Discovering the Universe through Knitting with Elisabetta Matsumoto

Why do we overlook the power of traditional crafts in modern innovation? In this episode, Eleanor Drage speaks with Elisabetta Matsumoto, a physicist and textile expert, about how practices like knitting and smocking can inspire breakthroughs in science, technology, and design.

Their conversation explores Matsumoto’s research on textile techniques that are shaping more sustainable approaches to fashion and opening new possibilities in medical technology, including applications in pelvic surgery. She also reflects on how the logic and flexibility of textiles can inform new ways of thinking about programming and materials. Challenging the idea that innovation must always be high-cost and high-tech, Matsumoto highlights how longstanding craft traditions can lead to more accessible, inclusive, and sustainable solutions. From textile geometry to real-world applications, the episode reveals how these often-overlooked practices are helping reimagine the future of technology.

Elisabetta Matsumoto is a physicist and textile researcher whose work bridges soft matter physics, geometry, and traditional craft practices. Her research focuses on the structure and behavior of textiles, exploring how techniques such as knitting and smocking can inform innovations in materials science, engineering, and design. Through an interdisciplinary approach, Matsumoto examines how craft knowledge can contribute to more sustainable, adaptable, and inclusive technological systems.

Reading list:

How one physicist is unraveling the mathematics of knitting

Elisabetta A. Matsumoto's work

Transcript:

Eleanor Drage (00:53)

In this episode, I talked to the incredible Elisabetta Matsumoto, a professor in the School of Physics at the Georgia Institute of Technology. She uses really cool geometry to understand the mechanical properties of materials, like how smocking patterns can help prevent organ prolapse after pelvic floor surgeries by making the pelvic tissue stronger and stretchier. This is such a great episode on the intersection of feminized crafts, maths, and technology. And I hope you enjoy the show.

Thank you so much for joining us today on The Good Robot. Kerry is not here today, so it's just gonna be me. But I'm very excited to have you. Can you introduce yourself and explain to everyone what brings you to the topic of gender, feminism, and technology?

Elisabetta Matsumoto (01:42)

Hi, Eleanor. It is absolutely a delight to be here. So I am Dr. Sabeta Matsumoto, and I am a physicist who studies textiles.

So I'm really interested in looking at how making textiles from a hand-making point of view sort of lends itself to understanding mechanical properties. So one thing we might ask is why is your t-shirt stretchy? Or can I make a fabric a custom response?

Eleanor Drage (02:16)

Very cool. And we'll be talking about fabrics and craft and all these things that I love so much, a bit later on. First, I wanted to ask you this, this big question that we ask all our guests, which tends to be very difficult to answer. So what is good technology? And perhaps you can respond from the perspective of somebody who works with textiles as a technology. And how can it help us? What does good technology look like, and how can we work towards it using craft, textile, and knitting?

Elisabetta Matsumoto (02:46)

So that's a very hard question. Good technology, I guess, is technology that sort of helps humans out. And so this was something that we obviously needed to clothe ourselves, so things that helped us make fabric, such as looms, started. And now that we have mechanical looms, which I guess are from the Industrial Revolution, they made it into an industry, both good and bad, giving rise to us being able to buy fabric. And now we use sewing machines to put that together, which is amazing. But now we also have problems with cheap labor and a workforce that we're not respecting. So I think the thing with textiles is that things that help us connect to them are also, unfortunately, things that are used to create capitalism. So there's a little bit of good and a little bit of bad in pretty much every technology that we've come up with.

Eleanor Drage (03:48)

That's perfect. I think what I wanted to go on to ask you next is about how knitting has been disavowed in masculinist computer science for such a long time. And I love something that a collaborator of mine said recently, Felienne Hermans, who's a feminist programming language designer.

And she said that knowing a programming language is one of the biggest gifts you can have, next to knitting. I love having those two things together. And knitting and computer science have had this long history. Judy Wajcman, who's a sociologist of computer science, showed us these cartoons from the 80s, where they were trying to get more women into computer science, and these cartoons and contact me if you're listening and you want to have a look at them, say, “If you can knit, can sew”, and they're from Microsyster. So, as a physicist, can you tell us why you love knitting and what geometry can gain from this kind of exploration of what you call the topology of soft materials? So maybe you can explain what a topology of soft materials is, and also, just from your heart, you know what it is about knitting that you really enjoy.

Elisabetta Matsumoto (05:01)

Yeah, so I absolutely love knitting. So I definitely came to knitting well before I came into physics and anything computer science related. But I definitely really resonate with the comment about programming being a really wonderful gift apart from knitting. But I think to myself, I kind of view it more as a type of programming language. But you're not telling a computer what to do using your fingers to tell fabric what to do. So it seems that every stitch you take is really encoding a word or a letter or a piece of code directly into the material. And as you continue to make your piece, you can start to see how those bits of code work together to make an object. 

So, you can start with almost a simple hello world style program, which would be a garter stitch, which is just knitting, and then you flip it, and then you knit, and then you flip it, and you knit. It's the first thing you would learn, but at some point, you might learn how to do increases and decreases, and you can learn how to make things like lace, and for me, making lace was really this very visual connection between the stitches you're making, the code you're writing, and the final product. So you can often feel like you're following a recipe and not really knowing what's going to happen. And then after a while, you're suddenly seeing things coming off your needles that look like leaves or flowers or things that come out of it. And then as you do it, you learn what the logic is of the combinations of stitches that build up a different set of shapes and a different texture of your piece.

Eleanor Drage (06:58)

It's so beautiful. It reminds me of a piece by a designer in Berlin, Kasia. I need to remember her name, but she makes latex designs out of 18th-century lace patterns. And even though, as the difference in material makes, it has a totally different effect, right? But you still have that kind of, there's something of the design there that feels traditional, even in this very untraditional fabric. I guess.

So what does the material matter then? Whether you're making lace or you're creating code, is the difference then a question of the materials that you use? Rather than the way that you're making whatever it is that you're creating?

Elisabetta Matsumoto (07:37)

Yeah, I would argue that the code part of it is what stitches you take and not the material you take them in. So there's something universal, whether I would be using a woolen yarn or if someone is, I don't know, knitting with an LED strip or something like that, you're still basically doing the same things, and you're encoding the same program. But, I guess it's like a Python code or is it a C++ code, maybe that has a different feel to it because it's coming from a different place, but it's still executing the same function in a way. But it's really cool that she's combining those materials. I love mixed media things.

Eleanor Drage (08:22)

It is, it really is. And it shows how you can pull something from one era and make it new. The other thing I quite like about the pieces, but also about knitting in general or sewing in general, is that it's very intuitive, perhaps in a way that coding isn't, although maybe that's arguable. In knitting, knowledge is passed down through generations of hand knitters. That tends to be where the knowledge is, right? Is that true? Do you think knitting is more intuitive than coding?

Elisabetta Matsumoto (08:51)

I think intuition comes from maybe a different place. I mean, if you have learned from your mother or your grandmother or even a male relative, I mean, I guess that's less common, but certainly there's definitely plenty of men who knit. So I don't want to exclude them, but learning from a person has a very different feel to it. So you learn their intuition, and you learn what mistakes they make before you've made those mistakes sometimes. And you learn rules of thumb, or you learn different shortcuts; there's this historical lineage to what you learn. So the things that my mother would have taught me are very similar to things that your mother would have taught you, but they will have been filtered through my mother's experience and your mother's experience. And I think part of that is that it lives in the textiles we make. There's a lot of experience that just comes from being able to touch them and having crafted something. There's a slight chance I don't know, people really, really like things that they've touched or can touch. And also, things that they've made are really important. So I think in some sense, like craft in general has a really intuitive aspect on one hand, other people might feel the same way about coding, that it's very neat. It lives in a couple of documents on your computer. It's just a different aesthetic. And there are lots and lots of different ways of thinking about coding languages. So you can think of the ones where you type everything out in some language, or there are the ones where you have little boxes, and you draw wires in between them. They do the same thing, but they feel very different.

Personally, I can't stand the ones that you pull wires in between. I need to have the text to document. But that's just a personal choice. So someone might like knitting, someone else might like crochet, or lace making, or pottery, or something like that.

Eleanor Drage (11:07)

And these different modes of creating, I guess, appeal to different kinds of people, different kinds of brains, or potentially, depending on whether you've been encouraged to do craft or encouraged to do code, people express that desire to create, to pattern the world in a particular way according to what they've been encouraged to do. I'd love for you to tell us a little bit more about the way that your work influences manufacturing. And it might seem incredible to think that craft and manufacture, through experimental crafts, are improving the way that we do manufacturing. Can you tell us a bit more about that?

Elisabetta Matsumoto (11:50)

Yeah, so manufacturing is such a complicated thing to think about. I mean, we have supply chains starting from people picking cotton or people taking petroleum, refining chemicals, and turning them into fabric. And then those get spun into thread and woven or knitted into fabric, and then the fabric gets chopped up and sewn into garments. And then the garments get sold to us in the industry. And I'm sure I've skipped about a thousand steps in that process. So I think if we're thinking about manufacturing, we can kind of think about almost any point of view. So the point of view that we started with, and definitely aren't, hopefully not going to end there, but we started by thinking about how people would hand-knit something like a sweater. And so people often will make different types of stitches in different places for reasons of having different mechanical properties. So you might have ribbing near the cuffs and collar. Some places you might use what's called seed stitch to make it a bit more isotropic, so it doesn't stretch too much in only one direction. And these are things that people would have done from an intuitive point of view. But also because they start with one type of yarn and knit the whole garment from one type of yarn. And in industry, often, instead of using the combination of stitches, what we would call a topology, they tend to combine lots of different fibers together. So you might have elastane, which is what spandex and Lycra are made from. You might have polyester, you might have cotton and wool, you have so many different things, and they blend them in very specific proportions, but keep the stitch pattern, the topology of it all, the same. So in the code analogy, it's maybe not quite the Hello World, but it's maybe one more program after that that's giving the fabric its bulk structure. So we're interested in saying, well, instead of those specific yarns, what can we do on the stitch pattern side? Can we emulate the results that you are going for, but by modulating what the stitches are, rather than the material itself? So that's what we're interested in from a manufacturing point of view. And that leads to less waste from petrochemicals, but also, I think, a thing that's really fun to think about from textiles, particularly knits, is that you can make garments, like the flat pieces of garments, with zero waste. You can knit out just the shape of the pattern pieces you want. Or some machines, if you've got a really fancy machine, you can just take the whole garment and knit out the entire garment. And so that means that there's no waste from the patterning process, from cutting and sewing things together as well.

Eleanor Drage (15:03)

So, is it that you're using one bit of fabric, but because you're sewing it in a particular way, you don't need to use it, you're just using that one bit of fabric? I'm not a non-knitter; I don't even have to explain.

Elisabetta Matsumoto (15:15)

Yeah. So, maybe I could say if I were making, if I were sewing myself a cardigan, say, I would have made, I guess, five parts. So I would have a back shape, left front, a right front, and two sleeves. And when you're doing the knitting, you have the shape of each of those pieces already built into the object. So would have the place where my armhole starts, and the sleeve joins in. I'd have the place where the top of the sleeve meets my shoulder. And these are all geometric parts of this that would exist only on their own. Whereas if I were sewing, I would take a large flat piece of fabric and then cut out those shapes. And so there's only so well you can smush those shapes together into a piece of fabric, and there's always going to be waste left over.

Eleanor Drage (16:12)

Not only are you transforming sustainable clothing, but you're also making waves in medicine. And I know that you're doing incredible collaborations, for example, using smocking to reenact surgeries for pelvic organ prolapse. Can you tell us what that is and what you're doing to resolve it?

Elisabetta Matsumoto (16:32)

Yeah, so this is a project that we are absolutely so excited to work on right now. So we were approached by some pelvic surgeons from Emory. So, Dr. Robert Kelley and Dr. Natalia Calzada Jorge. And so they deal with reconstructive surgery on pelvic floor dysfunction. And this is such a hugely common problem, something between 25% and 50% of people with uteruses will experience this at some point in their life, most commonly either after childbirth or with age. particularly. It has some genetic components to it, but when pelvic floor dysfunction becomes bad, you can end up with some of your internal organs protruding through into your vagina or your bowel.

And this is incredibly uncomfortable. It makes daily life really, really, really challenging for many people. And so different surgeries are designed to strengthen the pelvic floor and the walls in between those organs. And I guess there's a tradition that involves making almost a billfold and sewing it up. That was invented over a hundred years ago. And it definitely improves people's quality of life, but something about the surgery makes it almost heal too well. So it heals to the point of going back to the tissue it was before. In the 70s or 80s, there were manufacturers of products that were solving or doing meshes for hernias.

And they were saying, “What is pelvic organ prolapse but a hernia of the vagina? Let's put this mesh there.” And that, of course, got rushed through a lot of regulatory services. A huge number of surgeries using a vaginal mesh have happened. And a vagina is not the same as your lower abdomen. I hope I do not need to mention this to any of your listeners, but it had a lot of problems related to scar tissue forming and loss of muscle tone, which led to several lawsuits. So this is a surgery that I guess can happen in certain cases, but very problematic. So the idea the surgeons had was coming up with a new technique that would enable us to create a stronger pelvic floor, but just out of your own tissue, using a natural healing process.

And so we took inspiration from smocking. So this is an embroidery technique that has existed at least to the Middle Ages. There are pictures of people using smocked garments, and smocking basically is an embroidery technique that is used to give some amount of elasticity to woven fabric. So woven fabrics, particularly ones made out of cotton, linen, wool, or natural fibers, do not have a lot of stretch to them. But by making stitches in a specific pattern, you can allow those fabrics to have some stretch. And so this was used in any number of garments, from a tunic that might have the cuffs. And the cuffs around the wrists and neckline that were smocked, which would definitely make it easier to use a garment as someone is growing. So when I was a child, my mom made me a smocked dress, which I think lasted me from when I was about two till when I was about eight, just because of the amount of elasticity it has. So the idea was to repurpose some of the ideas from this embroidery technique and try to see about using them to improve the outcomes with pelvic surgery. So we're nowhere near there, but we're exploring, trying to see how much additional strength we can impart into a fabric by varying the smocking pattern.

Eleanor Drage (21:03)

So is the idea that this mocking is a fabric that's used internally, or is it?

Elisabetta Matsumoto (21:09)

Now we actually want to take the stitches directly into the pelvic fascia. So we're using the same smocking like stitches used in the way sutures would normally take place. So you would create the new stretch out of the original pelvic tissue. So you'd just be encouraging it to heal in a way that has a little bit more mechanical support than it did before.

Eleanor Drage (21:38)

That's fascinating. So you're actually through with changing the way the stitching is done, changing the properties of the organic material that we have.

Elisabetta Matsumoto (21:47)

That's the goal. We don't know if it's going to work yet, but I'm so, so excited to think of this historical embroidery thing, being able to help modern women. I just kind of love the women helping women through women's technology. But will, I will also say that the team of five people currently has two people who identify as women and three people who identify as men. So we are not a monolithic team of scientists.

Eleanor Drage (22:21)

That's okay, men can do smocking too, it's allowed. It's also Smocking, one of those fabrics that is coming back into fashion. It's interesting how it's not just the idea that knitting is a woman's art, but also the different materials themselves that come in and out of fashion, that seem a bit antiquated. I think smocking was. Also for pajamas, nightwear, and women's undergarments.

Elisabetta Matsumoto (22:48)

I'm not 100 % sure about it. I guess I've seen some paintings and engravings of it from the Middle Ages, but I actually don't know the recent modern history of it. I should probably read up on that.

Eleanor Drage (23:04)

I'm pretty sure that I had a smocked dress that had that kind of stretchy thing in the middle of the chest.

Elisabetta Matsumoto (23:09)

Yes, it's very common in anything. It's like the one-size-fits-all things that you can often buy at craft fairs and things like that. Because it will stretch to fit any chest size.

Eleanor Drage (23:15)

For girls, which is very sustainable and actually, just looking on the internet now, smocked, I don't know what to call it. I mean, it is a smock, it's not a dress, and they are seen on men of all varieties. And if you have a look online, it's a garment that you would associate with women, but definitely is something that men have worn for a long time, and interestingly, it became feminized. I guess it's the kind of outfit you might not imagine a man wearing today. Isn't it? Isn't it? I was watching, I was reading yesterday about how in the 15th century pink was originally worn by men, it came from bois de Brazil

Elisabetta Matsumoto (23:50)

Fashion is fascinating.

Eleanor Drage (24:11)

And it was definitely a male colour, and it wasn't until much later that it became feminised. And now we think of pink and blue as these, as has always been a binary that associates itself with gender, right?

Elisabetta Matsumoto (24:27)

Yellow and green are the best baby colors because then the baby can decide once they get old enough.

Eleanor Drage (24:33)

Right, exactly. We should. I don't know. We should. I'll be doing that. I'll be doing that. I wanted to just end by asking you about ambiguity because we're going back to what you said that knitting is a kind of coding. It's not just analogous to code. It is a form of coding. And it allows for a lot of ambiguity and loose interpretation, unlike a lot of binary programming languages so it's yes or no and there's not much there's not much ambiguity in programming languages. So, one of the things that Felienne, whom I mentioned, is really concerned about is how we can create languages that allow the people who write them not to have to decide at the moment that they're writing it, but to leave doubt within the programming language.

And is there a way that we can look to knitting patterns to inject more doubt and ambiguity and context-specific adaptation, to adapt programming languages for different uses? How can we look to knitting to create more of that?

Elisabetta Matsumoto (25:37)

So I love that idea. Thinking about it is really fascinating. It would be really interesting to see ways of making computers flexible in interpreting the same way that humans and animals can do about things in different think fabrics, and I know knitting better, so I'll use knitting as the analogy, and have a fantastic way of doing just this. I mean, for knits, it's probably primarily mechanical, although there are plenty of people who will knit with wires and create circuits just built into the garments. But from a purely mechanical point of view, there are lots of patterns that would have, I guess, we would call them metastable states. So it can have one confirmation and one context and be perfectly happy, and then another confirmation and another context and also be perfectly happy. And then being able to get in between those requires some process of mechanically moving the fabric around. So this could be something like putting on and taking off a garment. You might be able to have it encode something. I mean, even, I guess, if we put something on and take it off and it's inside out, that's already doing some amount of recording something about the two states of inside and right side out.

For the context-specific programming, if you'd want it to return information. So I guess an example of this would be something like a strain sensor. So you can imagine that once it reaches an amount of extension, it triggers, I don't know, an electronic signal that gets recorded, I guess you could imagine using this for something like a compression stocking. After a certain signal turns off, and you've got enough tension on your stockings, but if it slides down, you get a signal that says, " Pull up your stockings.” So you can imagine something like that being, I mean, that's not exactly context specific because that's designed specifically for stockings, but maybe there would be a garment where you could use that for stockings, but you could also use it to, I don't know, measure size of a pregnant woman's bump or something like that and use that as a way to signal, I don't know, something about garment change or needing to see a doctor or whatnot.

Eleanor Drage (28:20)

Very cool. What's nice about that is it's a low technology in inverted commas. It doesn't require power. It can be done by anyone. And hence it's probably not used. We know that a lot of the things that the reasons that AI has become masculinized are because it's deemed to be hard. These other technologies, like knitting, are deemed to be easy, or you could do it anyway; you can do it on the go in a way that doesn't require a date, a huge data centre, or loads of power, loads of energy.

Elisabetta Matsumoto (28:48)

I would challenge Elon Musk to learn how to knit.

Eleanor Drage (28:52)

Right, exactly. But sustainability is forcing us, requiring us to find other ways of building technologies. And some of the most exciting things that I've seen in innovations in AI have been innovations for low-resource scenarios like microscopes that can be put together using 3D printed bits of materials and can be used all over the world equally, solar-powered sensors, and that kind of thing. And I think that this ingenuity to me is really exciting. Those are the innovations that I'm really thrilled about, not just things that require digging up vast waves of lithium and quartz.

Elisabetta Matsumoto (29:32)

Yeah, I think it's really fascinating also that thinking about the lo-fi tech is also very democratizing because it's not just that tech is hard, but it's also expensive, which makes it an elitist structure as well as a masculinizing structure. So if we can have microscopes that we can take anywhere. I think a lot of what's overlooked is that every indigenous culture has come up with its own forms of textiles and crafts, and they're very, very smart in terms of building things for both form and function, and we have basically come in as a Western society and been like this, yes, that, no. But I think we've probably lost just so many solutions to problems by not listening to the technology that people have developed for centuries in their own communities all around the world.

Eleanor Drage (30:40)

Thank you so much for joining me today and talking about all these incredibly interesting things. We will have a reading list to go with the show, and do check it out. It will have some things that we've talked about today. But thank you so much, Sabetta

Elisabetta Matsumoto (30:54)

Thank you, Eleanor. It's been an absolute delight.