TRACING THE TRUTH

What cello moulds were used in the Stradivari workshop? The question has long gone unanswered, despite the number of artefacts – and even intact moulds – that survive. Philip Ihle examines 17 of the cellos made before 1700 to find out how many moulds may have been used before the mighty ‘B form’ came into play

The c.1690 ‘Barjansky’ cello surrounded by the outline of the uncut ‘Mediceo’ from the same year

Throughout his career, Antonio Stradivari would experiment with different patterns and forms for his instruments, but at the start of the 18th century he became more consistent. We are fortunate that twelve of the internal moulds he used for his violins have survived and are now in the collection of Cremona’s Museo del Violino. This allows us to match them to his finished instruments and track their development (see my article ‘Variations on a Theme’, May 2019). However, the Museo contains no internal moulds for his cellos – and even if it did, most Stradivari cellos have been cut down over the past 300 years. This makes it hard for us to work out how they would have originally looked – or matched up with the moulds. In 2020 I was commissioned by the Art & Strings publishing house to investigate how the c.1690 ‘Barjansky’, formerly owned by Julian Lloyd Webber, would have looked when new, which also gave me an opportunity to examine the age-old andunder-researched question of the cello moulds. In this article I will examine the dimensions of the Stradivari cellos built before 1700, when he started to make cellos of smaller dimensions that have proven more suitable for the whole repertoire.

I was intrigued by the question of whether all the cellos could have come from one mould, and, as in the 2019 article on violin moulds, I have used the relatively new medium of standardised digital instrument photography to provide fresh insights into the matter.

Cello mould attributed to Antonio Stradivari, cat. no. E.901.1 in the collection of the Musée de la Musique, Paris

Cello mould, cat. no. E.901.3. The letters ‘VC’ can be seen inked in the centre of the upper bout

MOULD PHOTOS COLLECTIONS MUSÉE DE LA MUSIQUE/CLAUDE GERMAIN, 2018. BARJANSKY PHOTO LEONHARD RANK

Although the Museo contains no moulds of its own, there are two attributed to Stradivari in the collection of the Musée de la Musique in Paris (see below). Both were bought by the museum in 1880, at the sale of Jean-Baptiste Vuillaume’s estate. The one on the right, E.901.3, carries the letters ‘VC’ in ink, probably standing for ‘violoncello’. The lettering is very similar to that found on Stradivari’s moulds in the Museo del Violino. Both moulds have been significantly altered, making it difficult to match them to Stradivari’s surviving cellos. The dimensions of the left one (E.901.1) are: length 734mm, upper bouts 315mm, C-bouts 205mm, and lower bouts 404mm. Its thickness is 32mm. 

The Museo’s collection, however, does contain a number of paper templates for cello neck and f-hole placements, marked in similar fashion to the violin moulds and templates. These led Simone Fernando Sacconi, who first catalogued the collection, to speculate that the cellos in Stradivari’s workshop were built on moulds called: P; T; B; Q; FN; and ‘Violoncello di Venezia del 1696’. In his 2010 book on Stradivari, Stewart Pollens gives a precise list of these artefacts (MS284–288 and MS291–306) and writes that ‘approximately ten forms are known to have been used’. Also looking at patterns for bridges, corners, and top- and bottom-blocks, Pollens finds four more letters referring to cello forms (and deletes the ‘Violoncello di Venezia’). His full list is: A, B, B picola, BO, MO Bo, FN, P, T and Q.

In 2004 there was an exhibition of Stradivari’s cellos in Cremona. In the exhibition catalogue, Bruce Carlson gives a good account of the development of their dimensions, drawing on the Hill brothers’ 1909 Antonio Stradivari: His Life and Work 1644–1737. There is little disagreement between the two. Here is part of Carlson’s text in the chapter ‘Evolution of the Stradivari Violoncello’:

The first known surviving cellos by the Master date from approximately 1680 and are all of large size, a direct continuation of the Amati tradition. Nearly all of these early examples have had the body dimensions reduced. A fine specimen of this early type, dated 1690, is from the Conservatory collection in Florence, currently located in the Accademia. This rare, uncut example exceeds 79cm in body length with a relative body stop of over 43cm […] Despite the productions that the Master may have seen from his colleagues who were already making cellos on a smaller format, he makes no apparent move until the close of the 17th century. Approximately 25 cellos of this large type can be accounted for today. From the year 1699 a slightly smaller model is introduced. Interestingly enough this date coincides with the Master’s abandonment of experimentation with the series of violins made on the ‘long pattern’ which had begun in 1690.

This corresponds with the Hills’ comment in their book: ‘We estimate that between 1680 and 1700 Stradivari made at least thirty cellos: twenty-five are known to us; and how instructive to note, that all – without exception – were made of large size!’ In the same publication the authors state there are only three cellos from that period whose form and dimensions remain unaltered: the 1690 ‘Mediceo’ (currently in Florence’s Galleria dell’Accademia); the 1696 ‘Aylesford’; and a cello at the time owned by the Marquis de Piccolellis and dated 1697. This last has now been re-dated to 1699 and is housed in the US Library of Congress under the name ‘Castelbarco’.

NOT ALL PHOTOS ARE FREE FROM DISTORTION, AND WE MUST BE CRITICAL WHEN USING THEM

FIGURE 1 The ‘Barjansky’ cello overlaid with a CT visualisation

BARJANSKY PHOTO LEONHARD RANK. CT SCAN DR RUDOLF HOPFNER

In this investigation I compare the shapes of 17 early Stradivari cellos, speculate on their original proportions and show how they might have been altered by restorers.As with my earlier article on violin moulds, I worked with photos and carefully accessed their most obvious sources of error: optical distortion and imprecise scaling due to unreliable measurements. Apart from these technical issues, we must expect a degree of deformation due to shrinkage and warpage of the wood, and also some tolerance during the making process, even when built on the same mould. Owing to the relatively high ribs and thin moulds, this variation from one cello outline to the other would naturally be bigger than in violins.

During the restoration work of the c.1690 ‘Barjansky’ cello undertaken by Leonhard Rank, he compared the plates with prints of photos taken by Jan Röhrmann for the eight-volume Antonius Stradivarius work published by Jost Thöne, and sent me a video of it. The outlines matched very well and there was no distortion visible. Also, at a later stage, Rudolf Hopfner produced a visualisation of a CT scan of the same cello (figure 1) which confirmed the same outline.

I used only one source of high-quality photos and cut them out in Photoshop to create precise outlines. The above test and the four positive comparisons I made with violin outlines overlaid with Jan’s photos convinced me that my source material is not significantly distorted.

THE MAJORITY OF STRADIVARI’S EARLY CELLOS HAVE BEEN CUT DOWN OVER THE CENTURIES TO ACCOMMODATE MODERN REPERTOIRE AND VIRTUOSITY

For the scaling I relied primarily on the measurements in Antonius Stradivarius. I compared them with other sources and, where possible, double-checked them with colleagues who had first-hand data. I discovered the back length of the ‘Castelbarco’ is frequently published incorrectly as 749mm; luthier John Montgomery, a consultant for the Library of Congress, confirmed the length as 772mm when measured with callipers, and 774mm when taken with a tape measure over the arching.

The table above gives: the name of each cello; the various sources I used for its back length; the length I found in this research; the stop length as given in the Thöne books; and finally, the group to which I assigned each cello (more on that later). I added the stop length as it is of particular interest: it shows us what the restorers who cut down each cello assumed would be a good, modern stop. The uncut ‘Mediceo’ cello has a stop length of 430mm whereas the ‘Barjansky’ was cut down to 410mm. The majority of cellos, however, were cut down to today’s most common standard of c.400mm.

To explain the most significant reason for inconsistency during the making process we need to look at the mould itself. The internal moulds for cellos, just like those for violins, have no corners; the outline is interrupted to leave space for the corner-blocks, and Stradivari used small templates to mark the corner lines on the blocks. A cello’s rib height is around 110mm and the moulds in the Musée de la Musique are only 32mm thick; as a result, if a block were glued at an angle it would matter much more than on a violin. Furthermore, even if two cellos were built on the same mould, the differences in the blocks could have added a few more millimetres of flexibility to their C-bout dimensions. How important a factor this is will become clear from the following outline comparisons.

FIGURE 2 Outlines of all 17 pre-1700 cellos, aligned to the horizontal tangent of the right C-bout

FIGURE 3 All 17 outlines, grouped into colours as defined in the table below

ALL DIAGRAMS PHILIP IHLE

Key to the outlines depicted in figure 3

Figure 2 shows the outlines of all 17 cellos superimposed on top of one another. These line drawings were all carefully scaled to the measurements in the table, and straightened to the central, uncut part of the instrument. All have been aligned to the horizontal tangent of the right C-bout, which I chose because it is an uncut spot, somewhat related to each cello’s body stop. The instruments vary in length from 720 to 790mm, the shortest being the ‘Flatback’, and the longest the ‘Mediceo’.

As stated earlier, the majority of Stradivari’s early cellos have been cut down over the centuries to accommodate modern repertoire and technique. Every cello considerably bigger than the gold-standard 756mm of the ‘B form’ risked being cut down. To reduce them it was common for luthiers to make cuts in the upper and lower bouts, hiding the joints underneath the new purfling line. Many cellos were also cut along the centre joints to reduce them in width. This reduced the distance between the C-bouts – but the C-bouts themselves were never cut at all. Hence, comparing these against each other can give us the most reliable clues for understanding on what mould the cellos were built, as opposed to comparing the complete outlines.

Knowing that the C-bouts were not altered during the process of cutting down a big cello, I was thrilled and very surprised to find some C-bouts were over 20mm shorter than others. Here was something to work with! If this difference were down to inconsistencies in the corner-blocks, it would result in wonky rib mitres in the side view and different front and back outlines, but that was not the case with the ‘L’Évêque’ (with short C-bouts) and the ‘Mediceo’.

My next step was to align the cello outlines to the single C-bouts, to try and find groups of similar ones. Together with my assistant Gianmaria Stelzer, we settled on four groups and assigned them each a distinct colour: gold, green, red and grey (figure 3). The length of the C-bouts was the most important criterion when matching the outlines.

FIGURE 4 1690 ‘Mediceo’ 1694 ‘London’ 1698 ‘Magg’

FIGURE 5 1683 ‘L’Évêque’ 1688 ‘Cazenove’ 1690 ‘Mediceo’

The work on the blocks and linings of these cellos is at times very fast and rough. When I worked at Florian Leonhard’s shop I remember seeing the 1694 ‘London’ cello open, when it was in for repair. The interior work was hastily done and looked like it was by a much less skilled hand than Stradivari, whose typical work at the time is neat and masterly. It seemed to me that there was division of labour in the workshop. Despite that rough interior work, its outline matches almost perfectly with that of the ‘Magg’ cello (figure 4). To find two cellos so similar in outline demonstrates that the speed of the work did not prevent consistent outlines from being achieved. This is evidence that these two cellos were indeed built on the same mould. Incidentally, the maple used for the ‘Magg’ and ‘London’ backs seems to have come from the same tree.

It is also worth noting that both cellos were apparently cut down to the same standard in the upper bout, probably targeting a body stop of 400mm. The shortening in the lower bout, however, seems to have been more random, as if there had been less of an established standard for the overall body length. Looking at two very similar examples of the grey group (figure 5) the 1683 ‘L’Évêque’ and 1688 ‘Cazenove’ show this in even more extreme fashion, with the lower bout of the ‘L’Évêque’ of similar length to the ‘Mediceo’, and the upper bout being more than 30mm shorter. The other two cellos in this group, the ‘Flatback’ and ‘Cazenove’, are also from the same decade of the two we are looking at. The curves in the C-bouts of this group are characteristic not only for being short, but also for being rounder than the others.

Figure 6 (page 39) shows all the outlines of the ‘gold’ group: the ‘Archinto’, ‘Mediceo’, ‘Barjansky’, ‘Bonjour’ and ‘St Senoch’, ‘Murray’, along with a representative of the much-celebrated ‘B form’ cellos (the 1710 ‘Rothschild’, ‘Gore-Booth’) for reference. Looking at this group, the consistency of the alterations in the upper and C-bouts is staggering. It shows either a monopoly of violin restoration in one specific workshop, or the complete dominance of Stradivari’s ‘B form’ in the opinion of the restorers – possibly both! I also note that the outline of the ‘Magg’ in figure 4 is very similar to that of the ‘Rothschild’, ‘Gore-Booth’. It looks as if the restorers who cut down the early Strad cellos had an outline of one of the ‘B form’ cellos which they used as a target.

All the cellos in this group have very long C-bouts. All except the ‘Mediceo’ were considerably cut down in length and width. There is great consistency in the upper bout: they are all cut by about 22.5mm in length. In the lower bout, the ‘St Senoch’, ‘Murray’ stands out, having been cut down more than the others.

FIGURE 6 1689 ‘Archinto’ c.1690 ‘Barjansky’ 1690 ‘Mediceo’ 1696 ‘Bonjour’ 1698 ‘St Senoch’, ‘Murray’ 1710 ‘Rothschild’

FIGURE 7 1683 ‘L’Évêque’ 1690 ‘Mediceo’ 1697 ‘Castelbarco’ 1698 ‘Magg’ 1710 ‘Rothschild’

DIAGRAMS PHILIP IHLE

The green group can be seen well in figure 3. With six representatives it is the most numerous group. The ‘Castelbarco’ is the second biggest of all 17 cellos. It is particularly interesting because, as I quoted from the Hill book at the beginning of this piece, it is one of the few that are famously uncut. When comparing it with the other five cellos of the green group, I notice that the 1680 ‘Stradner’ is very similar in outline, which suggests that it might not have been cut down either.

Figure 7 shows one representative of each of the four groups I identified for the pre-1700 cellos, plus the 1710 ‘Rothschild’ as reference. The ‘Mediceo’ (gold) and the ‘Castelbarco’ (grey) have not been cut down. The upper bouts are fairly similar, but the lower half is too different to have come off the same mould. The ‘L’Évêque’ (dark green) from the grey group with the shortest C-bouts and the ‘Magg’ (light green) are more similar and I am not certain that they could not have been built on the same mould. Possibly Stradivari altered the mould around the upper corners, opening the C-bouts to facilitate bow clearance.

Did all the pre-1700 cellos come from one mould? No, surely not. Were they all of big proportions? Well, yes, biggish for sure, but they were not all as huge as the ‘Mediceo’ at over 790mm. I estimate that nearly half came from the Stradivari workshop with a body length of approximately 772mm, which sits right between the post-1700 gold standard of 756mm and the ‘Mediceo’ length. This estimate is based on the fact that the ‘Stradner’ and the ‘Castelbarco’ were not cut down,and that they are representative of the most numerous group in my research – the green group with 6 out of the 17 cellos. There is an overall trend for a development from short C-bouts to longer ones. Then, at the turn of the century, the short C-bouts come back again, together with the short body length of the B-form.

Considering Stradivari’s workload at the time, it is hard to imagine he made many different cello forms during the 1680s and 90s. He built a wide range of instruments, including harps, guitars and different-sized violins and violas. The cello – marked ‘VC’ on his templates and one of the moulds in Paris – was a relatively recent addition to music and its dimensions not yet stable. My research suggests that he used four moulds, but I can imagine a scenario where, rather than making a new one, he chose to move the upper corner-blocks up by a centimetre in order to facilitate bow clearance or a shorter string length. Or perhaps he cut down a mould in the upper bout and scribed on new horizontal C-bout tangents to get a similar effect. His violin moulds were made of walnut, but his cello moulds were made of soft wood. This was much easier to get hold of and work with, but a mould would still have taken him a day or two to make.

It would be interesting to get tracings or undistorted images of the moulds in Paris and compare them with these outlines. They might fit one of the four groups, or maybe even a couple.

This is an edited version of an article published in Monograph of the Antonio Stradivari Cello c.1690 ‘Barjansky’, published by Jost Thöne Verlag, now available at The Strad Shop: bit.ly/3dA2gKk