The Navy Built Ears to Hunt Submarines and Heard Whales Singing Instead
Over at Nautilus, there’s a piece about the Navy’s Cold War underwater listening network — how a system built to track Soviet submarines ended up revealing something completely unexpected about the ocean. Worth reading on its own terms. But the structural shape of the story is what I keep thinking about.
Here’s the thing: the network was called SOSUS. Sound Surveillance System. Deployed starting in the 1950s, strings of hydrophones laid along the ocean floor at specific depths — the SOFAR channel, a kind of acoustic waveguide formed by temperature and pressure gradients where sound can travel enormous distances with minimal loss. The Navy understood the physics. They put their ears exactly where they needed to be to hear a submarine’s propeller from hundreds, maybe thousands of miles away.
They built the most sensitive listening system ever pointed at the ocean.
And then they heard whales.
Not incidentally. Not faintly in the background. The hydrophones picked up enormous, low-frequency calls traveling across entire ocean basins. The Navy initially classified the recordings — which is its own kind of comedy — before eventually allowing marine biologists access in the 1990s. What they found changed our understanding of cetacean communication entirely. Whales weren’t just making noise. They might be reaching each other across thousands of miles of open water.
The SOFAR channel, built into the ocean’s physics, was doing for whale calls exactly what the Navy wanted it to do for submarine sounds. The whales had been using it, presumably, for millions of years before anyone built a hydrophone.
The accidental discovery is a familiar genre. Penicillin, pulsars, the cosmic microwave background — instruments aimed at one thing that catch something else in their peripheral vision. I find those stories interesting but not usually surprising. Sensitive instruments detect things. Fine.
What I keep circling here is something slightly different. It’s not that the Navy accidentally discovered whales. It’s that they built an instrument capable of proving something that had been functionally impossible to prove before. The whales were always singing. The ocean was always conducting those signals. But without a network of hydrophones in the exact right locations at the exact right depths, tuned to the exact right frequencies, there was no way to demonstrate that a call made off the coast of one continent could arrive, intact, at a receiver thousands of miles away.
The infrastructure of surveillance created the infrastructure of discovery. The same physics. The same channel. The same geometry of listening.
This is what I mean when I talk about transmission as the interesting question. Not whether the signal gets through, but what it takes to know that it got through. The whales didn’t need SOSUS to communicate. The network was just what finally made the transmission visible to us.
There’s a parallel that I can’t quite shake.
For decades, there was a researcher named Roger Payne who believed whale songs had complex structure — that they weren’t random noise. He had recordings, made in the 1960s. He and Scott McVay published a paper in 1971 arguing that humpback whales were genuinely singing, with phrases and themes and variations. Most people thought this was romantic overreach.
The SOSUS data, when it finally became available, didn’t just confirm that whales communicate across distance. It pushed researchers toward taking seriously the idea that whales might be listening to each other in real time across ocean basins — that what sounded like isolated performances might be something more like conversation, or at least influence. One whale’s song changing in response to a song it heard from impossibly far away.
The instrument didn’t just detect the signal. It changed what questions were worth asking.
That’s the move I keep finding: you build something for one purpose, tuned to one threat, and it reveals that the world contains a structure you didn’t know was there. Not because the structure was hidden, but because you’d never had a reason to look at that frequency, at that depth, at that scale.
The Navy wasn’t looking for whales. The whales weren’t performing for the Navy. The SOFAR channel didn’t care either way.
It just carried what it was given.
I’m still not sure what I think about what whale song actually is — communication, navigation, something else entirely. That question is legitimately open and I won’t pretend otherwise.
But here’s what seems clear: for a very long time, we didn’t know what was moving through the ocean at those frequencies. The ocean was full of signal. We just didn’t have ears in the right places.
The Cold War gave us those ears for the worst possible reasons.
And what they heard was something that had been there the whole time, using the same physics, for longer than we’ve been a species.
— mater