In my most recent post, I reported on the inspiration and thought process behind the construction of my Johanna’s Jacket: five knitters used a work by another knitter as a jumping-off point for their own very different jackets, and the others’ work informed the knitting each knitter did. Now I’m going to get very specific about what I did and how I made my design decisions.
When I committed to joining the Johanna’s Jacket KAL, the only thing I knew I wanted to do was to use the bright colors I had dyed in 2018 and 2019 with friends in the UK using cochineal, madder, and indigo, dyes that are outside my repertoire of weeds and waste. I wanted to use my weed-and-waste dyed yarns too, since that’s what I mostly have, but for some reason I hadn’t made much use of the pinks, oranges, and blues I had dyed with Liz and Gwen, and that was just unconscionable. Of course I was going to use stranded knitting because that’s my comfort zone, and I had been playing with improvised double-knitting during my January 2019 Swatchathon and thought it was just about the funnest thing ever. When Priscilla, one of the members of my Ravelry group, started knitting her Johanna’s Jacket using a stranded pattern of 3-stitch squares in a small checkerboard pattern, I started thinking about contrasting the little checkerboard with larger checkerboards and playing off the checkerboards against circles and concentric circles. So I started by making big fold-up pirate cuffs in double-knitting with irregularly spaced circles of various sizes floating around like bubbles, in a pattern that I made up as I knitted. For each cuff, I chose colors that were similar but clearly different, an orange madder against a greenish yellow, maybe from onion skin, for one cuff, and a cochineal salmon pink against a mustard-ish yellow, from fig leaves, for the other.
I used the theme of “same but different” throughout the execution of all of the color work. Each sleeve followed the same rota of little checkerboard, big checkerboard, concentric circles, then repeat in reverse, but in different color combinations with each repetition, for each sleeve. I intentionally did things that can only be seen if the viewer is very near the jacket and looking very closely, using low-contrast color combinations in the little checkerboards and three-color knitting in the concentric circles, with the third color as the edging of the circle and chosen to blend into its color, just to toy with the eye and give a bit of dimensionality to the circles. I don’t know if anyone will ever notice these details, since they don’t show up in photos and disappear at distances more than a foot away, but they were fun to knit and it amuses me to notice them when I’m wearing the jacket. I like to think these hidden details register to the viewer on a subconscious level as vibrations and dimensionality and the perception of complexity within a unified structure.
I knitted first one sleeve, then the other, up to the elbows, while I was thinking about how and how much to increase the volume of the very narrow sleeves to grow into the body of the oversized, boxy jacket shape. My solution was to work an increase into each repeat of the concentric circles motif, starting with the top of the second set of circles in the first sequence. I had 15 circle motifs, so that was one added stitch per motif, which increased the size of the circle in a very subtle way, and I put another set of increases into the tops of those circles, and then again at the top of the last set of circles, adding 45 stitches over 14 rows. You really have to be looking for the make-1 increases to see them. Of course sharp-eyed knitters will see them, as they see everything.
After forming a kind of gusset with those 45 stitches placed around the circumference of the sleeve, I wanted to straighten the curve I had created by adding stitches in an even distribution all around the knitting. I considered various methods, and what I decided to do was to knit some short rows at the start of the downward-headed panel I was going to knit from 40 stitches at the armpit of the sleeve to make the sides of the panel longer than the center of it, thus squaring the curve between the armpit and the side of the body. An inch or so down, in the next bit of plain knitting, I did another pair of short rows, and then it looked as if I had squared the curve. My short rows weren’t the most elegant thing I’ve ever knitted, but they succeeded pretty well at their purpose, and if anyone is scrutinizing my armpit so closely that they can detect the flaws, they deserve what they get.
I knitted the side panel of the first side and got a few inches into the side panel of the second side, and I wanted to take a progress photo for the KAL thread. So I laid the two pieces on my porch table and was all set to press the camera button. And then I realized something very bad had happened. The splits for the pirate cuffs were facing in opposite directions, one facing upward, according to plan, and the other facing downward, which definitely wasn’t the plan. I looked more closely at that second sleeve and saw that the increases for the sleeve shaping were on top, which meant that I had built the downward-facing side panel from the shoulder side of the sleeve rather than the armpit. Well, stuff happens. If you knit, sometimes you get disoriented, you miscount, you misread something, you forget something. It happens, and I assign no dishonor to it. But can you fix it? Better yet, fix it without frogging (knitter jargon for “rip-it, rip-it”)? I have a thread on my group called “Ingenuity After Stupidity”, and here was a new topic for it.
No. I was not going to frog the perfectly good work I had already done in that second side panel. That meant detaching the panel from the sleeve and reattaching it in the right place. There wasn’t really an easy way to do that, since there were short rows that I was going to have to undo, so I resigned myself to unpicking the stitches connecting the short rows in blue with the stranded knitting that followed and tinking (knitter jargon: “tink” is knit spelled backwards, so it means to unknit) the blue stripe back to where I made my fatal error, and knitting back down to the actual armpit. Then I reknitted the short rows and did a Kitchener join with the stranded knitting. That was awkward because I was trying to replicate the shape of the knit stitch into two-color knitting that was going the wrong way, with the stranded yarn looping over the needles rather than tidily attached to the adjoining stitch as it would be facing downward instead of upward. The result wasn’t perfect, but as I just said, if anyone is examining my armpit close enough to see the wonky join, that’s their problem.
When both of my side panels were the length I wanted, I picked up stitches at the sides of the panel I had just knitted and knitted the sleeve stitches that were waiting were for this moment, up and around the shoulder, and picked up stitches on the other side of the panel. I resumed the patterning rota that I had used on the sleeve and began increases at the center line of the shoulder for shoulder shaping, a stitch on either side of the center stitch spaced slightly less than an inch apart, that is, two stitches every eight rows. Then I made pockets, to be hidden behind a slit at each side of the knitting, which was now growing sideways. I knitted the fabric of the pockets on my knitting machine, using one of the preprogrammed fairisle patterns. I’m still not very comfortable with the electronics of my knitting machine, and I had never changed colors in the middle of a fairisle pattern, so a lot of trepidation and learning curve went into this endeavor. I didn’t have the hang of securing the new yarn when I changed colors, so my edge stitches fell off the needles and I had to make repairs each time. I think maybe stationery clamps might be an effective way to weight the new yarn so that it stays on the needles, so I’ll try that next time. For the pockets, I used screamingly supersaturated commercially dyed Wollmeise yarn, as a contrast to my gentle plant-dyed colors, like a subversive secret you can only see on the wrong side of the jacket.
At the end of the patterning sequence I knitted after picking up the stitches from the side panels, I grafted the stitches of the top three-quarters of the pocket to counterpart stitches on the body of the knitting, positioned so that the hem and the pocket edge aligned, and I cast on new stitches in their place, to make up for the stitches that connected with the pocket. From there, I did one last sequence in my rota, and then it was time for something different. So far, with the exception of the improvised double-knitting of the cuffs, the patterning was all repeats of small geometric patterns. Now I wanted to riff on the regular repetitions I had been making by improvising square shapes for about 80 rows, which would take me to the other end of the pocket, and then I would join the end of the pocket to the end of this new improvised stranding section. I had a sequence on one of the sleeves of the larger checkerboard pattern in a golden brown and a lighter camel tan, and I felt that that sequence needed a response on the other side of the jacket in similar colors. That design decision was also based on the practical consideration of the colors available to me, which were lots of browns and yellows. If the colors one gets from plant dyeing were a food pyramid, yellows and browns would be at the base, so I had plenty for creating a yellow gradient and a brown gradient for this stretch of knitting.
Melissa, my friend and local yarn store owner, and I were sitting at the Lovelyarns back table one day while I was approaching this part of the knitting. She told me that my use of color and geometric shapes was reminding her of Friedensreich Hundertwasser, and when she showed me his work, I saw what she meant. Then we looked at the Google Images page for Gustav Klimt, and there was a bit of a painting that had irregularly sized and spaced brown quadrangles on a yellow background, or maybe it was yellow quadrangles on a brown background, and it was so exactly the idea I wanted to achieve that I had to look away from that image immediately, before it paralyzed my ability to do something original. I didn’t want to have a detailed idea of what Klimt had done because I didn’t want to find myself incapable of doing something different. Now I’m looking at the Google Image page, and I don’t know which painting had grabbed me, because it could have been just about any of them. It’s a go-to Klimt leitmotif.
I plopped down irregular alternations of a golden brown and a green-yellow-brown and began randomly knitting without a plan. That’s always uncomfortable for a while, because it looks rather dreadful until some kind of regularity begins to emerge, at which point it’s time to bust up at least some of those repetitions by reversing the colors or building different right-angle formations by extending or contracting the areas of background color and varying the size of and distribution of the quadrangles, and I referred back to my checkerboards of concentric squares in regular repetitions by putting squares within squares in irregular placements. Also there were long sequences of uninterrupted background color columns, although they’re functionally horizontal stripes since the pattern is viewed side-to-side. I built a brown gradient and a yellow gradient, and in general there’s a good bit of contrast between the two. The effect is clearly Klimtian, but I did things with the balance between regularity and the disruption of regularity and the placement of color reversals that he didn’t do.
Then I applied my same-but-different concept to the other side. I used my collection of pale pink-oranges from a variety of plants– fermented hibiscus blossoms, fermented pokeberry that had faded with light and time, madder exhaust overdyed on yellow, amaranth, faded fermented paper mulberry– paired against a blue gradient from black bean dyes and indigo. I couldn’t get a smooth gradient for either set of colors. I was also frustrated for quite a lot of the knitting of this segment by the fact that I couldn’t find a way to break out of the color columns I had established at the beginning in a way that I hadn’t already done on the first side, although now that I examine that first side, I really didn’t do it that much there either. Instead, I varied the size and shape of the quadrangles within the columns. As the segment grew, I found that irregularity within the regular borders of many of the columns was a nice design element, and the fact that I didn’t have smooth gradients gave interesting emphasis to certain parts of the knitting. I also liked the areas of low contrast between the pink and the blue.
After I had knitted both sides as far as the reach of the pockets, I knitted the loops of the cast-on edge of the pockets together with the live stitches of the improvised stranding. These pieces extended up the slope of my shoulders and ended at the nape of my neck. The fronts would be the start of an 8″-wide double-knit shawl collar/button band, but first I had to connect the two sides at the back with a 6″-wide panel that I would join to the live stitches of the sleeve pieces by means of modular knitting, knitted bottom up from a simple cast-on. The end stitches of the panel were knitted together with the live stitches of the stitches of the sleeve/body pieces that were still on their needles, the right sleeve piece on the right edge of the panel, the left sleeve piece on the left side of the panel. For the stranded pattern, to be knitted back-and-forth, I decided to refer back to Johanna’s original jacket, using a traditional northern European flower pattern that was similar to one of the chart’s Johanna had shared with us, which was a stylistic contrast to my Klimt- and Hundertwasser-like squares and circles. But I felt that traditional as this pattern was, its shape was both a square and a circle, which connected the stylized flower shape with the dominant themes of my composition.
As for the colors, it was now green’s turn. If yellow and brown are part of the base of the plant dyer’s food pyramid, the base isn’t complete without green. In fact, green might be the base itself with yellow and green sitting on top of it. Or not. It’s a discussable topic. Anyway, I had enough greens for a very smooth gradient from dark to light and back to dark, without repeating any one green. The contrast colors were various pinks, blues, and yellows, and I made a point of repeating them in high-contrast and low-contrast combinations with the greens. As I’ve said elsewhere, I really like low-contrast color combinations because they force the eye to stop and look closely in order to figure them out, which the eye will do when there are high-contrast color combinations nearby to attract the eye initially. I should note that the low-contrast effect works best when there are high-contrast versions of the same patterning, which teach the viewer’s eye what to look for in decoding the combinations that are harder to see. In my aesthetic, design success is measured in the amount of time the eye spends working out the elements of the piece and the number and direction of the trips the eye takes from one design element to another in its mission to sort out similarities and differences. To this end, the colors of the flowers lead the eye back to previous appearances of these colors in the squares and circles motifs and then asks the brain to sort out how their appearance is modified by other colors and shapes in each context.
Once the center back panel had grown to match the length of the two sides at the nape of the neck, I put all of the stitches– up the right side, around the neck, and down the left side– onto a single needle for the knitting of the final phase, the double-knit shawl collar/button band. The colors would be a green spectrum and a pink spectrum, green as the main color on the facing side and pink as the main color on the reverse side, and I would improvise dots, bubbles, and swirls. I wasn’t so very concerned with getting smooth gradients for either color spectrum, although I’m not sure that I was consciously intending to create stripes within the patterning. Once the striping was there, I was fine with it. I had experimented during January 2020 Swatchathon with using intarsia to introduce different colors as dots on the facing side and patchwork squares on the reverse side, which I planned to do in the last couple of inches of the knitting, in the spaces between and after the buttonholes. So I established the double-knitting and started knitting, and knitting, and knitting. And then I continued to knit. And then I knitted some more. Double-knitting is twice as much knitting as plain knitting, and improvising patterns as one knits means that every stitch is a decision, done twice, in reverse, meaning the knit stitch of the pairs is in one color and the purl side of the pair in the other color. It took four months and a couple of weeks to complete the double-knit shawl collar, but it was four and a half months of the most fun knitting I can possibly do. When I finished I was sad, although wearing the jacket is a joy and just looking at it makes me happy.
I am very pleased with the pink-green combination in the the shawl collar/button band, which is the most conspicuous part of the jacket, and I especially like the colors that appear at the fold of the shawl collar, where the pink side dominates. I like that long sequence of the cochineal pink and the high contrast with light greens, and then the smooth gradient into orange, rust, and muted purple. It’s a good way to frame my face. Someone suggested to me, while the jacket was still in progress, that I had quite the jumble of colors and patterns going on there, but I think that the colors and patterning of the button band/shawl collar pulls everything together, with the help of the intarsia dots. I really never thought that the colors and patterning were at all incoherent because the same-but-different principle provided structure as well as diversity to keep things interesting without being excessively confusing, although there’s no accounting for taste. But I think that the double-knitted part brings it all home.
There’s a lot going on in the button band/shawl collar. The reverse side is as complex as the facing side since double-knitting is a positive-negative form of patterning, so there are a lot of parts of it that don’t show. In some places, like the neck, it’s the pink-dominant side that shows. If it’s buttoned, the intarsia dots on the button side of the band don’t show. I’m glad that the patchwork rectangles of the intarsia dots are visible when the jacket is unbuttoned because I think that’s a fabulous detail, but the facing side is pretty wonderful too, and it doesn’t show much. I’m sad that all these details can’t be absorbed all at once, but I’m glad about it too. I like these little secrets and levels of complexity.
And the cuffs!