How Upside-Down Models Revolutionized Architecture

Some buildings are designed upside down. For buildings like this, the architect crafts a model, maybe of the whole building, or maybe just a piece of it, and then the model is suspended by its base from above. The architect works on it like this, tweaking things until they're satisfied, when they flip it over and build everything right side up. This convoluted process has created some of the most incredibly beautiful and sophisticated buildings in the world, and it's been happening for hundreds of years. Sometimes it's the only way to solve a very particular kind of problem. The architect Christopher Wren ran into such a problem while working on the most significant building of his career, 10 years after he'd started construction. By 1685, the foundation and the ground level of St. Paul's Cathedral in London were already complete. It was common practice during those days for an architect to continue designing such a large and complicated building even while it was being built, and Wren made all kinds of models during that time to work out the design. Wren's initial design, like this one, was actually rejected for being too radical. So the one that was under construction was a kind of a compromise that Wren wasn't exactly happy with. So Wren went back to the drawing board and tinkered with and sketched out a new design that sprang from the parts that had already been built, while replacing everything above. He figured that with this same foundation, that he could build the tallest building in London. This meant capping everything with the highest dome ever built. The king approved the new design from Wren's sketches, but the problem was that Wren had no idea how to make a dome like this. It would have to be 365 feet or 111 meters off of the ground, and this impressive height was the driving factor for the design. For a traditional dome design to reach this height, it would have to span an opening that's 160 feet or 49 meters wide. But this made it much too heavy for the walls below. Existing techniques for building this just couldn't work. So he consulted his friend, Robert Hooke, who claimed that he had a new approach. Design it upside down. Hooke was a engineer that had already been fascinated by studying how materials behave when they hang from above, or what's called funiculars. He used chains, springs, and weights, and set them up in various configurations and observed how each element affected the other. He noted how moving a weight would result in the chain taking on a different shape, just by virtue of how everything responds naturally. I used this kit by MOLA that they sent me to recreate one of the experiments here. They're not a sponsor or anything, but these kits are really great for exploring structural behaviors and things. Anyway, chains are strong only when you're pulling on them. They're able to resist with incredible strength. This is described structurally as working in tension. They resist pulling, but not pushing. That's what's called compression. Tension and compression are opposites from one another. And Hooke's genius was that he realized that the chain in his experiments was calculating the perfect shape for it to remain in tension, since that's all it can do. Today, we might call this process something like material computation or self-organization. That's when material and physical behaviors are simulating real-time responses to design operations. The system becomes this kind of calculator that spits out the right shape for a given system. And the leap that makes this useful for Wren is that Hooke explained domes as the exact opposite of the chains. His famous line was, as hangs the flexible line, so but Domes are the exact opposite of the chain. Because they're built with stacked masonry or stone, they only work in compression. If you pull on it, the individual stones will just come apart. So if you flip the shape of Hooke's chain experiments upside down, the forces flip, and this shape is the perfect compression system. The shape of this chain doesn't look like a traditional dome, which is just a half of a sphere or circular. Instead, the shape observed by Hooke is narrower at the bottom and sort of flares and tapers outward toward the top. That explains why Wren's earlier designs, why they had to be so thick. Unless this shape could fit inside of the thickness of those rounded versions, it would collapse. Only the material along this line is really doing the work of carrying the weight. The rest is just almost waste. Hooke and Wren traced the shape of the chain model onto paper, flipped it, spun the shape around 360 degrees, and used it for the dome. This allowed them to build taller, over a narrower opening, with less material than anyone thought possible. But they didn't get it exactly right. Years later, mathematicians were able to determine, with more precision, the formulas for finding this ideal shape. Luckily, Wren was close enough to be able to make it work, and it still stands today. But their math shows that this shape should be a catenary curve, as it's called. Today, this shape is pretty well known, and it's designed into all sorts of incredible constructions, and we don't have to design it by working on a model upside down. We have the math to be able to draw the correct geometry and be able to incorporate it into the design directly through drawing. But there was this gap of time, when the principle could only be reproduced and determined physically in mock-ups that were translated into building instructions. And this had to be done with a model upside down. The most famous upside down model sits in the Sagrada Familia Museum in Barcelona, Spain. And like Hooke's experiments, this one too uses strings and weights to find the most ideal forms for an efficient, structural solution for a building. A building that would be 10 times larger than the size of this model. The model was made by the architect Antonio Gaudí, while creating the design for the Church of Colonia Guell. Gaudí spent about 10 years of his life making this model, almost exactly 200 years after Wren and Hooke's work. And just like today, catenary forms were well known by the time that Gaudí wanted to be able to use them. They were easy enough to calculate for anyone who had studied architecture or engineering or geometry. In Spain, where Gaudí lived, they were often deployed by the military, and then that shape extruded into making warehouses to store gunpowder. All these structures were made without upside down models that preceded them. So why did Gaudí make them? Gaudí was famously enamored with what some call organic forms. These kind of flowing shapes of seamless material where elements gradually transition into some other element with a complex fluid curvature. And for this church, this meant incorporating multiple catenary forms that connected in really complex ways. That's why it's called a polyfunicular model. And it allowed Gaudí to work on the design with the assurance that the building would stand up in the end. Gaudí, he also just didn't really like making drawings of his designs before construction. He didn't really trust them, and he preferred to make models instead. This allowed him to visualize the space more fully beforehand. But this upside down model was a precursor to the other kinds of models that he would use to show the construction crew. So while he was working, he would photograph it, he would print the photo, and then he would use it as a guide in the right side up models that he would make out of white plaster. And then it's in these models that he would gain a more full understanding and command of the overall space. And he would add details with the assurance that the whole thing would stand up structurally in the end. The Sagrada Familia, the building that this model is in, has similar forms, and it's really Gaudí's masterwork. He was designing at the same time as the church we were just looking at. In fact, he devoted the entire rest of his career to this, working on it until his death in 1926. Like Wren's process, where the construction and design were happening kind of at the same time, that was true too for this Barcelona Cathedral. During that time, Gaudí created a full, comprehensive model, all out of plaster, that dictated every single part of the final design. And it was meant for construction crews to follow as instructions for creating the real building. The models were the only thing for anyone to be able to look at and understand the architect's intentions once he was no longer there. So construction went smoothly for 10 years after Gaudí's death until 1936. And that's when the models were destroyed as a political statement by an anarchist group during the Spanish Civil War. Without the evidence that Gaudí left behind, construction on the building, it had to stop. And the building sat in its half-finished state for decades. The models were eventually pieced back together as could, but they were basically useless. There were these huge pieces that were missing, and construction crews couldn't really build off of a model anymore, at least not even a fragmented one. They needed actual construction document drawings or blueprints. But no one could recreate Gaudí's complicated, organic design. Traditionally trained architects just didn't have the tools or the know-how to be able to complete the forms. That is, until someone imagined the models upside down. In 1979, a recent architecture school graduate realized that Gaudí was essentially thinking of the building like a landscape. The complicated ceiling is just like an upside-down mountain range, you know, with hills and valleys. And how do you draw a landscape? Well, you use topography lines, these imaginary slices that cut through the ground that allow you to describe the changes in height. And then the three-dimensional hills can be shown in two-dimensional lines on paper. And that's how you can understand the forms of Sagrada Familia. Burry translated and drew the construction documents for the entire design, drawing it like mountains and a landscape. The building is due to be completed in 2026, 100 years after Gaudí's death. Despite the fact that the cathedral appears smooth and solid, structurally, it really works a lot like those chain models. Structural forces collect into specific lines that take them down to the ground. The chains are basically translated into stone structural ribbing. The surfaces between the ribs, they're filled in, but they're not really acting structurally. They don't help the ribs out at all. If that entire surface were part of the structure, it would behave more like a shell. And that's where forces flow evenly across a surface. And as you can imagine, that is much more complicated form and set of equations to be able to calculate. But when done, the resulting thin shells are some of the most impressive spaces and feats of engineering that you can imagine. The same year of Gaudí's death in 1926, an engineer was born that discovered a way to design perfectly efficient shell structures. And of course, it involved building models upside down. For years, Heinz Isler was designing and making thin shell structures, along with notable folks like Luigi Nervi and Felix Candela. All these engineers were using complex mathematics to be able to arrive at solutions, which seem almost impossible, spanning dozens, if not hundreds of feet with only inches of material. But these were limited to simple, known and mathematically predictable shapes. But Isler was channeling Robert Hooke's experimental mindset. And he figured that hanging fabric is the same for shells as hanging strings are for arches. So he hung some burlap from a wooden frame. And to preserve his structurally perfect sculpture, he took it outside into the cold Swiss winter and sprayed the fabric with water. In a few hours, his fabric was coated in only one millimeter thick of ice. And that incredibly thin membrane that was made spanned three meters across without any internal support. And he went on to test all kinds of starting points to observe the results. He also built a few, mostly as canopies for indoor tennis. In some ways, this process of using material properties of models while upside down is like a structural failsafe. While working on it like this, the project inhabits this other universe with these special rules where no matter what you do, the outcome will be structurally efficient and plausible to build. The architect and the designer working this way can do what they want with the assurance that it'll all work out in the end. But it also makes it hard to understand the space that you're creating. I mean, it's upside down. I'm not. And architects like Antonio Gaudí, he didn't really use the upside down model to be able to create forms. He kind of mistrusted the chain models. Instead, those plaster versions are what really mattered to him. While the upside down chain versions were just there to ensure the rigid structural solution. Also, Isler, he wasn't an architect. He was an engineer. While Christopher Wren and Robert Hooke together formed an architecture engineer duo. That's why I think examples like Isler's don't quite have the same level of spatial sophistication overall in the final buildings, even though they're incredible engineering achievements. All these examples use this flipping process when the task was just too complicated for technology at the time to handle, or the tools for calculating everything were just too cumbersome to be able to make decisions quickly for the architect. It demonstrates the power of what's possible when architects and engineers work closely together. It also shows how we can look at buildings in truly isolated and fragmented ways, separate from how they're eventually going to be built. And the fact that you can work on it like this while it's upside down, I think is a cool step that might lead to completely new and super efficient solutions. And that's always something worth exploring. The magic of turning things upside down doesn't just work in architecture. It also applies to fields like industrial design. That's exactly what Henson Shaving does with their razors and how they've found a way to provide a better, more satisfying shaving experience that's less expensive in the long run. Most razor companies lure you in with a cheap handle and then make their money by hooking you on proprietary disposable cartridge heads. They charge you like two bucks a piece for each one just because it has five blades and a slimy lubricant strip. The next thing you know, you've spent hundreds of dollars trying to keep a decent shave. Turns out one blade is actually better if you mount it correctly. So Henson precision engineers and crafts handles and heads that you buy once. This gorgeous hunk of aluminum holds a single blade along its length, exposing only 27 microns. This means there's no way for the blade to flex, bend, or cause irritation. Henson's expertise, it comes from making parts for satellites and space probes. They've engineered stuff that's on Mars right now. So when we say that 30 degrees is a perfect angle for a blade to shave the skin, you can trust them. And these blades, they only cost 10 cents a piece. You benefit with this upside down business model by investing in a well-crafted personal item and leaving that disposable plastic crap at the grocery store. But because you're here, you will get 100 blades for free. Head to hensonshaving.com, pick whichever razor you like, and add 100 pack of blades to your order. Then when you're checking out, enter the promo code stewardhicks, all one word, and the prices of the blades will be deducted from your order. You probably won't have to spend another dime on shaving for at least a year. And I've loved my razor for a few years at this point, and it still looks and feels fantastic. So treat yourself or someone else with a gorgeous razor and get those free blades with promo code stewardhicks. Links down below. Enjoy, and thanks for watching.