Analysis of the origins of a large Franco-Flemish double-manual harpsichord.  Would a Ruckers by any other name sound as sweet?

  

Introduction

            An extremely fine 5-octave double-manual harpsichord, on temporary deposit in the Russell Collection, has recently been entrusted to me for examination and study.  The general appearance of the instrument has all of the usual characteristics of a late eighteenth-century French instrument.  Indeed, at the time of a restoration made in 1971 it was found to bear the signature of Jacques Barberini who worked in Paris in the second half of the eighteenth century.  It has a French carved and gilt stand, case paintings in the style of François Boucher, a painted soundboard and lid, etc.  It also has a Ruckers soundboard rosette with the initials ‘HR’ and this appears to be genuine.

 

 

Plate 1 - Soundboard rosette

Franco-Flemish double-manual harpsichord

Private ownership, Brazil

 

Closer examination reveals that the tail and bentside have been lengthened to increase the width of the instrument on the bass and treble sides and the soundboard and bridges have also clearly been extended at both ends of the compass.  These can be seen in Plate 2.

  

            

                  Plate 2 - 1786 Ravalement join in the bentside shown by the arrows (left) and the treble extensions to the 8' and 4' bridges and soundboard (right).

Franco-Flemish double-manual harpsichord

Private ownership, Brazil

Both of these photographs were taken before the linseed oil varnish was removed from the case and the soundboard.

 

Indeed, the instrument bears all of the usual signs of an eighteenth-century ravalement of a seventeenth-century Flemish instrument.  And the presence of the ‘HR’ rose suggests that it might even be a ravalé Ruckers.  This paper is a partial analysis of this instrument and concentrates on the determination of its original state and origins.

 

 

The eighteenth-century state

            Before commencing on the analysis of the original state however, it is necessary to say a few words about the eighteenth-century state.  The instrument is now a typical 5-octave F1 to f3 double-manual harpsichord.  The style of the decoration and stand strongly suggest that it was re-built and decorated in France.  The stand is somewhat higher than normal and is extremely fine in its carving and gilding.  The external decoration is in the style of the Parisian Martin brothers and is also extremely rich and elaborate.  According to the owner the signature or calling card of Jacques Barberini (known dates 1775-91) was found inside the instrument when the instrument was restored in Rio di Janeiro by Roberto de Regina in 1971.  However, despite a great deal of effort having been made to find it, no trace of this signature could be found on any of the external surfaces of the instrument normally exposed to view.  In addition, an endoscopic examination of the interior of the instrument was made and a separate examination by Willie Hendry using a video webcam was made in order to try to find the Barberini signature.  No trace of this inscription was found using either method.  Since the present baseboard is not the original it therefore seems likely that the Barberini signature referred to by the owner must have been written on the original baseboard and that all traces of this signature disappeared when the baseboard was replaced in 1971.  

            On the other hand there is a faint inscription written in pencil or silverpoint on the front surface of the lower belly rail.  This appears to read:  “?? fait par hoffmann a Paris”.  It has not been possible to transcribe the first part of this inscription (which probably consists of no more than one short word, probably just “Refait”).  At any rate this is probably the signature of Nicolas Hoffmann who is known to have been active in Paris from 1781-90).  Various features of the present state suggest that the ravalement was carried out in two stages, and this seems to be consistent with it having been worked on separately in different periods by Barberini and Hofmann both of whom are known to have lived and worked in Paris in the second half of the eighteenth century.

             The instrument was given four registers in the second of the ravalement states and it appears once to have had a pedal or genouillère to operate the registers.  However, one of the registers (presumably a peau de buffle) as well as the pedal or genouillère appears to have been removed in the 1971 restoration when the eighteenth-century baseboard was replaced.

  

 

The original state

            I now want to concentrate on the main thrust of this paper which is a new and unusual way of determining the city of its origin which is different from my usual analysis of the baseboard layout.  The ‘HR’ rosette (see Plate 1) now found in the soundboard is too small to fit into the hole into which it has been placed suggesting immediately that the instrument was not originally a product of the Ruckers workshops.  The rose hole is 81mm in diameter, whereas the rose itself has a diameter of 65mm and is that usual for an ‘HR’ rose of this type.  The rose appears to be genuine and of the type that was used by Ioannes Ruckers in surviving instruments dating from 1598 to 1617.  This suggests that the instrument has been falsified by placing a Ioannes Ruckers rose in it and indeed close examination of the rest of the instrument shows that it possesses none of the usual characteristics of the normal Ruckers design and construction.  So where then were the original parts of the instrument made?

             A close examination of the pinning of the bridges shows that they have not been re-pinned in the course of the eighteenth-century ravalements.  Instead the original pinning has been re-used  The pinning of the 4' bridge and the 4' hitchpin rail indicates that, separated by the 12 notes of the normal chromatic octave, there were the usual doubled strings for the eI/gT notes like those of a Ruckers double-manual harpsichord.

             The distances of the short 8' and 4' bridgepins from the inside of the spine were measured and these are given below in Table 1.  The underlining of the measurements in the two right-hand columns indicates the transitions from those pins on the original portions of the bridges (shaded light grey) to the pins on the ravalement added sections (shaded dark grey).  Using the positions of the doubled sets of strings given by the doubled sets of 4' bridge and hitchpins it is possible to assign the pitches of the original notes of the strings which would have been placed at each of the bridge pins.  The difference in pitch of one semitone between the original upper-manual notes and the notes of the present compass indicate that the person who carried out the ravalement encountered the van Blankenburg Problem in the usual way which I have discussed and analysed in my book on the construction methods of the Ruckers family.  The notes for the original strings of each bridge pin are shown along with the present notes in the two left-hand columns of Table 1 below. 

 

     

Distances from pin to the spine

Orig. l.m. note

Orig. u.m. note

Present Note

Short 8'

4'

 

 

f3

900.2

887.9

 

 

e3

886.9

874.4

 

 

eI3

870.2

859.9

 

 

d3

855.9

845.6

 

 

cT3

841.7

831.1

 

 

c3

826.8

817.3

f3

c3

b2

813.6

803.6

e3

b2

bI2

799.4

789.7

eI3

bI2

a2

784.9

775.4

d3

a2

gT2

770.9

761.3

cT3

gT2

g2

757.3

746.6

c3

g2

fT2

742.8

731.2

b2

fT2

f2

727.4

718.3

bI2

f2

e2

712.6

703.6

a2

e2

eI2

698.0

689.1

gT2

eI2

d2

684.0

673.3, 676.0

g2

d2

cT2

670.4

660.0

fT2

cT2

c2

654.2

645.6

f2

c2

b1

639.1

631.3

e2

b1

bI1

624.9

618.4

eI2

bI1

a1

601.0

605.0

d2

a1

gT1

597.0

589.6

cT2

gT1

g1

584.2

575.4

c2

g1

fT1

571.6

561.5

b1

fT1

f1

557.1

547.2

bI1

f1

e1

542.3

533.7

a1

e1

eI1

528.5

519.1

gT1

eI1

d1

515.7

504.4, 505.5

g1

d1

cT1

502.0

490.4

fT1

cT1

c1

486.1

476.6

f1

c1

B

470.5

461.0

e1

b

bI

456.1

448.3

eI1

bI

A

441.5

434.2

d1

a

gT

428.0

420.0

cT1

gT

G

414.3

407.1

c1

g

fT

400.8

391.4

b

fT

F

385.9

377.5

bI

f

E

371.9

363.3

a

e

eI

357.4

349.6

gT

eI

D

344.0

334.0, 335.0

g

d

cT

330.5

321.2

fT

cT

C

316.5

308.2

f

c

B

303.7

294.6

e

B

BI

290.1

280.6

eI

BI

A

277.2

267.7

d

A

GT

264.7

254.0

cT

---

G

251.7

240.5

c

G

FT

239.1

228.1

B

---

F

227.8

215.9

BI

F

E

214.3

203.8

A

E

EI

201.9

191.0

E/GT

---

D

189.8

178.4

G

D

CT

175.9

166.0

D/FT

---

C

162.8

151.8

F

C

B1

150.3

137.0

C/E

 

BI1

139.5

122.9

 

 

A1

123.0

108.5

 

 

GT1

112.1

97.0

 

 

G1

101.2

83.5

 

 

FT1

89.3

70.4

 

 

F1

77.5

58.3

Table 1 - Distance of the short 8' bridge pins and the 4' bridge pins from the spine in millimetres.

The measurements with a darker grey background correspond to pins on the ravalement additions to the bridges.

 

            These distances are plotted in the graph shown in Figure 1 below.  Here the distances of the 8' bridge pins to the spine are indicated with small circles, those of the 4' pins with a small square and the ends of the original sections of the bridges are marked with vertical lines.

 

 Figure 1 - Graph plotting the distance of the bridge pins from the spine for the present compass.

 

            The graphs of the bridge pin positions seen in Figure 1 show that the bridge pinning of both the 8' and the 4' bridges is two distinct sections.  Above the present 8' played note c (the original lower-manual note fT)[1] on both bridges the bridge pins have one spacing relative to one another, and below c they have another slightly smaller spacing.  This is indicated by the slight change in the slope of the line joining the plotted points above and below the 8' played note c and above and below the 4' played note c1 (which are, of course, played by the same keylever). 

            This slight change in the spacing of the bass bridge pinning is quite normal and a feature of almost all harpsichords made in all schools of harpsichord building throughout the whole of the historical period.  The reason for this can be explained as follows.  In order to improve the quality of the bass notes it was found to be necessary to move the bass end of the bridge as far from the spine side of the instrument as possible.  This gave a wider spacing on the bass side between the end of the bridges and spine liner.  This gave a more flexible and unhindered vibrating soundboard area near the bass ends of the bridges.  Without this additional space the soundboard between the ends of the bridges and the spine soundboard liner would have been stiff and inflexible and the quality of the bass sound would have been adversely affected.  Therefore the strings were positioned parallel with one another and with the spine above the present played note c, but from this note downwards to the bottom note the strings angled away from the spine (originally this would have amounted to the bottom 15 notes from C/E to fT on the lower manual – see Table 1).  At the soundboard bridge end the strings gradually became further and further from the spine than they were at the nut, and therefore they became gradually less and less parallel to the spine. 

            Only the original spacing of the strings between the vertical lines was used in the analysis below.  However, it is also clear from Figure 1 that the person who carried out the ravalement was meticulous in the way he spaced the strings on the bridge extensions.  The graph shows that the same spacing as that used by the original maker was used for the treble notes above the ends of the original bridges, and a different spacing was used in the bass below the original ends of the bridges.  But the spacing of the strings on the bridge extensions matched closely that used by the original maker.  That is to say, the slopes of the lines plotting the pin positions is accurately similar for the pins added on the bridge extensions (dashed line extending the line fitted to the points for the original section of the bridge) as for the original pins in each part of the compass.

             In order to calculate the spacing of the strings in each part of the compass, the measured points have been fitted to a straight line using a regression analysis and the method of least squares.  Each section of the graph can be represented mathematically by a simple straight-line equation of the form y = mx + b where y is the position of the pin, x is an integer corresponding to the number of the note, m is the regular spacing of the notes along the straight line, and b is the ordinate where the line crosses the y axis (ie it is the value of y when x=0).  A regression analysis was carried out for each of the straight lines segments shown in Figure 1, and the value of m, the slope or the regular spacing of the strings, and b, the value of y when x=0, were determined for these.  The results are shown in Table 2 below.

 

                                                                 Bass section           Middle and treble section

                                                                   m            b                          m            b

                                      8' bridge:             12.67       74.8                     14.18       46.2

                                      4' bridge:             12.89       61.8                     14.17       37.3

                                       Average:             12.78                                  14.175

Table 2 - The values of the constants m and b in the straight-line equation y = mx + b found from the regression analysis of the bridge pin positions for the bass and treble sections

 

            Here it is the regular spacing of the strings, given by the constant m, which is of interest for each of the bridges and for each section of the pinning above and below the present played note c.  Not surprisingly the value of m is very nearly the same for both of the bass sections and for both of the middle and treble sections of the two bridges.  This reflects the fact that the strings require the same spacing in each section so that the 4' and 8' strings stay the same distance apart throughout the compass of the instrument.  The two values have therefore been averaged in order to get the string spacing for each section.

            For this instrument the treble bridge pin spacing, expressed in terms of the  Antwerp duim[2] is:

At first sight this seems a rather strange and arbitrary spacing  However, this spacing is very close (error is only 1%) to 0.550 mm/duim and, when expressed in terms of the Antwerp voet with 11 duimen[3], then this same ratio gives:

 

            This suggests very strongly that the maker spaced the middle and treble strings, jackslots and keytails out such that 20 of each occupied a space of 1 Flemish voet.

 

            In the bass the average 4' and 8' string spacing was only 12.78mm/note.  Hence the spacing of the bass bridge pins, again expressed in terms of the duim, is:

 

In a similar way this value, in turn, is very close (error is 0.3%) to 0.50 duim/string = ½ duim/string.

            This, in turn, suggests very strongly that the maker spaced the bottom 15 bass strings on the 8' and 4' bridges such that each occupied a space of a half a duim, or so that 22 strings occupied a space of exactly 1 Flemish voet

            The simplicity and elegance of the way the strings were spaced on the bridges is striking.  However, these are totally different from the way Ruckers/Couchet spaced their strings.  For example I have been able to show that the Ruckers spaced their strings so that 1 octave of strings was just 6½ duimen giving an average treble spacing of 13.802mm instead of the 14.175mm found here.  This difference is illustrated most clearly by comparing the 3-octave spans of the two.  If the octave spacing of the keyfronts was the same as that of the keytails in the middle and treble parts of the compass then three octaves of notes or 3 octaves of strings with 36 notes would have a spacing of:

 

 36 x 14.175mm = 510.3mm

 

This is considerably greater than the 500mm found universally on the instruments built in the Ruckers tradition and, although the analysis leading to this result is rather complicated, is one of the strongest and clearest indications that this harpsichord was not built by one of the members of the Ruckers/Couchet family.

 

Conclusions

            In its original state this instrument must have had 2 keyboards.  The presence of the doubled 4' hitchpin and bridge pins for the original eI/gT notes indicates that the keyboards must have been displaced in pitch from one another by an interval of a musical fourth like the normal Ruckers double-manual harpsichords.  An analysis of the placement of the doubled pinning indicates that the upper-manual keyboard must have had a compass of C/E to c3 and the lower-manual keyboard must have had the usual compass of C/E to f3.  All of this strongly suggests that this harpsichord was originally made in Antwerp.  However, the simplicity and elegance of the use of the Antwerp small duim in the design and layout of the lateral spacing of the original bridge pins confirms conclusively that it was made there.  Any attempt at using the voet and duim used in Amsterdam or Brussels, or indeed the French pouce, or the unit of measurement of anywhere else in the nearby regions yields numbers which to not show the same simple and elegant design as that demonstrated here.  Only the Antwerp small duim gives such results.  Clearly the use of the local of unit of measurement permeated all aspects of harpsichord and virginal design from the layout of the baseboard to string scaling design to the regular lateral spacing of the strings in the treble and bass.

            This harpsichord therefore adds its name to the ever-growing number of instruments that are being recognised as harpsichords made in Antwerp, but not by one of the members of the Ruckers/Couchet family.  It is also an instrument in which the ravalement has been competently, professionally and accurately carried out.  Once restored it probably will sound just as sweet as the numerous examples of French ravalements of seventeenth-century French instruments or of other non-Ruckers Flemish instruments.  How many of us could distinguish a French ravalement of a genuine Ruckers or Couchet double-manual harpsichord from a ravalement of a non-Ruckers Flemish instrument or of a French seventeenth-century harpsichord?  I could not.

 

- Written and prepared by Grant O’Brien, July/August, 2003

This paper was first published in the Early Keyboard Journal, 22, 2004, p. 49-80

Much more is now known about this harpsichord and can be found on this website at:  http://claviantica.com/Franco-Flemish_harpsichord.htm


[1] Note that the Ruckers and some other Flemish makers designed and marked out their instruments around the notes c and fT.  Pin holes for positioning the bridges during glueing are often visible at these notes on the soundboards and bridges of their instruments.

[2] In Antwerp the small duim had a length equal to 25.48mm and was used to measure out smaller dimensions such as the case height and string scalings.  See pages 69 and 70 of my book Ruckers.  A Harpsichord and Virginal Building Tradition, (Cambridge University Press, Cambridge, 1990) where the length of the small duim = 25.48mm is derived.

[3] The Flemish voet or foot had an unusual subdivision into 11 duimen or ‘thumbs’, and not into 12 as was much more common elsewhere in Europe.