It is difficult to describe the joy to be had from a private viewing of the beautiful and tiny mandore in the Victoria & Albert Museum. The instrument, repaired in 1640 by the mysterious Monsieur Boissart and probably dated to c. 1570, is exquisite. You can view photographs and read my observations about it in the second of these three articles. The first article traces the history and pre-history of the mandore, with its origins in the lute and gittern familes. This, the third and final article, is a record of the design and making of a new mandore based on, but not a replica of, the V&A’s instrument. The affable and ever-accommodating maker was Paul Baker; the delighted and very lucky player is Ian Pittaway, the author of this piece. This was to be a new journey for us both: me never having played a mandore, Paul never having made one. Includes a video of the completed mandore playing 3 pieces from the John Skene mandore book of 1625-1635, accompanied by photographs of its construction.
(1) Having seen the original in the V&A with Paul, me taking photographs from every angle and Paul taking critical measurements, Paul’s first task was to create a simplified design of the outline, derived from our measurements and photographs, designed on a CADD system (computer-aided design and drafting). The only change was to slightly modify the shape to give me my requested 9 frets on the neck. Historical illustrations variously show a few as 6 frets to as many as 9, all tied gut, with no wooden on-body frets.
(2) The next step was to create a card and foam model to gain an idea of the body shape in three dimensions: we had measurements at critical points but Paul needed to fill in all points between, so this model acted as a belt and braces check. The spikes on the card represent drill holes on the inside of the body, in order to give accurate depth measurements for carving from solid wood, as per the original.
(3) Before arrival, the block of pear wood had been steamed to kill any resident insect larvae and to release internal stresses so it doesn’t twist, thus rendering it stable. When it arrived, it looked unfeasibly small and very orange. A quick check showed that it was, indeed, the correct size and, as we will see, the wood continually changed colour as it was worked. This picture shows the block of pear wood with the drilling template glued on, viewed from the top of what will become a hollow bowl. All points are marked with depth gauges in millimetres for drilling and then carving.
(4) And here is the inside carving of the bowl done. These are not giant clamps – just a very small mandore.
Designing and carving the back
(5) The inside profile was used as a guide for the outside profile, the first stage of which we see completed here. The wall thickness was 6mm, as the carving on the outside was still to be done.
(6), (7) Paul and I had many long conversations about the design. I didn’t want a replica of the original, which would anyway have been quite impossible. Originally I thought of something simple on the back: a pair of wings on each side, emanating from a simple spine travelling up the centre. This idea was based on a 15th century French painted wooden carving of the angel Gabriel, also on display in the V&A. We then discussed designing a flying snake sprawed across the back, an idea uniting Gabriel’s angel wings with Medusa’s snakes on the Boissart mandore. We soon discarded the idea of angel’s wings or a winged snake altogether. We agreed that there were so many rich themes on the original to choose from, and chose decorative flowers, with their tendrils and petals, and the snakes from Medusa’s head. Here we see the beginning of the design. I wanted one snake travelling across the back – Paul thought it would be better with two interweaving snakes, and from that point the design came entirely from Paul’s imagination; with the scroll design on the sides and tail based on the original. We see a radical change of wood colour, which continually changed through carving: removing the outer surface exposed the natural lighter colour underneath; then exposure to light darkened the wood amazingly quickly, in a matter of weeks, rather than decades as with other woods. Oiling the wood darkened it further by filling the pores and smoothing the surface, and it continued to darken naturally after oiling, as we will see.
(8) The bowl design is now fully drawn and the carving has begun, inspired by the original general theme of flowers, leaves, snakes and scrolls. I asked Paul how he came up with the design: “I gazed at it for half an hour to work out where the stems ought to go, then drew it … and rubbed it out … then drew it again … and rubbed it out … then drew it again … then rubbed a bit of it out … then drew it again …” It was pencilled directly onto the mandore, far less complex a process than trying to design it as if flat and then having to take account of visual distortion around the curves. Paul describes the pot/urn from which the flowers grow as “typically renaissance”. He also happens to have a pot very much like it in his back garden!
(9) The work is continually affecting the pear wood colour, now turning decidedly pinky. We discussed the gaps between the snakes needing not to look empty. Paul came up with a spear with entirely gratuitous bosses, as “otherwise it would just be a stick”. Here we can see the carved relief in good light, especially on the flower petals. Experimental carving was performed first on scraps of wood.
(10) Now we see the beginning of more detailed carving to give the impression that one snake is moving underneath the other, and that the spear is under both of them. Paul has turned out many a medieval gittern, which also requires a carved bowl, but in that case smooth on the outside except for a central spine. The small size of the instrument and the level of magnification needed for viewing the work required a high level of concentration, so progress was made in short bursts with plenty of cups of tea in between.
(11) What looks like two giant screwdrivers and a giant novelty pencil and, in (14), a giant rubber, are all standard size, giving a good indication of the scale of the mandore. The screwdrivers were modified by Paul to become tiny chisels.
(12), (13), (14) The carving is advancing, with the foliage now more complete and with more dimensional shaping on the snakes. These photographs show clearly how pear wood changes colour as it is worked, from orange to pinkish then light mauve, lightening as it is carved down, then darkening very quickly over the next few weeks.
(15), (16), (17) The criss-cross skin texture has now been added to the snake bodies, and their eyes have now been carved. Design detail has been added to the urn. The leaves have been shaped and leaf spines carved. The wood colour has now evened and darkened substantially. The neck, yet to be carved, is darker.
(18), (19) Now we have the darkened colour with the first coat of oil. The oil also has the effect of emphasising the carved edges and darkening carved crevices as the oil finds and fills them.
(20) An application of wire wool has taken off the rough edges and smoothed out the carving. The neck has lightened through being worked on: the basic shape of the pegbox has now been formed.
(21) On the original mandore, a double line on the very edge of the neck and a decorative detail runs down both sides of the neck. (22) Paul copied the edging and imitated the running design. This is a typical 16th century carving, simple but highly effective. Paul couldn’t find anything on the market small enough to create this effect, so he made his own tool by recutting a miniature turning gouge.
(23) Paul decided to continue the decoration on the neck through to the back of the pegbox and, since the theme is snakes and flowers, to pencil on then carve a flower motif on the back of the pegbox, the size of a 1p piece, (24) with leaf veins and pollinated centre details. Like all the detail on the mandore, this was carved with the aid of a headset magnifier.
(25) Now the back is complete, and tuning pegs made. The colour is now brown with a red tinge, partly the result of oiling, but largely the result of time.
The rose, soundboard and fingerboard
(26) The standard procedure for carving a rose is to glue a parchment template on the reverse of the soundboard, carve through it, then turn over the soundboard and add detail to the front. This is the original roughly carved Boissart rose, obviously executed without a template, with guide cuts on the edge of the rose as Boissart worked by eye alone. For reasons given in the second of these three articles, this indicates that Boissart was a repairer on the cheap, making a new soundboard, rather than the maker of the entire instrument.
(27) From the Boissart rose, Paul followed the repairer’s intentions, replicating them more tidily on a CADD system, (28) in order to recreate the rose.
(29) The inside of the bowl shows the carving pattern due to the preserving oil soaking through the porous pear wood. The bars are now on the back of the rose to support it, stained black to render them invisible; with the usual bars either side of the rose and a third bar midway towards the tail. This is a simplified version of what you’d find inside a lute, which would have more bars, but in this case only a single bar is needed as the vibrating area of the soundboard is so small.
(30) Now the soundboard is on; the edge yet to be tidied up and ebony edging yet to be added.
(31) Ebony edging has been added to the soundboard and neck, with an ebony bridge and ebony fingerboard. A new feature has been added, not on the original: light stripes between the ebony edging and fingerboard.
A visual issue has arisen: the ebony fingerboard would usually finish a little before the points of the edging, so here the fingerboard falls short of the visual norm. Paul explained that, since the inside of the bowl is scooped out and there is a butt joint (wood is joined simply by butting the two pieces together) between the fingerboard and soundboard, extending the fingerboard further would be to extend it over an internal gap, thus making it inherently weak.
(32), (33) I wanted a longer ebony fingerboard. Here is the internal gap in the body at the neck that the soundboard was sitting over, and that an extended fingerboard would have sat over. A brass clamp to glue in an extra piece of internal supporting wood enabled a supported extension of the ebony fingerboard.
The pegbox snake
(34) We have two carved snakes on the back of the mandore. When we discussed what manner of carving to have on the head of the pegbox, I asked if a snake head was possible. One of the great things about working with Paul (besides his unfailing good humour, patience, generosity and vast store of knowledge) is that ‘not possible’ barely figures in his vocabulary. After looking through hundreds of pictures of snakes, he decided that a cobra, with its expanded hood adding extra width, would create the best visual image. Having found two photographs of a brass cast of a cobra head for a walking stick, front and side views, he printed them out as templates.
(35) The side profile could be stuck on for carving, but the front profile had to be worked by eye. A one pence piece gives the scale. (36) Once it had been roughly carved, then smoothed, (37)(38) it was oiled and given scales. The scales were made using a modified circular punch usually used for taking holes out of baroque guitar roses, now ground down to make semi-circular scale shapes.
(39), (40), (41) I asked for the snake head to have eyes and a tongue: here we have eyes in black paint and a tongue made from ebony, with two fangs of bone, shaped with a small file and inserted with small pliers and a large magnifier.
(42), (43), (44) And so we have a completed mandore, modelled on some of the themes of the Boissart mandore, with an original design for which Paul has the majority of the credit, with initial input and requests from me.
The usual string calculations had been done. I wanted to play in various historical mandore tunings, requiring a two semitone leeway downwards on strings 3 and 5 and up to a minor third tuning downwards on string 1, so I asked for gut strings, as I thought they would be more tolerant to retuning that nylgut.
So the moment of truth came, and I played those gut strings, in the main tuned g’’, c’’, g’, c’, g, with the possibility of the first string to f’’ or e’’, and an alternative tuning of f’’, c’’, f’, c’, f. The top string sounded sweet and sang well, the second was rather dead, the third and fourth moreso, and the thick, heavy fifth string had all the look, feel, resonance and sound quality of a stair rail. So we spent several hours experimenting with different string gauges, with gut and with Aquila’s alternative, nylgut, which is nylon treated to feel, look and sound like gut.
We discovered two altogether surprising things. Firstly, it wasn’t that nylgut worked better than gut; it was that, for a mandore, or at the very least for this mandore, nylgut worked and gut did not. Historical paintings show that some 5 course mandores had single strings on 1-4, with only the 5th doubled with a higher octave. Perhaps we’d just discovered why an octave string was necessary on the bottom course to brighten the dead sound. But the sound of the rest was unsatisfactory, too. So secondly, we discovered that not only was nylgut superior to gut in its sound and resonance, but that the instrument sounded so much sweeter and more resonant at string tensions significantly below what the calculations predicted. So, after much experimentation, we now have Aquila nylgut on 1, 2 and 3, Savarez KF (their alternative to gut basses) on 4, and an Aquila DE on 5 (nylgut overwound with copper then varnished, thus it sounds like gut, lacking the unnatural sustain old overwounds used to have, and with the advantage of being significantly thinner than gut, especially on a tiny instrument like this).
(45) I’ve made several references to the mandore’s small size. The size of the latches on the case, (46) and the photograph of it in my hands illustrate this well. The vibrating string length is a mere 281mm, exactly as on the Boissart-restored mandore in the V&A. This makes it, at first, a challenge to play, as the left hand fingering I’d usually use for the lute isn’t possible: there just isn’t the space. The right hand, too, has to be re-thought: I’ve found the lute’s thumb inside technique to be nigh on impossible while holding the instrument steady which, again, due to its small size, takes a little getting used to. A strap button was fitted but a strap proved impractical as it just got in the way. As with all new instruments, I found a little perseverance and experimentation soon pays dividends and in a short time the mandore became comfortable.
So what does the mandore, and this mandore in particular, sound like? Within hours of receiving it, Paul recorded me playing three pieces from the Scottish handwritten John Skene mandore book of 1625-1635: Comoedians Maske, I longe for your virginitie, and Ioy to the personne, which you can hear by clicking on the picture below. The video will open in a new window, and the playing is accompanied by images of the mandore being built.
The instrument the early music revival has forgotten
The mandore is the instrument the early music revival has forgotten. There are three surviving and substantial mandore books – the John Skene ms., Scotland, 1625-1635; the Tresor ms., France, c. 1626; and Francois Chancy’s printed Tablature de Mandore, France, 1629 – plus music in other sources if you’re prepared to search. The music is well worth playing, yet very little of the material is commercially available, and none in mandore tablature. Instead, copies of copies of the three surviving mandore books are passed from person to interested person, once owner and seeker have found each other. The commercial unviability of those books is unsurprising, since there has never been a mandore revival, as there has been a lute revival, and thus there are almost no mandore players. That, to my mind, is a great shame, since it is a gap in the soundworld early music enthusiasts are trying to recreate. With my new mandore, I’ll be doing my bit to fill that gap.
All photographs by Paul Baker except the article header photograph and pictures (21), (26), (39), (40) and (41) by Ian Pittaway.