. The string stops of the sets these later writers knew were about 80, 60 and 40 cm (with an alto between the tenor and treble). The difference could be just be that the string stops of the early sets were at the maxima of their ranges, but it is likely that the early French viols were bigger (like the Italian viols) and played at a lower pitch level than the usual French standard. There is no evidence of viols playing with other instruments then, so there is no reason to expect conformance to a standard.
Jambe de Fers illustration of a viol (missing in the sole surviving copy but reprinted in Mersennes book) showed a bridge rounded enough to play individual strings when bowed near it. Yet a c.1585 drawing of a 5-string French viol40 shows quite a flat bridge, implying that a chords-plus-melody style of playing was also used That particular viol also had a lower-back fold as well as the usual one in the upper back, a feature found on a good number of surviving English treble viols (often 'restored' away), presumably to make playing against the shoulder more comfortable.
Mersenne credited Jacques Maduit (1557-1627) with the addition of a 6th string to French viols41. The tunings of 17th century French viol sets as reported by Mersenne and Rousseau were basse: D-d, taille (tenor): G-g, hautecontre (alto): c-c and dessus: d-d. The calculated ranges of string stops at the expected ton de chapelle standard for these tunings are 65-84, 51-63, 41-47 and 37-42 cm. But Mersenne stated that the English tuned their viols a tone lower than the French to 'render the harmony softer and more charming', when from the nominal pitches and pitch standards, we wound expect little difference. This echos Praetorius's statement that when playing alone the English viols played at the same pitches as his viols, which would be a minor third lower than they normally tuned. If the two observations were of the same practice, this would imply that Mersenne's viols were tuned to a standard a semitone lower than ton de chapelle. This standard was called 'opera' or 'theatrical' pitch42 later in the century.
The earliest evidence for the use of a soundpost on viols is given by Mersenne. He called it the soul of the instrument, and wrote that it was directly under the treble foot of the bridge, and can be lifted up again through the sound hole when it has fallen down. The soundpost was not mentioned in his discussion of fiddles, which might indicate that in a fiddle it was still permanently fixed when the instrument was opened up, as was the bass bar in both types of instrument. It is likely that the soundpost spread amongst viols after about 1600, as it did with fiddles of all sizes, when its combination with the bass bar led to the great success of the violin in Italy.
Ensembles of viols in England
In the 16th century, there is the same kind of English evidence of viol playing as in the rest of Europe, though there is very little on viol characteristics and none on tunings. We can expect a strong Italian influence after 1540 when viol players from Italy joined Henry VIIIs musical establishment and dominated royal bowed-string playing for the rest of the century. Soon, playing viols had become part of the education of choirboys, and in the middle of the century, professional choirboy ensembles of viols were commonly playing as entertainment for functions in London43.
It is likely that the choirboys used smaller viols than the usual Italian sizes both because their fingers were smaller than adults, and because they could play at the easy transposition of a fourth lower than the 10 ft FF organ pitch that they sang in. With these smaller sizes, there was no reason for having less than six strings on each. These viol sizes were the only ones used in public viol performances, and these performances helped spread the popularity of viol playing outside the households of the nobility. This was probably crucial in establishing the smaller sizes as the norm in England, making sense of Rousseaus statement that the English reduced the sizes of their viols earlier than the French did.
The 17th century sources that gave the tunings of the English set of viols indicated that they were D-d,G-g,d-d (with no alto44) at what Mace called 'Consort pitch' (the same as south German chorthon and Italian corista, and about the same as French ton de chapelle). The calculated possible string-stop ranges for these pitches are 67-82, 53-62 and 38-41 cm. Talbots (c.1694) measurements lead to string stops of the consort bass and tenor viols of 81 and 61 cm respectively. We can expect that the string stop of the treble was about 40 cm. Talbot didnt measure a treble viol possibly because he couldnt find one. The playing of viols in sets had faded by then, and many treble viols had been converted to tenor violins (violas)45. His only mention of a tenor viol tuning was A-a, which was when he was apparently copying the Italian tunings mentioned by Mersenne. This was the only English source that reported a tenor tuning other than G-g. Supporting the point that this mention was not relevant to English tenor viols, is that the string-stop range of a AA-a viol is 46-55 cm, which does not include Talbots string stop of 61 cm.
For the string stops of 81, 61 and 40 cm, the set could be tuned several semitones lower than the standard pitches. A common retuning for the bass was for the 6th string to be tuned a tone lower, to double C fa ut. According to Praetorius, the English tuned their viol sets lower than their usual pitches when playing alone (without voices, lute or organ) because they preferred the sound that produced. This was down to the pitches he tuned his own viols (of the same sizes) to. This was at 3 semitones lower then their nominal pitches, which were relative to the pitch standard a tone lower than his. Mersenne wrote that the English tuned their viols a tone lower than the French. Since the relevant pitch standards and viol sizes were essentially the same, and the nominal tunings were the same, either the English viols he heard were playing a tone lower than their usual pitches, or they played at the same pitches as Praetorius reported but the French viols played at a semitone lower than ton de chapelle.
During 18th century, interest in viols decreased. Often with re-necking, original small solo basses (see below) were used as cellos, original tenor viols were used as small cellos, and original treble viols were used as violas. Consort basses were replaced by smaller basses with a wound 6th string. The original ones could have been used as a double bass viol (the English name for the violone) with a wound 6th string (as the German ones were), but English interest in using contrabass instruments was low, and they disappeared. The old wood of discarded instruments has always been highly valued for repairs on currently-used instruments.
When sets of viols were revived, mainly by Arnold Dolmetsch at the 1890's, surviving small solo bass viols of cello size (with string stops under 70 cm) were what the new sizes were based on. They played well with D-d' tuning at modern pitch (or somewhat lower), and new tenors were developed at 3/4 of this bass's size, and trebles at half its size. Evidence for larger original sizes was disbelieved (and often still is). Surviving original tenor sizes were thought to have originally been small solo basses, and surviving original treble viols were thought to have originally been alto viols tuned a tone below the trebles (as 17th century French ones were tuned). Such alto viols work well in playing many parts in the music, and documentary evidence for their original use in England was sought, but not found.
Soloistic and special-purpose viols
A major use of a stringed instrument has always been for simple accompaniment of the players singing. In addition it could have often played instrumental interludes. Any viol (or lirone) of a set could be used this way, but a bass was most common. Italian examples were in the 1539 Florentine Intermedii46 and Ganassi47. A large fraction of the extensive lute song repertoire published in England early in the 17th century specified a 'viola da gamba' as an alternative (not in addition) to the lute in accompanying the voice. With only the need of balancing with the voice, no need to conform to any pitch standard, and with fingering easier on smaller viols, such a bass was usually considerably smaller than the optimal size of the bass of a set. Almost all of the surviving 17th century English bass viols are of the smaller sizes made for this purpose (they survived because they could be used as cellos after the demise of viol playing). In the last decade of the century, using a metal-wound 6th string became acceptable, so small basses became suitable for any bass-viol functions. Makers such as Barak Norman started to make new ones to be used this way. Such small sizes, with a wound 6th string, were also characteristic of 18th century bass viols elsewhere.
As mentioned above, a lirone could be physically distinguished from a viol in 16th century Italy only by the curvature of the bridge, and perhaps also of the fingerboard. The linguistic implication of the name lirone has led organologists to confuse it with the lira da gamba, otherwise known as lirone perfetto, lirone doppia, lira grande or arciviolata lira. The lira da gamba didnt play any individual voice consistently, being essentially a continuo instrument. It was generally of the size of a small bass viol. Evidence for the lira da gamba starts from the 1570's and continues for more than a century. By the 17th century, it was usually just called lira, because the lira da braccio was in such disuse that there was no ambiguity. It had about a dozen strings, mainly tuned in a repeating sequence (from lowest to highest) of a rising fifth and a falling fourth. The bridge had very shallow curvature, with five or six strings bowed at one time being usual. To establish its lira credentials, it usually had an individualistic body shape, a leaf-shaped peg disk and a few bourdons that could be plucked by the left thumb. The advantage it had over the lirone was that it could offer bowed chords in any key with equal facility. Disadvantages were that the fingering patterns were unique (with no transfer with other instruments), and continuing a melodic line within the chords to another string required skipping strings.
In the 1570's, when roped-gut strings became affordable and generally available, a specialist viol called viola bastarda appeared in Italy. The size was that of a small bass, and the bridge was curved to play single lines. Its purpose was to give solo interpretations of well known part songs. Sometimes it skipped amongst the parts played in a highly decorated way, and sometimes it did what the lute did, involving arpeggiated chords. It was a style of playing, in different (especially extended) tunings, and not a special instrument recognisable by any criterion other than size when seeing it. The extended tunings included fourths (a viol characteristic) and fifths (a fiddle characteristic) and the name could have come from considering it as an illegitimate offspring from a union of these two families of instruments.
Of the five viola bastarda tunings listed by Praetorius, four were used in English viol tablatures before 1610. A different one used was sometimes called 'bandora set' (when playing on any viol) or Leero way (on a bass viol). We also find that tuning used on a particularly small bass viol of about 65 cm string stop (it had to be that small because of the large stretches required in the tablature), called 'lyra viol'. After about 1610, new lyra viols were fitted with metal sympathetic strings (in unison with the bowed strings)48. Within a few decades, the sympathetic strings were largely discarded49, and the lyra viol continued as just a soloistic viol played in various tunings from tablature. Later in the century, fingering demands were more relaxed, and larger lyra viols were preferred, a string stop of about 72 cm, two frets shorter than a consort bass, being typical.
By about 1640, the English player Walter Rowe working in Germany, played a viol called barretone, which was a lyra viol with a modified neck so that the thumb of the stopping hand could pluck the metal strings running underneath the fingerboard. This instrument persisted (mentioned in the Talbot ms), and much was written for it by Haydn. Around 1650, the principle of adding metal sympathetic strings to a bowed gut-strung instrument was picked up by the Norwegian Hardanger fiddle, a local version of the violin.
In 1679, John Evelyn recorded hearing a German playing a viol a'amore, which was an ordinary violin but with 5 wire strings played Lyra way (i.e. with chords and a special tuning). Rousseau (1687) mentioned that the viole damour was a treble viol with brass strings. Daniel Speer (1687) wrote that the viola di lamor was a viol with iron or silver strings and special tunings. In early 18th century Germany, the name viola damore applied to a soloistic treble viol in a variety of tunings with two types of stringings. The first had only six strings, with the 1st gut, the next three of iron or brass, and the last two were metal-wound. The other had six or seven playing strings of gut with metal-wound basses, and about the same number of iron or brass sympathetic strings. If the number of sympathetic strings was double that of the bowed strings, the instrument was called 'English violette'. An Italian book by A. Sgargi (1747)50 describes an amateur's shoulder viol (viola da spalla) called viol d'amour or viola angelica with six or seven bowed strings of gut (with the basses copper-wound) and brass sympathetic strings tuned chromatically.
Around the middle of the 17th century, a new kind of soloistic viol appeared in England, called division viol. Its string stop was about 76 cm, one fret shorter than a consort bass. In his method for playing this instrument, Simpson51 indicated that he preferred the design that had the shape of a violin, with the Bellyes digged out of the Plank (by contrast, other English viols then had their bellies made of bent staves glued together). The division viol that Talbot measured (c. 1694) was of this type and the body dimensions were very close to those he gave for the bass violin. That was the usual bass violin in England (except for the royal music establishment, which used the French basse de violon), and apparently came from the Italian basso da braccio. It is likely that the division viol had an Italian origin, with the Italian model having a basso da braccio body on which was a longer and wider viol neck. The Talbot division viol seems to have been of this type, while the example Simpson illustrated had a longer body and shorter neck, giving the same string stop.
The division viol was tuned the same as the bass viol of a set, and its playing style was like that of the viola bastarda without most of the chords. The Italian instrument popular at the same time as the division viol, and with the same playing style, was the violoncino. The first we hear of the violoncino is around 1640, and it seems to have been a revival of the viola bastarda (which we last hear of in the 1630s), but without extended tunings and apparently with a basso da braccio body. The name means small violone, which was appropriate since it was tuned like a viol and was smaller than the violone, which was the only viol still popular at the time. When there was a small resurgence of interest in other viols in Italy later in the 17th century, violinistic exaggeratedly pointed corners were as likely in the design as the right-angle corners of English and French standard viol shapes.
In the middle of the 17th century, a school of soloist bass viol players developed in France. Rousseau credited one of them, Sainte Colombe, with introducing to the bass viol both a 7th string tuned a fourth lower (AA) than the sixth, and strings wound with silver. In the 18th century, we hear of violes voutées, which were viols with violin-type bodies (having arched backs). They could well have been the French equivalents of the division viol and the violoncino. Some of their ordinary viols then, with flat backs, also had the violinistic sharpened waist corners.
At various times, we find viols smaller than any that were normally used in sets. There is early 16th century evidence in paintings. In a Portuguese one, a 5-string viol is playing with a string drum and a tiny rebec52. In another, the famous painting St. Cecilia by Raphael, a very small broken 6-string viol lies at her feet53. In the latter case, there is reason to suspect that the relative sizes of the viol and Cecilia had been altered for allegorical reasons (the apparent size is also rather small for a 6-string 2-octave open-string range).
A sopranino di viola was called for in the 1589 Florentine Intermedii, played to acclaim by Alessandro Striggio. The name implies that it was smaller than a soprano. This seems to be the violetta picciola that Zacconi called a particularly small viol. When reading this in Zacconi, Praetorius couldnt imagine what this double-diminutive viola could be, so he listed the name both as a fiddle and viol. The fashion for this soloistic viol was short lived (one was listed in a German inventory at that time54), and one by Juan Maria da Brescia survives in the Hill Collection in Oxford. Its string stop of 30 cm implies that it was a g-g viol in corista.
Late in the 17th century the dessus de viole, (a 6-string d-d viol), became quite active in playing the generally available violin repertoire in France. (Similarly, earlier in the century, most of the English string music that was not exclusively for viols had the 'treble' part deliberately ambiguous as to whether a treble viol or violin was intended). By the end of the century, a smaller dessus de viole (with a string stop about 3 or 4 frets smaller) with a metal-wound 6th was developed to play at the higher ensemble pitch standards popular then55. This instrument could also be used to play the violin repertoire in first position if it was nominally tuned higher at a lower pitch standard. It was then called a pardessus de viole. The tuning was that of a 6-string g-g viol with the third between the 4th and 5th strings, and the pitch standard was about a minor third lower than modern (called the 'very low French chamber pitch' by Quantz.
Around 1725, a new 5-string pardessus tuning became popular, particularly amongst ladies. The tuning g,d,a,d,g had the high strings like the pardessus de viole and the low strings like the violin. Holding and bowing was like a viol, and frets were used. Earlier pardessus viols were played with the new tuning. Violins were converted by replacing the neck, pegbox, bridge and tailpiece. New instruments were made with either traditional viol bodies or with bodies that were various compromises between a viol and a violin. This new instrument was called either pardessus de viole à cinq cordes or quinton56. Either name was sometimes used for any instrument with that tuning, but the pardessus name was mostly used when the body design was like a viol, and the quinton name was mostly used when the design had obvious violinistic aspects. The popularity of this instrument peaked around 1750 and was waning when the French Revolution terminated it.
1 J. Tinctoris, De Inventione et Usu Musicae (prob. Naples, c.1487), translated by A. Baines, Galpin Society Journal III (1950), pp. 19-26.
2 I. Woodfill, The Early History of the Viol (C.U.P. 1984), p.80.
3 W. F. Prizer, 'Isabella d'Este and Lorenzo da Pavia', Early Music History Vol. 2 (1982) p.110.
4 illustrated in Plates 40 and 44 in Woodfield op. cit.
5 illustrated in Plate 35 and possibly 45 in Woodfield op. cit.
6 illustrated in Plate 33 and 37 in Woodfield op. cit.
7 an Italian example with three double courses is illustrated in Plate 53 in Woodfield op. cit.
8 illustrated in Plate 53 and 54 in Woodfield op. cit.
9 illustrated in Plate 52 in Woodfield op. cit.
10 S. di Ganassi, Regulo Rubertina (Venice 1542) and Lettione Seconda (Venice 1543).
11 R. Baroncini 'Contributo alla storia del violino nel sedicesimo secolo: i <sonadori di violino> della Scuola Grande di San Rocco a Venezia' Recercare Vi (Rome 1994), Appendice A, pp. 136-185. The records of sonadori de lironi in sets occur often from 1530 to 1622.
12 G. M. Lanfranco, Scintille di musica (Brescia 1533); A. della Viola, ms. leaf in copy of M. Savonarola's De tutte le cose che se manzano comunamente (Venice 1515) dated c.1536; S. di Ganassi, op. cit.; D. Ortiz Trattado de glossas (Rome 1553); A. Marinati, Somma di tutte le scienze (Rome 1587).
13 S. Cerreto, Della prattica musica (Naples 1601).
14 M. Mersenne, Harmonie Universelle (Paris 1636), Bk. IV, Prop. V.
15 L. Zacconi, Prattica di musica (Venice 1592).
16 A. Banchieri, Conclusioni del suono dell' organo (Bologna 1609).
17 D. P. Cerone, El Melopeo y Maestro (Naples 1613).
18 M. Praetorius, Syntagma Musicum II, De Organographia (Wolfenbüttel 1619).
19 E. Segerman, 'Further on the Pitch Ranges of Gut Strings', FoMRHI Quarterly 96 (July 1999), Comm.1657, pp.54-8.
20 The theory is in E. Segerman, 'Some theory on pitch instability, inharmonicity and lowest pitch limits', FoMRHI Quarterly 104 (July 2001), Comm.1766, pp.28-29
21 A. Agazzari, Del sonare sopra il basso (Rome 1607), transl. into German by Praetorius in Syntagma Musicum III, p.149, and into English by O. Strunk, Source Readings in Music History - The Baroque Era (New York 1965), p.69.
22 A. Virgiliano, Il Dolcimelo (c.1600) ms in the Civico Museo Bibliographico Musicale in Bologna. See E. Segerman, 'Virgiliano on instruments and transpositions', FoMRHI Quarterly 97 (Oct. 1999), Comm.1672, pp.26-8.
23 S. Virdung, Musica getutscht (Basle 1511).
24 Munich University Library 4o Cod, ms. 718 (1523/4)
25 M. Agricola, Musica instrumentalis deudsch (Wittenberg 1528).
26 H. Judenkünig, Utilis et compendaria introductio (Vienna c.1518).
27 H. Gerle, Musica Teusch (Nuremberg 1532).
28 frescos from "Schloß Goldegg" (1536) and woodblocks on the title pages of the superius, altus and bassus part books of the copies in the Basel Universitätsbibliothek of 'Reutterliedlin' printed by Christian Egenolph in Frankfurt am Main (1535).
29 M. Agricola, Musica instrumentalis deudsch revised edition (Wittenberg 1545).
30 D. Hitzler, Newe Musica (Tubingen 1628).
31 A. S. Instrumentalischer Bettlermantle (mid 17th century), manuscript in the Special Collections of Edinburgh University Library. See E. Segerman, 'Violins, citterns and viols in the Edinburgh 'A.S.' manuscript', FoMRHI Quarterly 91 (Apr. 1998), Comm.1576, pp.38-44.
32 J. J. Prinner, Musicalischer Schlissl (1677), ms. in Music Division, Library of Congress, Washington, D.C., U.S.A.
33 B. Bismantova, Regule per Violoncello e Violone (Ferrara 1694).
34 A. Planyavsky, The Baroque Double Bass Violone (Scarecrow Press 1998), translated by J. Barket.
35 I Woodfill, op. cit. p.196.
36 I Woodfill, op. cit. pp.199-200.
37 P. Jambe de Fer, Epitome musical (Lyons 1556).
38 S. Mareschall, Porta Musices (Basle 1589).
39 J. Rousseau, Traité de la viole (Paris 1687).
40 T. Dart, Galpin Soc J. X (1957), p.88, drawings by Jacques Cellier (c.1585) in plates between pp.62-3.
41 M. Mersenne, Harmonie Universelle (Paris 1636), Premiere Preface Generale au Lecteur. His eulogy of Maduit is in Bk.VII, Prop. XXXI.
42 G. Muffat, Florilegium secundom (Passau 1698)
43 I Woodfill, op. cit. p.212-8.
44 T. Robinson, Schoole of Musicke (London 1603); J. Playford, An introduction to the skill of musick (London 1674); T. Mace, Musick's Monument (London 1676); J. Talbot, Christ Church (Oxford) MS 1187 (c.1694) - see E. Segerman 'The Sizes of English Viols and Talbot's Measurements', Galpin Soc. J. XLVIII (1995), pp.33-45.
45 E. Segerman, Viol-bodied Fiddles, Galpin Soc. J. XLIX (1996), pp.204-6.
46 O begli anni del oro, o secol divo by F. Corteccia.
47 Io vorei Dio d'amore in Lettione seconda (1543) Chapter 16.
48 In March 1609, Edney and Gill applied for a court privilege (that was not successful) for 'the sole making of viols, violins and lutes with the addition of wire strings beside the ordinary strings for bettering of the sound, being an invention of theirs not formerly practised or known.' - D. Lasocki, Galpin Soc. J. XXXVIII (1985), pp.130-1, fn.59. Praetorius (1619) wrote that the English version of the viola bastarda (i.e. the lyra viol) had metal sympathetic strings.
49 after the tuning of the sympathetic strings could no more be in unison with the bowed strings because the wire that could tune as high as gut became unavailable - see E. Segerman, 'A Short History of the Cittern', Galpin Soc. J. LII (1999).
50 M. Tiella, 'On Francesco Antonio Sgargi's book [...] La viola da sei, o sette corde, in Bologna 1747 per Tommaso Colli a S. Tommaso d'Acquino', FoMRHI Quarterly 100 (July 2000), Comm. 1719, pp. 36-9.
51 C. Simpson, The Division-Viol (London 1665).
52 illustrated in Plate 86 in Woodfield op. cit.
53 illustrated in Plate 50 in Woodfield op. cit.
54 'viole de gamba oder die grosz geigen ... 2 pasz, 4 tenor, 2 discant und ain clainer discant' in the 1596 inventory of Archduke Ferdinand's instruments in Ambras Castle - quoted in Woodfield op. cit., p.193.
55 E. Segerman, 'The anomalous size of the pardessus de viole', FoMRHI Quarterly 105 (Oct. 2001), Comm*., pp.*.
56 M. Herzog, 'Is the quinton a viol? A puzzle unravelled', Early Music (Feb. 2000), pp.8-31.