Hemstitching as Surface Language

I first came across the hemstitcher at RNC, a factory owned by Ramdat Harihar. I was eighteen, still at Parsons. Ramdat showed me work he was doing on a hemstitcher for Oscar de la Renta: spirals and spirals of beautifully hemstitched floral print chiffon, which was later going to be cut to be turned into picot finished ruffles.

He picked up a panel and began cutting it, and before my eyes the transformation began. The hemstitching wasn't the finish. It was setting up the finish. Once Ramdat started cutting along those perforated lines, the picot edge just appeared, clean and precise. The whole thing had been built into the fabric before the shears even touched it.

That moment stayed with me. What I'd seen wasn't a finishing technique, not really. It was a construction language hiding inside a finishing operation, and once you see it that way you can't unsee it.

The Singer 119W1 is a double-needle, double-bobbin industrial hemstitching machine, first manufactured in the 1930s and the rarest of the Singer line of double-needle hemstitchers. Three assemblies work in unison: two vibrating needles forming an over-edge hemstitch on each side of the opening, two gear-driven rotary sewing hooks, and one piercer that perforates the fabric between the two needle lines. It runs at 2,250 stitches per minute. Piercer sizes range from 1 through 7, producing hole diameters from roughly 1mm to 5mm.

What's happening mechanically is almost counterintuitive. The piercer punches through the fabric first, and then the twin needles stitch simultaneously on either side of that perforation. The result is a line of open holes bordered by locked stitches, fabric that is structurally sound despite the material that's been removed from it. The machine doesn't add to the surface. It subtracts from it, and the pattern is defined entirely by what's been taken away.

Most surface decoration in textiles works through addition. Embroidery adds thread, applique adds fabric, print adds pigment, beading adds weight and dimension. In every case the base fabric is the recipient, the thing being decorated. Hemstitching inverts that relationship entirely. The pattern is carved into the fabric. The holes are the design. The remaining threads are the frame. The visual identity of the surface comes not from what has been placed on it but from what has been taken away.

This puts hemstitching closer to intaglio printmaking, where the image is cut into the plate, than to screen printing, where it's laid on top. It also shares structural logic with architectural fenestration, the placement of openings in a wall where the void defines the rhythm and the solid material is just the frame. Mashrabiya screens in Islamic architecture work this way, and so do perforated building facades. The pattern is the absence, and that's a fundamentally different starting point from almost anything else in garment construction.

Hemstitching is a form of drawn thread work, and it is among the oldest textile techniques we have documented. Examples of drawn thread embroidery date to the 1400s, with possible origins in Egyptian and Coptic tombs reaching much further back. By the 12th century the technique was known as Opus Tiratum and Punto Tirato, from the Italian tirare, to pull or draw. The original function was purely utilitarian: withdraw a thread from a piece of linen near the fabric edge, turn up the remaining material and catch it in the hemstitching, and you've accomplished hemming and decoration simultaneously in a single operation. As ornamentation became culturally important the technique grew more elaborate, moving from functional edge finishing to complex interior patterning, but that dual utility never really went away.

Mechanization arrived in 1893 when Karl Friedrich Gegauf invented the first hemstitching machine, automating a bundling process that had been done by hand for centuries. The Singer 119W1, manufactured in the 1930s, represents the pinnacle of that mechanical lineage. And then there was a revival in the 1970s and 1980s through the heirloom sewing movement, which brought fine hemstitch techniques back on delicate fabrics like batiste and lace, drawing on 19th-century European traditions. The point being that people keep coming back to this, century after century, because the basic proposition is so good.

What draws me to hemstitching is that it can be a constructional stitch, a decorative stitch, and a finishing stitch, and in all three cases it produces a single-layer output without the bulk of seam allowances and turned edges. That versatility is rare in garment construction.

As construction, the hemstitch can join two fabric panels along a perforated line, creating a seam that is simultaneously decorative and structural. No seam allowance, no pressing, no bulk. As decoration, the open-work pattern system operates across the surface of the fabric, transforming its opacity, its hand, its visual rhythm. As finishing, it's the picot edge that Ramdat showed me at RNC, where the hemstitched line becomes the cutting guide for a clean, sealed edge. Three different jobs, one pass through the machine.

The technique and the machine are as old as time, and yet the capabilities feel futuristic. A single pass through the 119W1 can accomplish what otherwise requires multiple operations, multiple machines, and multiple layers of material. The resulting design is flat, clean, and structurally complete. That's hard to beat.

The 119W1 pierces and binds in the same pass, punching holes through the fabric while simultaneously stitching a border around each opening to keep it from unraveling. The line it draws is an absence and the stitch that borders the absence is the containment. It's a drawing tool that works by subtraction.

By varying piercer size, stitch density, line spacing, and path geometry you can produce pattern systems that range from delicate single-line tracery to dense grids of perforations that transform the fabric entirely. Change the piercer from a size 1 to a size 7 and the character of the line shifts completely. Tighten the spacing between passes and the fabric starts losing density, gaining translucency. Run lines in parallel and the surface becomes striped with absence. Cross them and you get a field of controlled transparency where the structural behavior of the fabric changes right along with the visual. The parameters are the vocabulary, the patterns are the sentences, and the fabric that comes out the other side is the text.

Swatch experiments are in progress. I'm testing different piercer sizes across a range of fabrics to see how the material responds, where it holds, where it gives, and what the 119W1 can do that I haven't tried yet. Findings will be documented here as the work continues.

In the zero-gravity textile research, the question is what happens when you remove an invisible force from the garment equation. Gravity has been co-authoring every silhouette ever made and its absence reveals dependencies we never knew existed. In hemstitching, the question runs parallel: what happens when you remove material from the fabric surface? The fabric has been relying on its own density and continuity to hold its shape, its opacity, its hand. Subtraction reveals what those qualities actually depend on.

Both bodies of work are really investigating the same thing, which is what absence reveals. And both lead to the same structural insight: that the design opportunity lives not in what is present but in the controlled removal of what was assumed to be necessary.

There's a practical overlap too. In weightlessness, bulk behaves unpredictably and every unnecessary layer of fabric becomes a liability. A stitch that can join, finish, and decorate in a single pass, producing a flat single-layer output without seam allowances, is exactly the kind of construction solution that zero-gravity garments demand. The hemstitch got there centuries ago.

Can hemstitch patterns be designed to change the drape behavior of a panel? A grid of perforations reduces fabric weight in a localized area, and it's worth asking whether that creates directional movement, where the denser region pulls the perforated region in a specific direction. Can density gradients produce controlled transparency that reads differently at different distances, the fabric shifting from opaque to translucent depending on the density of subtraction?

And then the big one: what happens when hemstitched fabric enters zero gravity? The perforations reduce the fabric's mass without reducing its surface area, and in weightlessness that mass-to-area ratio determines how the fabric responds to air currents, body movement, and inertial drift. A hemstitched panel may float differently than an intact one. The subtraction changes not just the surface but the physics. I don't have the answer yet, but that's the question I keep coming back to.