Illusion Knitting Turns Angles and Stitches Into Hidden Art

facebooktwitterreddit

Seen close-up and head-on, an illusion knit wall hanging might look like a mundane collection of stripes gently snagged by cat claws. But step a few paces to one side, and an image emerges. It can be simple: a checkerboard or a snail spiral. Or it can be complicated: a landscape view of the Great Pyramid of Giza, a portrait of Marilyn Monroe, or Vermeer's Girl with a Pearl Earring.

Whatever the image, the subtle trick on your eye that allows you to finally see this “illusion” isn’t much of a trick at all. It’s just knitting.

Knitting works like this: You build up a swatch of it by forming a row of yarn loops on a knitting needle, then pulling more loops through them, one by one, with a second needle. Each loop shows its rounded top on one side of your swatch, and its beginning-and-end-strand bottom on the other. A whole row of those rounded tops makes a puffy ridge; that's called a garter-stitch row. A whole line of those bottoms lies flat; that's called a stocking-stitch row. So, even though that seemingly cat-scratched wall hanging looks as planar as paper, because of those garter- and stocking-stitch rows, its surface is actually 3D. That's how you create illusion knitting.

As far as anyone knows, illusion knitting originated with a Japanese knitting teacher named Mieko Yano. In the early 1980s, she moved to Sweden to get married; packed along with all her earthly possessions was a slim booklet that explained how to make what she called “magic patterns.” At some point, the booklet was translated into Danish, which is how it came to the attention of another knitting teacher named Vivian Høxbro, who went on to publish her own book about the technique, which she called Shadow Knitting. Her designs were simple, but a slew of people have been experimenting with the parameters of illusion (or shadow) knitting ever since.

The simplest kind of illusion knitting uses one color of yarn. From the front, you see a swath of, say, green. From the side, you see an alternating checkerboard of green squares. Or take the knit below, which appears to be a multicolored grid straight-on but from an angle reveals circles within the grid.

How does illusion knitting show you two different images? From the side, unlike from the front, your eye catches on the raised garter-stitch ridges that delineate the pattern, and it glosses over the stocking-stitch valleys. Helping this along, a rough surface—the raised garter-stitch ridge, in this case—“tends to look darker than a smooth surface,” according to Derin Sherman, a physics professor at Cornell College in Iowa who studies optical illusions, among other topics. Sherman tells mental_floss, “That’s because, while light often gets caught in the nooks and crannies of a rough surface, it just bounces off a smooth surface”—our flat, stocking-stitch valley.

The kind of illusion knitting that gets you to Marilyn uses two colors of yarn: one light, one dark, in alternating stripes. The most basic explanation of how this works is that the light-colored yarn accentuates stocking-stitch valleys, pushing them into the background; the dark-colored yarn accentuates garter-stitch ridges, pulling them into the foreground.

Sherman says a good way to visualize how to create this effect is to imagine strips of clay, both dark and light, laid out on a table. “Where you want the picture to look dark, raise the dark clay stripe to create a small dark hill, and lower the white stripe to create a small light valley,” he advises. “Looking straight down shows dark and white stripes, but from the sides the hills stand out, so the patterns appear.” This bit of technique alone isn’t quite enough to make Marilyn pop out of some yarn, but it more than gets you started.

British math teacher Steve Plummer—who uses knitting and crochet to explain math concepts—creates complex images, including Charlie Chaplin in the style of Warhol, a tiger head, Rossetti's Sybilla Palmifera, and a 3D fractal Menger sponge, seen below. (All of the animations in this story come from Woolly Thoughts, the website of Plummer and fellow math teacher/knitter Pat Ashforth.)

The knitting itself isn’t complicated; even beginner knitters can do it. But any pattern first has to be made into a chart. That’s where the challenge lies. Plummer explains to mental_floss, “The smallest detail I want to show must be at least one stitch across. This determines the scale of the completed piece.” Once he’s figured that out, Plummer places a grid over his entire drawn image. “Each square on the grid represents one stitch, and each row of squares represents one row of knitting,” he says. He then decides which areas on the image will be dark or light, and colors the grid in accordingly. On average, it takes him 100 hours to chart one piece of illusion knitting.

To date, the most impressive use of illusion knitting might be by Austrian artist Tanja Boukal, who’s exhibited strikingly realistic portraits based on gritty newspaper photos of armed women prepared for combat. Is this as far as illusion knitting can go?

Sherman, who is not a knitter himself, sees the potential for more. He suggests the underlying formula could be enhanced by using different colors to shade ridges on either of their sides, so you’d see different images depending on whether you viewed the work from the left or right. But, he admits, “It would be hard for a human to knit.”

Knitted gauntlet thrown?

All animations courtesy of Steve Plummer and Pat Ashforth