In 1895, the Lumière Brothers filed the patent for the Cinématographe. The document does not merely protect the apparatus. It protects the intermittent film advance mechanism, the combination of shooting and projection in a single device, the principle of the claw that advances the film frame by frame. For seventeen years, any competitor wishing to produce moving images must circumvent these protections. Edison circumvents through a different mechanism. Pathé negotiates a license. Others abandon.

The patent has not merely protected an invention. It has drawn the shape of all inventions that followed by forcing them to avoid a precise territory.

A patent consists of claims. Each claim delimits a technical space of which the inventor receives exclusive use. What surrounds this space is free. What touches it is litigious. What enters it is infringement. Intellectual property jurists call this space the scope of protection. Engineers call it otherwise: the zone where one must not go.

The pharmaceutical industry has carried this logic to its extreme. Pfizer's patents on atorvastatin do not merely cover the molecule. They cover families of compounds, synthesis pathways, formulations, dosages, therapeutic indications. The protected territory is a volume in the space of possible chemical solutions. Each competitor seeking a cholesterol medication must navigate around this volume. Research does not begin with the question "which molecule works?" but with the question "which molecule works and is not in Pfizer's scope?". The juridical constraint precedes the chemical constraint.

This constraint is not merely a restriction. It is productive. The prohibition of the obvious solution forces exploration of solutions that no one would have sought otherwise. Circumvention engenders heterodox architectures, unforeseen synthesis pathways, alternative mechanisms some of which prove superior to the original. Watt's patent on the steam engine blocked development of the separate condenser for thirty years. This blockage pushed Trevithick toward high-pressure machines, which made the locomotive possible. Prohibition produced what permission would not have sought.

In the same movement, the patent makes possible the research that will circumvent it. Without the protection guaranteed by the patent, investment in pharmaceutical research is not justified. Pfizer's protected volume is the condition that has financed the research whose results it then constrains. The patent is simultaneously what closes the landscape and what finances its exploration.

Patent thickets are configurations in which several holders have filed patents on adjacent technologies such that no new entrant can innovate without violating at least one of them. The semiconductor industry has functioned thus since the 1980s. Intel, Samsung, TSMC each hold thousands of patents that partially overlap. The result is not total blockage but an economy of cross-licensing: each authorizes the others to use its patents in exchange for reciprocal access. Companies that do not possess sufficient portfolio to negotiate do not enter. The patent no longer protects an invention. It protects a negotiating position.

Doctrine

The patent does not describe what it protects. It draws what it prohibits, and what it prohibits produces what it had not foreseen. The space of accessible technical solutions is the negative of the space of filed claims. But this negative is not a void. It is a territory that prohibition has made explorable.

Open vector

A patent expires after twenty years. The space it prohibited becomes accessible again. But twenty years of exclusion have produced alternative paths, entire architectures built to avoid the protected territory. When the patent falls, the liberated territory is often empty, no one goes there because the industry has reconfigured itself around its absence. The explosion of innovation that followed the expiration of Watt's patents did not exploit his patents. It exploited all that his patents had prevented from being explored for thirty years. The legacy of a patent is not what it protected. It is the shape of the void it left.

References

H. Chevotet Researcher — Field Theory