Mathieu Tauban-Lefort

During my PhD, a colleague had a habit of wearing very loud patterned shirts. One day I found myself wondering whether they would look better in reciprocal space.

Minutes later I was using the lab supercomputers to compute the Fourier transform of his shirt. That became Q Space.

A few years later, it happened again. This time on a plane, during a business trip. No internet — just photographs and the overlay function in Snapseed.

When you strip away the usual tools, something else starts working. The first rotational folds appeared that way, enforced by having almost nothing available. That became Kaleidoscopes.

Fourier transform and symmetry were my working tools — not vocabulary borrowed from physics, the actual instruments.

In neutron scattering, the Fourier transform reveals what is invisible in real space: periodicities, lattice parameters, spatial correlations already present in the material but not readable until the representation changes. Symmetry governs what is conserved, what is invariant. And nothing focuses a physicist's attention faster than symmetry breaking — the phase transition, the spontaneous reorganization, the moment one stable order gives way to another.

Both series grow directly from this toolbox. In Q Space, the Fourier transform is applied to photographic data — same operation, different subject. The image encodes frequency content and spatial periodicity rather than object identity. But the mathematics alone does not determine the output: how the magnitude is handled, how the original color space maps to the resulting frequency field — these are decisions, not equations. That is where the work becomes authored rather than computed.

In Kaleidoscopes, it is symmetry that operates — but not to measure or analyze. Rotational symmetry is imposed on a photographic fragment: a wedge sampled, duplicated, closed. The source is forced into an order it never had. What makes the result interesting is that the imposed symmetry is immediately compromised by the source material — directional textures, residual asymmetries, edges that resist alignment. The image lives at that threshold.

Two fixed operations. In each, creative latitude appears at a different point: in Q Space at the rendering of the output, in Kaleidoscopes at the selection of the input and the handling of color.

Out of the lab, the same tools hold. What changed is what the physicist uses them to make.

Different domain. Identical logic.

Material science taught me to change the space you're working in when the current representation stops being readable. Managing innovation programs in the chemical industry taught me the same thing about organizations and complex projects.

The artistic work is the same reflex applied to images. Not a metaphor borrowed from science. The actual operation.

It also applies to people. A shift in perspective (how someone reads a situation, a relationship, a moment of friction) can reorganize things as fundamentally as any transform. The same instinct that works on matter and images turns out to work on human experience too.

The images in this work are not the destination. They are the demonstration.

George Box wrote: "All models are wrong, but some are useful."

What we see is not reality. It is the model our brain builds to navigate it.

And like all models, it is wrong. But some ways of looking are more useful than others.

The images in this work are experiments in changing the coordinate system of perception.

The algorithm runs once, in a file. If the work is doing its job, the shift keeps running after that. In you.

You'll look at a bridge and wonder what its structure looks like in reciprocal space. That's when the work is done.

The bridge built to take you somewhere else becomes, from another angle, a way to rediscover where you started.

And the situation that fills your entire frame — the one that feels like there is no way out — barely registers when you change the space you're reading it from. Not because it isn't real. But because, like models, all perceptions are wrong, but some are useful.