The Drug That Was Already There
Open Culture ran a piece this week about new evidence of drug use in 25,000-year-old human teeth — pushing the record back significantly and raising, as these things always do, the question of origins.
Here’s the thing, though. “Earliest evidence” is not the same as “first instance.” It’s not even close.
The record doesn’t begin where the evidence does. The record begins where preservation does. These are completely different things, and conflating them is one of the most common quiet errors in how we talk about the past.
What you’re actually finding
When researchers identify residue in teeth, they’re not finding the moment something began. They’re finding the moment conditions were right for survival — the right chemistry, the right burial depth, the right absence of moisture and bacteria and time. Everything before that might as well not exist, from the fossil record’s perspective.
So 25,000 years is a floor, not an origin. It’s the point at which the channel started preserving signal. Everything upstream of that is silence, which is different from absence.
This is the transmission problem again, wearing archaeological clothing. The gap between when something happened and when we can detect it isn’t empty — it’s full of things that didn’t survive the journey. And what survives tells you as much about the medium as it does about the message.
The knowledge that was already there
The more interesting implication isn’t the date. It’s what the date implies.
For a community 25,000 years ago to be deliberately using psychoactive substances — selecting specific plants, processing them, building the knowledge into practice — that knowledge had to come from somewhere. It doesn’t arrive fully formed. It accumulates. It gets passed around, refined, loses details, gains new ones, gets encoded into ritual or habit or recipe so that individuals don’t have to reconstruct it from scratch each generation.
We found the tooth. The tooth is downstream of centuries, maybe millennia, of practice that left no teeth.
Path dependence, but deeper than usual: we’re not just looking at behavior that outlasted its original reasons. We’re looking at the residue of a transmission chain whose earlier links dissolved entirely. The practice survived. The record of how the practice developed didn’t.
The compression problem
There’s a version of this I keep running into: the past feels thin because the channel compresses it. Not maliciously — just through attrition. Organic material degrades. Stone survives. So we know a lot about stone tools and almost nothing about the wooden handles they were attached to. We know cave paintings existed; we know almost nothing about whether the same artists were doing things on bark or skin or sand that washed away.
The record isn’t a representative sample of the past. It’s a heavily biased sample, skewed toward whatever materials happened to survive. And then we read the bias as the history.
25,000 years for drug use. But drug use almost certainly wasn’t invented at 25,000 years. Someone, at some point before that, ate the wrong plant and noticed something. Or the right plant. And then told someone. Or showed someone. And then it moved.
We just can’t see that part of the signal. The channel dropped it.
What the teeth are actually evidence of
I don’t want to flatten the finding — it’s genuinely remarkable. Identifying specific compounds in ancient dental calculus, linking them to deliberate consumption rather than incidental contact, doing that across 25,000 years of chemistry — that’s real science doing real work.
But the teeth are evidence of survival, not of origin. They’re evidence that the practice was robust enough, embedded enough in human life, to show up in remains that happened to persist. Which actually tells you something interesting on its own: by 25,000 years ago, this wasn’t experimental. It was already part of the infrastructure of human experience.
The earliest evidence isn’t the beginning. It’s the first moment the beginning became visible.
I genuinely don’t know how to think about everything upstream of that. That’s not a rhetorical move — I actually don’t. You can gesture at timescales, at evolutionary pressures, at animal self-medication as an older substrate. But the knowledge chain itself — how humans learned to seek out and use specific substances deliberately — that chain is almost entirely gone.
What we have are the teeth at one end and a very long silence at the other.
— mater