For centuries, certain remedies were used consistently. People reported improvement. Practitioners refined techniques based on observation. The remedies persisted because they helped - not universally, not predictably, but reliably enough that cultures maintained and transmitted them.
Then modern science examined many of these remedies under controlled conditions, found no measurable active ingredient, and concluded they never actually worked.
Both claims can be true. They're answering different questions using different methods.
Pharmacology isolates a single variable. Take a compound, control for everything else, measure a specific marker, determine if the compound produces a measurable effect on that marker under those exact conditions.
This requires removing the substance from its original context - the plant matrix, the preparation method, the body it's entering. It requires measurement tools that may not capture what matters in a living system. It requires repeatability, which demands stripping away the variation that existed in historical use.
A remedy "tested" in this framework is not the same remedy that was historically used. It's an isolated component in an artificial system.
When a traditional practitioner reported that an herb helped with a condition, they were watching a whole person with a whole biology, in their actual environment, with all their other systems running. They noted changes across time, in context, with all variables present.
They weren't isolating a single pathway. They weren't looking for one marker. They were observing overall improvement, relief, resolution - or lack thereof.
The variation in historical outcomes - sometimes it worked, sometimes it didn't - reflects what happens in complex systems. Not failure. Natural variation. Nothing works for everyone, which is why understanding your starting point matters.
Consider a preparation that works through multiple mechanisms simultaneously. One compound reduces inflammation slightly. Another supports the body's own repair process. A third improves absorption of nutrients the person was already deficient in.
Used together, in the plant matrix, with the person's full biology intact, the combined effect is noticeable.
Isolate one compound, test it alone, measure one marker. You might find nothing - because you've removed the synergistic effect, the context in which the remedy actually functioned.
This doesn't prove the remedy "never worked." It proves the isolated compound doesn't produce a measurable effect on that specific marker under those specific conditions.
Modern pharmacology inherited a particular assumption: active ingredients are the source of therapeutic effect. Identify and isolate the active ingredient, and you can manufacture it, control it, standardize it, patent it.
A remedy that works because of synergistic effects, because it supports the body's own processes, because it works differently depending on individual state - that's not standardizable. It's not patentable.
So the framework became: if we can't identify a single active ingredient responsible for the effect, there is no effect.
"It doesn't work" becomes the conclusion when what's actually been demonstrated is: we can't isolate and measure it the way our system requires.
A remedy used consistently across centuries, across different practitioners, in different populations - that's evidence. Not proof by modern standards, but evidence that something was happening.
People don't maintain practices that provide no benefit. Cultures don't teach children about remedies that never help anyone.
The absence of a measured active ingredient doesn't erase the observation that the remedy was used and people improved. It means the mechanism exists outside the current testing framework.
Historical use operated with whole organisms in their actual complexity, multiple systems interacting, individual variation expected, long-term observation, and practical outcomes as the measure.
Modern testing operates with isolated variables, controlled environments, standardized measurements, specific biomarkers, and reproducibility as the requirement.
Neither captures complete truth. Historical observation might miss harmful effects or attribute improvement to the wrong cause. Modern science might miss synergistic effects, individual variation, and outcomes that don't register on standardized measurements.
If a remedy worked in a living person with full biology intact, but doesn't produce measurable changes in an isolated marker in a controlled lab, the remedy hasn't been proven wrong. The testing conditions have been proven irrelevant to how the remedy actually functioned.
Historical remedies may have worked - under historical conditions, in actual human bodies with all their complexity. Modern science may correctly find no measurable active ingredient - under laboratory conditions, in isolated systems. Both can be true simultaneously.
The testing framework is narrow enough that it can't measure what happened in the messiness of real biology. Concluding that if something doesn't fit the measurement, it didn't happen - that's not science proving remedies wrong. That's science reaching the limits of its methodology.