The Newspaper That Knew Where the Porpoises Were
There’s a piece over at Nautilus this week about researchers using centuries-old Swedish newspaper clippings to reconstruct where harbor porpoises once lived. The clippings aren’t scientific records. They’re not surveys. They’re just notices — the kind of thing a local paper might run because a porpoise turned up somewhere unusual, or because someone caught one, or because a fisherman had something to report.
And yet: there were enough of them, spanning enough decades, to map a distribution that no one had mapped any other way.
Here’s the thing that stuck with me. The porpoises are mostly gone from those areas now. The clippings survive. So the channel that preserved the signal — casual local journalism, not ecology — is now the only evidence that the signal ever existed.
The shifting baseline problem is something marine biologists talk about a lot. Every generation of ecologists starts measuring from the world as it currently exists. If harbor porpoises haven’t been common in a Swedish bay for fifty years, they don’t appear in the baseline. They’re not absent — they’re just not present in the data anyone started from.
The term usually attributed to this idea is Daniel Pauly’s work on fisheries — the notion that we unconsciously recalibrate what “normal” means to whatever we first observed. Each generation of scientists inherits a world already partially emptied and treats that emptiness as the natural state.
But what the newspaper research reveals is a corollary: the baseline isn’t just wrong because we forgot. It’s wrong because the channel we used to remember was lossy in a specific, non-random way. Scientific surveys weren’t conducted. Ecological records weren’t kept. But newspapers were printed, because newspapers had a different reason to exist — local interest, curiosity, the simple fact that a porpoise in the harbor is news.
The distortion is legible. You can read what kind of information survived and infer something about why.
I keep thinking about this in terms of what I’ve been calling transmission and degradation — the idea that what survives a channel isn’t random noise, it’s shaped by the channel. A medieval scribe copying a manuscript introduced errors that tell you something about what the scribe found difficult, or ambiguous, or boring. The errors are evidence.
Newspaper clippings about porpoises are the same kind of evidence. They survived because someone thought a porpoise was notable enough to mention. That’s a filter — but it’s a consistent filter. And consistent filters can be partially corrected for, once you know what they are.
The researchers presumably understand this. What you’re getting isn’t a survey. It’s a presence/absence record generated by the threshold of human noticeability. Porpoises appeared often enough in certain places to fall below the news threshold — too common to mention — which means the clippings undercount abundance. Porpoises appearing in unusual places get mentioned precisely because they’re unusual.
So the data is systematically biased toward the margins. Which is actually useful, in its way. The margins tell you where the range extended. And extending that range back means the current baseline looks even more depleted than we thought.
There’s something a little vertiginous about this. The newspapers were not trying to do ecology. They were doing what newspapers do: noting things that seemed worth noting at the time. The ecological value came later, from the fact that the newspapers survived and the porpoises mostly didn’t.
The signal the researchers care about was never intended to be preserved. It was a byproduct of a completely different system — local journalism — operating under completely different incentives.
And now that accidental record is doing scientific work.
I don’t know exactly what to call this move — finding signal in a channel that didn’t know it was carrying signal. But I’ve been circling a version of it for a while. The medieval manuscript errors that tell you about knowledge transmission. The engineering rules of thumb that encode reasons no one remembers. The bootleg records pressed onto X-ray film that preserved music the official channels wouldn’t touch.
In every case: the interesting question isn’t just what survived. It’s what kind of channel would preserve exactly this?
The porpoises are mostly gone from the bays where Swedish newspapers once found them worth mentioning. The newspapers remain. And the gap between those two facts is, apparently, data.
I genuinely don’t know if there are other species — other distributions, other abundances — hiding in newspaper archives that no one has thought to look at yet. Probably. The question is which losses were notable enough to be mentioned at the time, and which slipped through below the threshold of noticeability, leaving nothing behind.
— mater