The wire strung-chitarrone after 1620: some considerations

                                                                                                             by  Mimmo Peruffo

 

In the present work I use the terms theorbo and chitarrone as perfectly synonymous, as they were in Italy in the 17^th century.  That the chitarrone could have been strung also with metal strings is a fact, well supported by some sources from the first half of the 17^th century, i.e. Praetorius' ’Syntagma musicum (1620): (1)

 

‘Deren sennd nun zwenerlen Arten; Die eine mit Gegensäitten: Die andere mit Messings und Stälenen Säitten. Und mit solchen Säitten beziehen auch etliche jesst die rechten gemeinen Lauten: Uber die Quarta und Quinta wird alsdann umb eine Octav tieffer/als sonsten gestimmet/gleich wie in der Theorba. Und das darumb/dieweil in der Theorba die lenge des Corporis, und die Messings Säitten/solches nicht anders leiden/und die rechte höhe nicht erreichen Können.

 

There are now two types: those with gut strings, and those with brass and steel [i.e. iron] strings, and some these days supply the common Lute with such strings, in which case the first and second courses are tuned an octave lower than otherwise, like Theorbo. And since the Theorbo has such a long body, it cannot be tuned so high with brass strings.’

 

In Piccinini's Intavolatura di liuto…(1623) we find the following passage. (2)

 

‘...Dico similmente, che il Chitarrone armato di corde di cetra, come s’usa particolarmente in Bologna rende armonia molto suave, & apporta leggiadra novità all’orecchio. Hora che gli hò levato alcune imperfettioni, e trovato altro modo di fabricare detti stromenti, che di bontà sono migliorati assaissimo, havendoli rimesso la quinta corda, e la sesta, & li contrabassi di fila d’argento, & ogni contrabasso con la tratta longa, e corta conforme il bisogno…

 

And likewise I say that the chitarrone strung with cittern strings, like we do especially in Bologna, produces a very sweet harmony and brings a pleasant novelty to the ear.  Now I have amended some imperfections and found a new way of building such instruments, which have enormously improved their quality, since I strung the fifth and sixth strings and the diapasons with silver wire, namely every contrabbasso, long or short according to need...’

 

We don't know how widespread such usage was (he specifically refers to Bologna); what is certain is that to string a theorbo with either gut or metal strings today presents no problem whatsoever and this may lead us to believe this would be the case also in Piccinini's time.

 However, on the basis of what we know about the mechanical characteristics of metal wires after 1620 ca (i.e. after the Meuler’s extra strong steel production), we are inevitably confronted with a series of queries.  To talk about metal strings means above all to talk about those used- for stringing spinets and harpsichords and were made of diverse materials, iron, brass and silver in particular.  We can reasonably assume that the same wire would be employed for citterns and other plucked instruments, such as the chitarra battente, the orpharion and the chitarrone, since we have no ground to believe that there may have been a specific manufacturing technology for producing metal strings for plucked musical instruments other than that employed for spinet and harpsichord string-making.

It is worth its while to spend a few words about the chitarra battente: its origin is assumed to be at the beginning of the 18^th century, lacking any iconographical evidence or musical sources from the 17^th century (3).

 

The surviving chitarre battenti are generally believed to be alterations of previous gut strung instruments, carried out by 18^th century luthiers: shortening the neck and re-fitting the peg-plate and raising the bridge higher up the belly in order to shorten the string length, scooping down the sides at the lower bout and bending the belly at an angle that will allow the bridge to exert more pressure while being kept in position by the strings that in turn are secured to hitch-pins inserted at the bottom end of the instruments are the necessary alterations. 

We can but wonder about the quality of the workmanship of such operations.  But  in the inventory taken in the workshop of the Roman guitar-maker Lorenzo Filzer (4) in 1657 we find “Dodici chitarre con corde de Cetra a b. 50 l’uno…” (12 guitars with cittern strings…) as well as many guitars “alla spagnuola" (in the Spanish fashion), “chitarre alla quarta” (a fourth apart) and "chitarrini alla taliana” (Italian small guitars, i.e. 4 course), as well as some 17^th century- paints. (5)  This document alone, however, should be enough to throw some doubt on some established points of view about the times of the chitarra battente.

  We do have a general idea about the mechanical properties of metal strings in the 17^th century.  The manufacturing technology does not seem to be much different from today's.  Atthanasius Kircher (6) tells us they could draw wires as thin as a hair, but we also know that the perfection of roundness of the gauge holes in the draw-plates left much to be desired when compared to what can be achieved today . A further difference lies in the degree of purity of the metals employed today to make harpsichord strings. 

According to Segerman (7), metallic impurities like zinc, tin, aluminium, antimony, or non-metallic ones like sulphur and arsenic played a role in contributing to make the metals of the past harder (and as a direct consequence more brittle by playing) than ours.  The brittleness of those metal wires was most probably a real problem, since it became inevitable to conceive a new way of fixing the strings different from that typical of gut strings (i. e. the traditional lute bridge, that was also the usual one on theorbos), as far as we can see on extant instruments.

The traditional bridge, in fact, must have proved itself inadequate since it forced the strings to be bent at angles that the wires of the time were, quite probably, unable to stand without breaking.  This would explain why the strings were attached to hitch-pins at the bottom side of the instrument, a method typical of citterns also adopted for metal strung mandolins in the 18^th century.  I think we can confidently say that wherever metal strings were certainly used we should also expect to find such a string anchoring method; however, the only example of an instrument of the lute-family showing such a feature, on the treatises, can be found on the plate XVI of Praetorius' ‘De Organographia’, namely the ‘Testudo theorbata’ .

 

The orpharion may seem to be an exception by certainly having metal strings and a lute-type bridge.  In fact, the back of the bridge is provided with metal hitch-pins but the essence of the matter does not change: strings do not reach –in this case- a dangerous bent.  The belly bent at an angle which is a feature typical of mandolins and chitarre battenti was probably indispensable to produce a higher downwards string pressure on the belly, since movable bridges are much lower than traditional fixed ones.

 

We have a general idea of the breaking load values of ancient metal wires thanks to the data recorded by Mersenne -even if their accuracy has been queried- (8), from other historical sources and by hundreds of measurements carried out on pieces of old spinet/harpsichord strings; although here we must point out that the dating of the wires is very often uncertain. But before I will recall some things about strings.

 

The ‘breaking point’ defines the tension at which a string of 1 mm² in section (about 1.13 mm in diameter) breaks. At equal vibrating string length, strings of different diameters –but of the same material- will break, theoretically, at the same frequency, known as the ‘breaking frequency’ (This may seem strange to those unfamiliar with this area of study –fatter strings are stronger than thin ones- but of course need to be cranked up to a higher tension to vibrate at the same frequency). Thorough the general equation for strings, it can be inferred that the product of this critical frequency times the vibrating length (in metres) is a constant K called “breaking index”.

 

Such relation when espressed in other terms (K/vibrating length) allows one to quickly calculate the theoretical upper limit for the frequency at which the top string of each historical instrument will break (see E. Segerman, footnote 7).

 

The common procedure  for tuning a lute, violin or viol for “solo” playing was, as is well known, to tune up the treble string, empirically, to the highest possible pitch, stopping short of breaking, i.e.before reaching its breaking point. By keeping slightly below breaking frequency, ancient musicians obviously aimed for an acceptable compromise between the top string’s playing life and safeguarding the best acoustical performance of the registers.

 

For  instruments with a definited nominal pitch-tuning it is to be expected that this rule could have taken a perfectly inverse route, exploiting, in practice, the maximium string vibrating length which the breaking point, or rather the breaking frequency allowed.

Thus, a playing pitch, the treble string worked in both cases at close to breaking point.

 

Based on these concepts, several scholars today suggest –with the metal strings for Plucked and Keyboard instruments- that working at one or two semitons below breaking frequency may be considered a reasonable choice suported by historical sources.(9)

 

Mersenne recorded the breaking point of some metal wires of his time, from which was easily determind their breaking index (10):

 

Silver:  155 Hz/m

Iron:    160 Hz/m

Brass:  150 Hz/m

 

Let us compare them to the breaking index of the strings commonly available on the market today, especially steel and phosphorous bronze, again from Segerman's table:

 

Iron:                            130-165 Hz/m

Phosphorous bronze:   145-165 Hz/m

Steel:                            280-295 Hz/m

Beryllium bronze :         210 Hz/m

Silver 750:                     150 Hz/m.

 

Two remarks: first, if the originar high purity of today's metals actually grants a higher degree of pliability, this probably explains why it is possible to use a traditional lute type bridge on the modern chitarrone; second, the breaking indexes of modern materials are noticeably higher than those of the past, and at least in one case (piano steel wire) even higher than gut, that is 260 Hz/m average (our tests on commercial gut strings).  It has already been made clear (11) that the fact that the breaking index of gut is higher than that of iron, although its breaking load is not, should not surprise us: metals possess, in fact, a higher tensile resistance than gut, but the highest pitch they can be tuned to, i.e. their breaking index, is inversely proportional to the density of the material. In fact, “ancient” iron with a mean breaking load of 100 Kg/mm² has a breaking index of only 178 Hz/m.  This explains why 17^th “gut”-guitars can have string lengths of 68-73 cm, whereas the battente (at the same standard pitch) must be kept in the 55-58 cm bracket. 

 

Other indirect information about the tensile resistance of metal wires is given us by instruments like the four course mandolin described in Fouchetti's Parisian tutor (c. 1770). 

The huge time lapse from Piccinini's and Practorius' time should not worry us overmuch, since within that timespan no technological improvement in the iron wire manufacturing seems to have taken place, and this is confirmed by the stringing layout.

 Fouchetti's mandolin has only the top string of gut, and the reason has already been pointed out:  the strongest iron strings of that time could not be tuned up at the pitch required for the treble at the given string length.

 

Segerman calculated the probable working breaking index of the 18^th c. mandolin gut-top string (12) at 176 Hz/m.  For the second course the breaking index falls perfectly well within the possibilities of  18^th century's brass wire. Thus, old iron strings couldn't stand then at working indices beyond 170-180 Hz/m.  The same facts apply to extant chitarre battenti from the seventeeth and eighteenth centuries having an average string length of c. 57 cm. lf the top course is tuned at 'e' that means a working breaking index of about 166 Hz/m at the Roman pitch of the time, i.e. about  ‘a’-390 Hz (13). 

Many ‘ancient’ pieces of spinet and harpsichord wire strings have been tested, but the range of breaking indexes turns out to be rather wide: this partly depends from the known fact that the tensile resistance is directly proportional to the number of times a wire has been pulled through the draw-plate the wire will be much more resistant but also more brittle by bending.  Here are some breaking indexes as ascertained by some researchers:

 

'Old' harpsicord iron: 158-188 Hz/m; mean 173 Hz/m. (14)

'Old' spinet and harpsicord iron: 164-187 Hz/m; mean 175 Hz/m. (15)

Old' spinet iron from the second half of the 17^th century: 159-195 Hz/m; mean 177 Hz/m. (16)

 

Other metals:

'Old' copper alloys: 112-138 Hz/m; mean 125 Hz/m. (17)

'Old' brass: 101-155 Hz/m; mean 128 Hz/m. (18)

'Old' brass: 148-153 Hz/m; mean 150 Hz/m. (19)