🔢 mathematics

Hawksmoor's Ceiling, Mercator's Projection and the Roman Pantheon

This paper proposes that the coffered grids of the Roman Pantheon and Hawksmoor's ceiling are generated by a conformal mapping (the inverse of Mercator's projection) that minimizes an energy functional, providing a parameter-free geometric explanation that aligns with empirical measurements and suggests feasible pre-modern construction methods.

Original authors: John Cardy

Published 2026-08-13

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: John Cardy

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you are trying to wrap a perfectly square gift, but the object you are wrapping is a bumpy, curved dome. If you try to lay flat, square pieces of paper onto a sphere, they will either bunch up, leave gaps, or get stretched into weird shapes. This is the fundamental puzzle of geometry on curved surfaces: you can't flatten a sphere without tearing or stretching it. This problem sits at the intersection of architecture, history, and a branch of math called differential geometry, which studies how shapes bend and twist in space.

For centuries, people have wondered how ancient builders and later masters managed to decorate domes with grids of "coffers" (sunken panels) that look almost like perfect squares, even though the ceiling is curved. The key concept here is conformality. Think of it as a magical map-making rule: if you draw a grid on a flat piece of paper and then stretch that paper onto a curved surface without twisting or shearing it, the corners of your squares will still meet at perfect right angles, even if the squares themselves get bigger or smaller. Another key idea is Mercator's Projection, the famous map of the world used by sailors. On a Mercator map, Greenland looks huge compared to Africa, even though it isn't, because the map stretches the poles to keep the angles of compass directions correct. This paper suggests that the secret to these beautiful domes is the exact mathematical opposite of that map.

The Mystery of the Squares on the Dome

Have you ever looked up at a grand, domed ceiling and noticed the grid of sunken squares? It looks like a checkerboard, but the squares get smaller as they get closer to the top, and they twist to fit the curve. For a long time, historians and mathematicians have been scratching their heads trying to figure out how architects pulled this off. How do you fit a grid of "almost square" boxes onto a curved dome without leaving ugly gaps or making the boxes look like trapezoids?

This paper takes a fresh look at two famous examples: the Roman Pantheon (built nearly 2,000 years ago) and the ceiling of the Buttery at All Souls College, Oxford, designed by the English architect Nicholas Hawksmoor in the 1700s. Both ceilings are doubly curved (like a dome), which makes tiling them with squares incredibly tricky. The author, John Cardy, proposes a surprisingly simple answer: these ceilings are the result of a conformal mapping.

In plain English, this means the architects (or the laws of physics guiding them) essentially took a flat, perfect square grid and "projected" it onto the curved ceiling in a way that preserves angles. Just as a Mercator map stretches the poles to keep compass lines straight, these ceilings shrink the squares near the top to keep the grid lines meeting at perfect 90-degree angles. The paper suggests that the pattern you see on the Pantheon and Hawksmoor's ceiling is actually the inverse of a Mercator projection. If you could take the ceiling and map it back to a flat rectangle, you would get a perfect grid of squares.

The "Magic" Math Behind the Masonry

The author didn't just guess; they used the math of differential geometry to prove that if you want a grid where the lines always cross at right angles and the boxes stay roughly square, there is only one unique way to do it. This unique way is the conformal map.

They tested this idea against real data. For the Hawksmoor ceiling in Oxford, they compared their mathematical predictions against photographs. By adjusting for how the camera distorted the image, they found that the predicted locations of the grid points matched the actual ceiling almost perfectly. For the Pantheon, they compared their theory against modern, high-precision measurements of the dome's dimensions. Again, the math held up: the sizes of the coffers followed the exact pattern predicted by the conformal hypothesis, shrinking in a specific way as they went up the dome.

Why Does This Happen? (The Physics of Stability)

You might wonder: "Did the Romans or Hawksmoor know about complex math equations?" Probably not. The paper offers a fascinating alternative explanation: physics.

The author argues that these structures are conformal because it's the most stable way to build them. Imagine the ceiling as a flexible sheet. If you try to build a grid that isn't conformal, the structure is under stress, like a rubber band being twisted. The paper suggests that nature (and smart builders) naturally gravitates toward the shape that minimizes energy. By minimizing the "elastic energy" (the stress in the material) and the "gravitational energy" (the weight of the structure), the ceiling naturally settles into this conformal pattern. In other words, if the grid weren't conformal, the dome might not have lasted 2,000 years. The structure had to be this way to stay standing.

How They Might Have Done It

The paper also plays detective, asking how these builders could have achieved such precision without modern computers or calculus.

For the Romans: They might have used a trial-and-error approach, slowly adjusting the size of each coffer until the structure felt stable. Or, perhaps an observer standing in the center of the room could have guided the builders, telling them when a new coffer looked "square enough" from their specific viewpoint.
For Hawksmoor: Since the Buttery ceiling is made of individual stone blocks, Hawksmoor could have realized that the blocks in each row needed to be identical. He could have had masons mass-produce the central blocks and then simply scale them down as they moved toward the curved ends, following a simple rule of thumb that mimics the complex math.

What the Paper Rules Out

It's important to note what this paper says is not the answer. The author explicitly argues against the idea that the pattern is based on simple, arbitrary ratios of whole numbers or that the builders used complex spherical trigonometry to "unfold" the sphere (which is mathematically impossible for a doubly curved surface). They also suggest that previous theories relying on artists' impressions or inaccurate measurements were missing the underlying universal rule. The paper doesn't claim to have found a "secret code" written in the stones, but rather that the pattern is a natural consequence of geometry and physics.

The Bottom Line

This paper suggests that the beautiful, grid-like patterns on the Pantheon and Hawksmoor's ceiling aren't just artistic choices; they are the result of a unique mathematical solution that preserves right angles. Whether discovered through advanced math, trial and error, or the laws of physics, the result is the same: a conformal map that turns a flat grid into a perfect, stable dome. The author is confident in their mathematical model because it fits the real-world data so well, offering a new way to understand how humanity has mastered the art of building on curves for millennia.

Technical Summary: Conformality in Hawksmoor's Ceiling, Mercator's Projection and the Roman Pantheon

Problem Statement
The paper addresses the geometric challenge of tiling doubly curved surfaces with (approximately) square coffers. Specifically, it investigates the ceiling of the Buttery in All Souls College, Oxford (designed by Nicholas Hawksmoor, 1661–1736) and the dome of the Roman Pantheon. Both structures feature a grid of ribs intersecting orthogonally to form sunken lacunaria (coffers). While the Pantheon's geometry has been subject to extensive historical and metrological study, no single unifying principle has previously been established to fix its overall geometry. Similarly, Hawksmoor's ceiling presents a complex tiling problem on a surface of revolution (a "half-capsule" composed of two quarter-spheres and a barrel vault) where the coffers vary in size and orientation but maintain a square aspect ratio and orthogonal intersections. The core problem is to determine the optimal distribution of these coffers on a curved surface without gaps or overlaps, minimizing the deviation from squareness.

Methodology
The author employs methods from differential geometry and variational calculus to model the ceiling as a smooth 2-manifold immersed in 3-dimensional Euclidean space.

Conformal Mapping Hypothesis: The paper hypothesizes that the coffer lines are the images of a uniform $M \times N$ square tiling of a planar rectangle under a conformal (angle-preserving) mapping. This implies that the mapping is the inverse of Mercator's projection (which maps a sphere to a cylinder/rectangle).
Isothermal Coordinates: The analysis utilizes isothermal coordinates, where the metric on the curved surface takes the form $(ds)^2 = \lambda^2(d\xi^2 + d\eta^2)$ . In such coordinates, a regular square grid in the plane maps to a grid of orthogonal curves on the surface.
Surface of Revolution: For ceilings generated by rotating a curve $\rho = \rho(z)$ , the author derives explicit expressions for the coordinates of the vertices. The spacing between coffer rows is determined by an integral involving the profile of the ceiling:
$\frac{m\pi}{N} = \int_0^{z_m} \frac{\sqrt{1 + \rho'(z)^2}}{\rho(z)} dz$
Physical Justification (Variational Principle): To explain why such a design might emerge, the author constructs an effective energy functional representing the sum of elastic (shear, compression, bending) and gravitational potential energies. By minimizing this functional in the long-wavelength limit, the author demonstrates that the equilibrium state corresponds to a conformal map. Non-conformal designs would possess unstable modes under long-wavelength deformations.
Data Comparison: The theoretical predictions are compared against photographic data of the Hawksmoor ceiling and modern metrological measurements of the Pantheon. The comparison involves applying a projective transformation to the theoretical 3D coordinates to simulate camera distortion.

Key Contributions and Results

Unification of Geometry: The paper establishes that the coffering patterns of both the Hawksmoor ceiling and the Pantheon dome are consistent with being conformal images of a planar square grid. This provides a unique geometric solution for any surface of revolution.
Explicit Predictions: For a hemispherical ceiling of radius $R$ , the vertical position $z_m$ of the $m$ -th row of coffers is predicted to be $z_m = R \tanh(\pi m / N)$ , and the radius $\rho_m$ is $R \text{ sech}(\pi m / N)$ . For the Pantheon ( $N=28$ ), the size of the coffers should scale with $\text{sech}(2\pi m / N)$ .
Quantitative Agreement:
- Hawksmoor Ceiling: The predicted vertex locations, when projected through a simulated pinhole camera, align excellently with photographic evidence of the All Souls ceiling. The fit requires only one adjustable parameter (camera height), which yields a physically plausible value.
- Pantheon Dome: The predicted ratios of coffer heights and widths match modern measurements (from the Bern Digital Pantheon Project) without any adjustable parameters. While the limited number of data points prevents probing the exponential tail of the distribution, the existing data does not refute the hypothesis.
Construction Protocols: The paper suggests that while the architects likely did not possess differential geometry, the conformal solution could be approximated through iterative physical methods (e.g., minimizing a discrete Dirichlet energy or observing "squareness" from a central vantage point) or by recognizing that the size of coffers in a surface of revolution should scale with the local height/radius.

Significance and Claims
The paper claims to offer the first quantitative analysis of the Hawksmoor ceiling and a new theoretical framework for the Pantheon's coffering.

Novelty: The identification of these architectural patterns as conformal mappings (and inverses of Mercator projections) is presented as a new proposition in the literature.
Physical Necessity: The author argues that conformality is not merely an aesthetic choice but a physical necessity for stability. A coffering design that minimizes the sum of elastic and gravitational energies for a fixed mass must be conformal to avoid instability against long-wavelength deformations.
Historical Context: The work bridges the gap between ancient/Baroque construction and modern mathematical physics. It suggests that the stability of these structures over millennia is a consequence of them naturally settling into a conformal equilibrium, accessible through empirical trial-and-error or simple geometric rules rather than advanced calculus.
Modesty: The author acknowledges that the conformal hypothesis is a mathematical idealization. They note that real-world constructions involve imperfections and that the Pantheon's complex multi-layered concrete structure may not adhere perfectly to simple rules, yet the underlying geometric principle appears to hold. The paper does not claim to have discovered the exact historical construction method used by the Romans or Hawksmoor, but rather proposes a plausible physical and geometric rationale for the observed patterns.