When I first started looking at this canvas, I was reminded of the episode of “A Stitch in Time” in which Ninya Mikhaila did a reconstruction of the garment in a portrait of a hedgecutter on a country estate. The garment in question was made of leather or buckskin and very plainly at least 50 years behind the fashion at the point at which it appeared in the photo. So it had been made for the Squire, and then gradually been passed from hand to hand, appearing to be almost indestructible, if not as good looking as before. So my first thought was to use soft cotton. However, I don’t have much soft cotton in my stash, and it turned out too heavy for the canvas anyway, so I used pearl cottons and Caron Collection “Watercolours” thread, one strand of the three. Linen stitch is what basketweave tent stitch looks like on the back, and although it can be a bit tricky to do, and it’s more time-consuming than basketweave tent, it’s worth the effort!
As with the lady at the cottage door, I blended several different shades of stranded cotton to create the approximate shades for the face, and used petit point to bring in a little more detail.
The hat was a bit more of a challenge, but in the end I decided it was a woven straw hat rather than a felt one, and created a criss-crossing effect, with something like herringbone stitch, but worked in a rather freestyle fashion.
The pouch hanging at his waist is stranded cotton, tiny cross stitches using the single crossing points of the canvas, and I decided to make his breeches in Kelim stitch, hoping it would make a sort of corduroy effect.
Not really, as it turns out, but it was worth a try!
The saddle is soft cotton, and I’ve used condensed mosaic stitch. I want the level of shine to contrast with the horse, which is probably going to be in Paterna.
I had three colours for the stone – a grey, a cream, and a pale rose. The idea is to echo the colour variation you often see in stone when you start looking, but also, at least some medieval masonry made ornamental use of colour variations. Again, remember that I am looking for a faintly mythological feeling to this, not a historically accurate one, so I don’t mind that I’m tweaking my representation of the Chateau de Tancarville until William’s kinsman wouldn’t recognise his own home. He might, after all, rather like what I’ve done!
I put some thought into whether I would use all three colours in the walls, and if so how, but in the end I decided that since the “sky” will be underside couched in gold, it would be better to put the rose or the grey on the topmost course. Grey felt a bit too lowering, like a stormy sky, and William had quite enough squalls to face in his life, so I chose the rose, and used it also for the mortar.
The narrow course at the base seemed to need a different treatment, still in the the same vein. I used six strands instead of the four I’m using elsewhere, and made a little narrow chequerboard, without any suggestion of mortar.
And here it is, the first turret of the gatehouse complete. I was a little anxious lest the sameness of everlasting split stitch prove exasperating or otherwise offputting, but as it turns out, that’s not the case. I’m finding the different effects of stitch length and direction endlessly intriguing, and while I don’t expect to spend the rest of my stitching days entirely on split stitch, I’m certainly expecting to enjoy William.
As you see, this sketch is from a long time ago. It was probably done in about ten minutes, in a pause on the road out of Palmerston North.
I found it a little while ago, and thought it would be interesting to see whether I could produce something pleasing in stitched form, using this as the basis. In particular, since it’s so long ago, and I couldn’t for the life of me find any other source material, this is all I have.
My first idea was to play with layered laces and gauzes, purples and blues because we were there in winter, and there was a lot of snow, bright blue skies and wonderful coloured shadows. There were some limitations in the laces and gauzes I could get hold of, so some of the adventure will be in making this work!
I haven’t drawn anything on the fabric, and I am cutting my laces and gauzes more or less freehand against the sketch. Compared with the painstaking stitching on William Marshall, the pieces are almost tacked into place, and there isn’t at all the same sense of quiet productivity.
I’m also contemplating a canvaswork, and more abstract, version of Manawatu Gorge, but just while I was writing this post, I came up with an idea for another layered gauze version. So I have more experiments to make!
In other news, Episode 76 of Slow TV Stitchery is now live: In which the Hunting Cat piece is mounted, and we discuss feature creep, tapestries as woven wallpaper with acoustic properties, and the depiction of layers of recollection.
I have a small lightbox now, so I used that to transfer the design. It was a bit of a battle, because the lines showed through so clearly I couldn’t be sure I’d got them all.
And I hadn’t – spot the non-deliberate mistake!
However, one horse’s belly can be added freehand. I haven’t put in the lines for the reins, either, because I’m anticipating putting them over the top of the stitching below, and that layer of stitching would cover the design lines anyway.
When I began the outline, I used the darkest thread I have, “Ebony”, and it looked a bit monolithic. After a couple of comments from Tanya Bentham, I unpicked it all, took the opportunity to restretch the fabric, which had sagged, and reinstated the outlines with some variation – navy blue for the building, brown for the horse, and dark grey for the armour. They are all very dark colours, so there’s only a slight difference, but there is a difference, and already the design looks livelier.
So, once I had the outlining in, I started with the conical slate roofs on the gatehouse turrets.
At this point, I should admit that those conical roofs are not remotely historical. The seventeenth century engraving that gave me the view of the gatehouse I liked the most didn’t have conical roofs, and I think I can be pretty certain that in William’s time those turrets probably had a man-at-arms on patrol, and no roof.
All that said, I’m not doing an entirely historical piece. I’m telling different stories here, and I want a slightly fairytale/mythological quality to the image. I think the conical roofs help with that.
Having spent some time in my stash, deciding that mixing even filament silks might be a bit unpredictable, I finally pulled myself together and placed an order with Devere Yarns. While I waited for them to arrive (it didn’t take long), I went back to Tanya’s book and read the early technique sections again. Homework is always more fun when you’re interested.
I do have some darks left over from the Amarna Family Group, so the outlining is sorted. I’ve got three stone colours, three shades for William’s armour, three greens for grass, three shades for the horse, the three heraldic colours for the shield, and a blue for the slate roofs on the turrets.
The blue is probably also going to be the colour I use for the border.
The Victorians remodelled Temple Church, where William is buried, found a strapping six-footer that they believed – based on what we’ve been told of him – to be William, and in due course, reburied him, and everyone else they’d moved, in the garden. So I’m planning a silk border, of blue laid-&-couched work, embellished with roses (for the garden) and common broom (for the Plantagenet kings he served so faithfully).
The grave slab is shown in a leaflet I got when we visited, which is no longer in existence, but is shown in an engraving from the 17th Century, and was thought to be associated with William. The lion looks right, anyway! I’m wondering whether to include that in my design for the border in some way, and if so, in what orientation..
I decided, when it came to it, that I would sign the Amarna Family Group on the gold, rather than on the support. I’ve been regarding the whole process of assembling the stele with something akin to terror, and signing on the velvet just seemed like bridge too far, especially since it would already be stretched on the frame at the time.
It looks a bit squiffy, I admit, but then it always does!
I’ve also painted the background calico with inktense. I’m going to try very hard to ensure that all edges are properly turned, and either concealed by the pile of the velvet or by other means, but knocking back the gleaming cream should help with the camouflage.
Both of these tasks were done while the fabric was still mounted on the frame, but I want that frame for William Marshall, so I’m going to have to get going…
Finally, I screwed my courage to the sticking point, and cut the threads holding the fabric stretched at the sides, at least half expecting some sort of terrible crinkled effect. But no, everything stayed nice and flat. Sighs of relief all round!
I want to pad the goldwork before I apply it to the stele, so the next stage was to use a photocopy of the goldwork to make patterns for the buckram and wadding I intended to wrap it around.
I could see what I was doing much better than the photograph suggests, I promise – although it wasn’t an entirely straightforward process, and I ran out of oomph at that point.
One of the nice things about the double thread canvases commonly used for painted canvas designs is that you can decide to go right off-piste with them, separate the threads, and use stitches at a different scale.
I’m going to do that with the characters – the woman, the horse and his rider, and the dog – on this piece, because in that way I will be able to use stash threads, and enjoy exploring the highways and byways of ornamental canvaswork stitchery. It may well unbalance the finished piece, so I may not regard it as a success when complete, but I will have had fun along the way!
This, however, was a false start. The stitch is Kelim stitch, or something like that, and the threads are pearl cottons. I was aiming for something that would look a little like a sturdy linen twill, but this didn’t seem to me to be the right choice after all.
One of the disadvantages of going down to petit point scale is that unpicking is a real nightmare. Not fun at all. <shudders dramatically>
I’m much happier with this. I’ve used trammed Gobelins stitch for the apron, and found a use for the dreadful lifeless slubby linen thread that DMC brought out – 20 years ago now? – which I bought because it looked interesting and have found a heartbreaking material to use every time I’ve tried. In this context, it worked very much better than I expected. There’s no shine, so it looks like a heavy, sturdy fabric, and the colour variations, much subtler than if I’d used the designated wool colours, recall the appearance of a crumpled linen fabric, its crisp white knocked back by dust and grime.
The skirt is a combination of a rather gorgeous rayon filament for the lighter sections with a pearl cotton for the darker ones, and I’ve turned Parisian Stitch from vertical to horizontal to help me to create contrasts with the shawl the woman has around her shoulders.
The shawl is in Parisian Stripe Stitch (there’s an extra row of small vertical stitches, which keeps the long stitches in alignment).
I may have ended up with not quite the right shade of green for the vessel the woman is holding. Something about it reminded me of the bright green glazed fragments that used to come out of trenches on Time Team, of happy memory, so I decided to go with the colour, and maybe adjust later when I have everything else in place. I’ve done my best to add a little shape and shading, but in due course I may stitch over this in curving long-&-short stitches. It’s straightforward Madeira stranded cotton, and I know I’ve not finished the skein!
And in other news, Episode 75 of SlowTV Stitchery is now live. In which we discuss the felt “coloured sketch” of the Dig House, John Pendlebury, the obstreperousness of William Marshall, and look ahead to finishing the Amarna Family Group.
Remember this? It was an exploration of canvaswork, a very long time ago, and when I’d finished it, I hadn’t the vaguest idea what to do with it. Then my cousin said she knew exactly what to do with it, and could she have it, please?
She and her local soft furnishing people turned it into a rather fabulous cushion, and it has been sitting happily kicking welly-booted heels in the air on a chair in her conservatory ever since.
Close Up Of The Corner Trim
Fast forward to some recent reorganising in her house, and she has a another chair that might be interested in a companion piece. Not the same shape, and not the same canvas, fortunately, and not necessarily the same colours either, just similar. I have to find the canvas before I can get excited about this one, so it will be a while, but I do have something to think about.
My cousin found a gorgeous drone shot of a knot garden in a National Trust property, and said “Start from this!”. The central tree might even end up as a sculpted piece of one of the pile stitches – that would be fun, wouldn’t it!
I’m thinking I might choose to use a brown canvas this time, and use as many variegated greens as I can lay my hands on. But as I’m not sure where I’ll find the canvas, my ideas have plenty of time in which to change….
So in due course I had finished the coral canvaswork panel, and took it off the frame. Then I had a slight anxiety about the size, and whether there would be space to go all around the sunglasses, and went around three sides with herringbone stitch.
The herringbone stitch rather emphasised the unevenness of one of the sides, but I’m choosing not to mind that, especially as the herringbone stitch had a secondary task; that of attaching some padding fabric to the the back of the canvaswork.
The fourth side was also herringbone stitch, but in that case, it’s the herringbone edging I often use on canvaswork pieces to cover and strengthen edges.
The finished piece was just folded in half and satin stitched to join the edges together. I didn’t herringbone it in this case because it was becoming so very difficult to stitch!
Finally, I took one of those flimsy black fabric cases that you sometimes get with a pair of sunglasses, and stitched it in around the top edge. Lovely though it is to have a bright, strong case for sunglasses, it’s not remotely protective if they keep falling out, and the drawstring will ensure that they stay put.
Since I’ve always been interested in heraldry, and deeply distrust online search results, which so often depend on something unexpected in the search terms, one of my bits of research involved an email to the College of Heralds, asking about William’s coat of arms. I received a commendably prompt and completely unperturbed reply from the Officer-in-Waiting, Rouge Dragon Poursuivant, telling me that:
“The left-hand side of the shield, from the viewer’s point of view, should be Or, meaning gold (or yellow). The right-hand side should be Vert, meaning green. The lion should be Gules, meaning red, and it should be rampant, meaning it is upight and standing on its left foot with its right foot slightly raised off the ground.”
You’d think embroiderers emailed for advice every day of the week! Maybe they do, of course…
I had painted a whole series of variants on the design, and then had another thought, namely – maybe I wasn’t exploring all the variations possible. I attempted to do the exploration on my computer, but went to bed that evening with a ligament in my arm squealing. Back to the paintbox!
So I fished out one of my largest blocks of paper, and painted the chateau with a bit more wall on either side, and an actual path, and then painted a separate William, so that I could move him around on the background and experiment with cropping.
I’ve left his shield as white and green in the design for now. I intend to use underside couching for the “sky”, but I’ve not yet decided whether to use couched gold or silk thread on the shield, and making sure the colours are different in the design will help me to remember I have a decision to make.
I’m still havering, but encouraged by Tanya Bentham, who wrote the book that started me on this, I’m thinking about the third of the four options in this post. As Tanya pointed out on Instagram, in the medieval period people messed around with scale and perspective quite cheerfully, so I have no need to make sure my knight will fit through the gate!
