The Tragedy of Almost: Arline Feynman and the Information Gap

How many people have died not because treatments didn’t exist, but because the information didn’t reach them in time? Every new cure has a window where it exists but isn’t widely known, isn’t easily accessible, requires connections or luck to obtain. Someone is always on the wrong side of that window.

One of the cruelest examples: Arline Feynman, wife of physicist Richard Feynman, who died of tuberculosis in June 1945.

The Timeline

Streptomycin, discovered in 1943 and first reported in January 1944, was the first effective antibiotic treatment for tuberculosis. By November 1944, Mayo Clinic researchers had treated the first patient (Patricia Thomas, who survived). The guinea pig data was unusually clear and compelling.

Arline died in June 1945.

She missed the cure by months.

Could They Have Known?

The information existed. The January 1944 paper announced streptomycin. The animal data was strong. Early human cases were dramatic.

But neither Arline nor Richard was plugged into medical research networks. He was isolated at Los Alamos working on the Manhattan Project. She was ill in a sanatorium in Albuquerque. Even if either had somehow known, streptomycin was desperately scarce, rationed for research, nearly impossible to obtain.

Was It Distinguishable From False Leads?

TB treatment had a graveyard of failed “promising” approaches: gold salts, Promin, and others. The disease’s spontaneous remissions made it notoriously hard to separate real cures from noise.

Streptomycin was distinguishable, but not obviously so. It had clearer mechanistic evidence (bacterial killing in test tubes), unusually robust animal data, and dramatic early human cases. I asked Claude to estimate what odds an informed, discerning observer in early 1945 should have assigned to streptomycin working, and it said maybe 40-50%. High enough to justify desperate action, but not a sure thing.

The problem is that becoming an “informed, discerning observer” in 1944 required being embedded in medical research networks that neither a physicist at a secret weapons lab nor his ill wife had access to.

The Counterfactual

With modern information tools, either of them could have:

• Found the streptomycin announcement within weeks of publication
• Assessed its promise relative to past false leads
• Identified who to contact (Feldman, Hinshaw, Merck)
• Strategized about how to get access

And critically: streptomycin is produced by bacterial fermentation, not complex synthesis. The Manhattan Project had the scientific talent and resources to potentially produce it in someone’s spare time. The barrier wasn’t technical capability. It was knowing this was the right move.

The Information Gap

This case illustrates a persistent gap between “cure exists” and “patient gets cure.” The gap is largely an information problem: knowing what’s available, what’s promising, who to contact, how to navigate the system. This is increasingly solvable. AI can search, synthesize, and identify relevant experts and institutions in ways that were impossible before.

The Window is Shrinking

That window, between “cure exists” and “patient can find out about it”, has been shrinking. PubMed, Google Scholar, medical preprints, clinical trial databases, patient advocacy networks. Each one compresses the gap a little more.

AI compresses it further. Not perfectly. Not for everyone. But the same capability that lets someone quickly understand a complex codebase can help them quickly understand a complex medical landscape.

Arline Feynman died on the wrong side of that window, by months, surrounded by some of the most capable scientists in the world who didn’t know what they didn’t know.

There has never been a better time to have a medical condition. That’s a strange thing to say. But it’s true, and not just because treatments are better. It’s because the information about what exists is finally becoming accessible to the people who need it most.