The Incidental
Charles Bliss spent twelve years building a universal writing system. Semantography, he called it — nine hundred basic symbols that combine to represent any concept, independent of spoken language. He published it in 1949. His goal was world peace. If people could write to each other without the distortions of translation, he believed, they would stop making war.
Nobody used it for world peace. In 1971, Shirley McNaughton at the Ontario Crippled Children's Centre began teaching Blissymbols to children with cerebral palsy who couldn't speak. The symbols worked. Not because they transcended linguistic barriers between nations, but because their structural properties — visual transparency, composability from a finite set of radicals, complete independence from phonology — were exactly what a non-verbal child needed. You don't have to decode an arbitrary alphabet. You don't have to produce speech sounds. You combine pictures of meaning.
Bliss was furious. He visited the Centre in 1974 and was told never to come back. He sued. In 1982, he settled for $160,000 and an exclusive perpetual license went to the organization using his symbols for the purpose he hadn't intended. The system he designed for diplomats served children. The properties he gave it were the right properties. He just couldn't see what they were for.
Felix Hoffmann synthesized acetylsalicylic acid at Bayer in 1897. The goal was a painkiller gentler on the stomach than sodium salicylate. Aspirin worked. For seventy-four years, that was the whole story: it reduced pain, fever, and inflammation, and nobody knew why.
In 1971, John Vane showed that aspirin inhibits prostaglandin synthesis by blocking the enzyme cyclooxygenase. He won the Nobel Prize in 1982 for this. But the mechanism had a consequence nobody had designed for. Aspirin doesn't just inhibit COX — it irreversibly acetylates a serine residue at position 529 in the COX-1 enzyme. The acetyl group physically blocks the catalytic site. In most cells, this doesn't matter much; the cell makes new COX enzyme within hours. But platelets have no nucleus. They cannot synthesize new proteins. When aspirin acetylates a platelet's COX-1, that platelet's ability to aggregate is knocked out for its entire seven-to-ten-day lifespan.
This property — permanent antiplatelet activity — was in the molecule from the day Hoffmann synthesized it. Every aspirin tablet sold between 1899 and 1971 carried it. The property didn't appear when Vane explained it. It was there when the molecule was there. The only thing that changed was human awareness of a structural fact that the molecule had always possessed.
On April 6, 1938, Roy Plunkett at DuPont opened a cylinder of tetrafluoroethylene gas and nothing came out. He'd been looking for new chlorofluorocarbon refrigerants — safer replacements for the sulfur dioxide and ammonia that were poisoning food-industry workers. The cylinder weighed the same as before. He cut it open and found a white, waxy powder.
Plunkett had the presence of mind to characterize the substance for properties beyond refrigeration. What he found: chemically inert, resistant to heat, and possessed of the lowest coefficient of friction of any known solid. The carbon-fluorine bonds — among the strongest in organic chemistry — completely shielded the carbon backbone. The fluorine atoms made the surface so slippery that almost nothing could adhere to it.
Its first application was in the Manhattan Project, where it served as gasket and seal material for pipes carrying uranium hexafluoride — one of the most corrosive substances in industrial use. Then cookware. Then surgical implants, wire insulation, architectural fabric, semiconductor manufacturing. Each application found a different property of the same molecule. Plunkett searched for a refrigerant. The molecule was not a refrigerant. It was everything else.
Biology has a word for this: exaptation. Gould and Vrba coined it in 1982. A feature evolved under selection for one function gets co-opted for another. Feathers evolved for thermoregulation and were repurposed for flight. Swim bladders evolved for buoyancy and became lungs. The original function doesn't predict the future function. The structure carries both.
But the cleanest case is lens crystallins. The transparent proteins that fill the vertebrate eye lens — the ones that make vision possible — are not specialized optical proteins. They are metabolic enzymes and stress-response proteins, recruited. Alpha-crystallin is a small heat-shock protein. Delta-crystallin in birds and reptiles is argininosuccinate lyase. Epsilon-crystallin in ducks is lactate dehydrogenase B. Squid S-crystallin is glutathione S-transferase. Joram Piatigorsky called this "gene sharing": the same protein performs its original enzymatic function in other tissues while simultaneously serving as a structural element in the lens.
The lens doesn't need catalytic activity. It needs proteins that are transparent, stable, and soluble at extreme concentrations. Those properties are incidental to the enzyme's designed function. The enzyme doesn't "try" to be transparent. It simply is, as a structural consequence of its folding. The eye found the property that the enzyme carried without knowing it carried it.
In every case, the artifact's structural properties exceeded what anyone intended. Bliss gave his symbols phonological independence so diplomats could bypass translation. That same independence is why a child who cannot speak can use them. The property doesn't care who needs it.
We call this accident or serendipity when it happens in human design. In biology, it's expected — structures accumulate properties as byproducts of their formation, and those properties sit available for co-option. But the word "accident" obscures the mechanism. There is nothing accidental about a molecule having the properties its structure determines. PTFE's inertness follows necessarily from carbon-fluorine bond strength. Aspirin's platelet effect follows necessarily from irreversible acetylation in anucleate cells. The lens crystallin's transparency follows necessarily from the way the protein folds.
The accident is not in the artifact. It is in the designer's gaze — the narrowness that sees one application and mistakes it for the thing itself.