FIG 1.0 — CrowLingo · Journal
Long-form essays on AI and crow communication.
Citation-driven, anti-hype, written for non-engineers but careful enough that practitioners in the field will find them load-bearing. Updated when there's something to say, not on a calendar.
80 essays · 6 sections
Editor’s picks · continued
3 of 80Frontier · 8 min
What 'translating' animal language would actually require
Why the popular framing keeps over-promising, and what the gap between 'AI can map a repertoire' and 'AI can translate' actually c…
Frontier · 6 min
The next decade of bioacoustic AI: a careful forecast
Where the field is plausibly going from 2026 to 2036. The honest version isn't 'we'll translate animal language.' It's both less a…
The Crow · 5 min
John Marzluff and the Seattle crow lab
Most of what the public knows about American crow cognition traces back to a single research program at the University of Washingt…
Jump to a section
The Crow
The species, the cognition, the social life.
33 essays
Why crows recognize human faces — the Marzluff experiments
A caveman mask, a college campus, and the most-cited cognitive finding in modern corvid research.
May 18, 20266 min · 1,320 wordsThe nine emergent clusters of American crow communication
What AI audio models recovered when given hundreds of hours of unlabeled crow audio — and how the names map to the old hand-labels.
May 18, 20268 min · 1,640 wordsWhy the American crow is the model species for AI bioacoustics
It's not just because they're charismatic. The American crow clears a specific set of empirical bars that make it the right species for the contemporary methods…
May 18, 20266 min · 1,290 wordsCorvid cognition: fifty years of breakthroughs
From the 1970s pioneers to the contemporary AI era, the through-line that connects what we know about how crows and ravens think.
May 18, 20269 min · 1,780 wordsCrow funerals and crow grief: separating evidence from anthropomorphism
What the careful research actually shows about crow gatherings around dead conspecifics, and where the popular framing reaches further than the data supports.
May 18, 20267 min · 1,480 wordsCrow tools and the cognitive ceiling
New Caledonian crows shape hooks from twigs. American crows don't, much. Why the difference matters for what we expect from corvid vocal cognition.
May 18, 20266 min · 1,340 wordsBernd Heinrich, Mind of the Raven, and the slow-naturalist tradition
Why a 1999 book by a Vermont biologist remains the methodological conscience of corvid research, even as AI methods take over the publication record.
May 18, 20266 min · 1,280 wordsWhat you'll hear in your backyard: a beginner's guide to crow vocalizations
If there's a crow within a hundred yards of your window, you're hearing it. Here's what each call type sounds like, what it means, and when to expect it.
May 18, 20267 min · 1,450 wordsHow to tell a crow from a raven by sound
The visual differences are subtle. The acoustic differences are not. If you can hear a corvid for ten seconds, you can usually tell the species.
May 18, 20265 min · 1,100 wordsWhy crows gather at dusk: the roost-formation phenomenon
Twenty thousand crows in one tree at sunset is not a glitch. It's one of the largest acoustic events in North American urban ecology, and it has a name.
May 18, 20266 min · 1,310 wordsThe cooperative-breeding family structure of American crows
Why young crows stay with their parents for years, what they do during that time, and how the family structure shapes everything else about crow communication.
May 18, 20266 min · 1,380 wordsDo crows mate for life? What the long-term pair-bond research shows
Mostly yes, with caveats. The empirical case for crow lifetime monogamy is stronger than for many bird species, and the social architecture that supports it is …
May 18, 20266 min · 1,280 wordsWhat the Wright laboratory is up to: contemporary corvid vocal research
While AI methods dominate headlines, careful behavioral work continues. The Wright lab at New Mexico is the cleanest contemporary example of corvid vocal resear…
May 18, 20266 min · 1,230 wordsHow smart are crows? A measured assessment
By every operational measure researchers have applied, very. By the unmeasured measure of 'how smart is smart?', the question keeps deflecting back at the asker…
May 18, 20267 min · 1,380 wordsThe unexpected richness of crow play behavior
Crows slide down snowbanks, drop sticks for the fun of catching them, ride wind currents for no reason at all. Play behavior is real in this species, harder to …
May 18, 20266 min · 1,240 wordsDo crows really bring gifts? Untangling the evidence
Internet folklore says crows leave shiny objects for humans who feed them. Some do. The pattern is more interesting and less consistent than the popular framing…
May 18, 20266 min · 1,240 wordsHow long do crows live? The longevity question, answered carefully
Wild crows usually live shorter lives than the popular framing suggests. The few that survive past childhood can live longer than most birds. The numbers tell a…
May 18, 20265 min · 1,080 wordsCarrion crow versus American crow: a comparative species note
They look almost identical, behave similarly enough to confuse generalist observers, and now show up in adjacent research papers using the same AI methodology. …
May 18, 20266 min · 1,230 wordsWhy corvids and parrots both vocal-learn: a convergent evolutionary puzzle
Vocal learning is rare across animals. Songbirds have it. Parrots have it. Crows have it to a degree. The evolutionary convergence is interesting — and it's why…
May 18, 20266 min · 1,240 wordsUrban versus rural crow behavior: what the acoustic record shows
City crows and country crows are the same species behaving differently. The behavioral divergence is now documented enough to be its own research subfield, and …
May 18, 20266 min · 1,230 wordsKaeli Swift and the crow funeral question
The crow-funeral behavioral pattern was first systematically documented by Kaeli Swift's doctoral work at the University of Washington. Understanding the method…
May 18, 20265 min · 1,080 wordsWest Nile virus and the American crow population collapse
Starting in 1999, an invasive virus killed millions of American crows. The species recovered in some regions and not others. Disease ecology is part of the spec…
May 18, 20266 min · 1,220 wordsCrow olfaction: the quiet sensory mode
Birds were assumed to have a poor sense of smell for most of the twentieth century. The assumption was wrong. Corvid olfaction is the quiet sensory mode that co…
May 18, 20265 min · 1,090 wordsClimate change and corvid range shifts: what the data shows
Bird ranges across North America have shifted measurably with warming since the 1970s. American crow is among the species that has expanded, which has consequen…
May 18, 20265 min · 1,090 wordsBernd Heinrich and the ravens of winter
Before contemporary corvid cognition was a research field, Bernd Heinrich spent winters in the Maine woods watching what ravens actually do. The methodology is …
May 18, 20265 min · 1,110 wordsAuguste von Bayern and the New Caledonian crow tool program
New Caledonian crows make tools. They modify the tools. They use sequences of tools. The systematic research on this is mostly Auguste von Bayern's program, and…
May 18, 20265 min · 1,100 wordsBird taxonomy in flux: what counts as a species
Bird species lists are being revised constantly as DNA evidence reshapes the field. The carrion crow, hooded crow, and northwestern crow have all had their spec…
May 18, 20265 min · 1,080 wordsListening to your local crows: a practical guide
The crows in your neighborhood are accessible bioacoustic subjects. With patience and the atlas as reference, you can develop genuine familiarity with their voc…
May 18, 20265 min · 1,050 wordsCrows in human cultures: a brief survey
Crows show up in human cultural traditions across nearly every continent. The cultural significance is older than ornithology and informs how people perceive th…
May 18, 20265 min · 1,090 wordsThe Bothell roost and other mega-aggregations
Some American crow communal roosts exceed 15,000 individuals overnight. The Bothell roost outside Seattle is one of the most-studied. Understanding why crows ag…
May 18, 20265 min · 1,070 wordsCrow play behavior: what it shows about cognition
Crows engage in behavior that looks like play — and the play interpretation has methodological support beyond casual observation. The cognitive significance is …
May 18, 20265 min · 1,070 wordsCrow sleep and roost acoustics
When the day's vocalizations end, the acoustic record continues. Crow sleep behavior is part of the species's daily cycle and has its own vocal patterns worth u…
May 18, 20265 min · 1,030 wordsCrows and humans: a coevolution sketch
American crow's ecological history with humans goes back further than people usually appreciate. The species hasn't just adapted to humans; humans and crows hav…
May 18, 20265 min · 1,070 words
Methods
How the AI works, in plain English.
18 essays
Self-supervised audio learning, explained for non-engineers
The training trick that made it possible to map a crow's vocal repertoire without anyone telling the AI what a crow sounds like.
May 18, 20267 min · 1,550 wordsBirdNET vs Perch 2.0 vs NatureLM-audio: the practical 2026 guide
Three audio foundation models, three jobs, three sets of tradeoffs. Which one to reach for, and why the answer isn't obvious.
May 18, 20268 min · 1,680 wordsWhy animal-language AI is harder than human-language AI
Same transformers, same self-supervised objectives, very different problem. The asymmetries that make animal communication research stubborn, even with foundati…
May 18, 20267 min · 1,490 wordsHow AI changed birdwatching: BirdNET, Merlin, and the citizen-science boom
The most successful AI bioacoustics deployment of the past five years isn't a research model — it's a free phone app that turned millions of casual users into p…
May 18, 20267 min · 1,430 wordsReading the vocal atlas: a beginner's guide to UMAP for bird sounds
What you're actually looking at when you see a 2-D scatter of bird vocalizations, and how to interpret what the geometry means.
May 18, 20266 min · 1,320 wordsSpectrograms decoded: what those squiggles actually show
Every modern bioacoustic paper includes a spectrogram. Most readers skim past them. Here's what they're actually showing and how to read one in thirty seconds.
May 18, 20266 min · 1,330 wordsVoxaboxen and the open infrastructure of modern bioacoustics
Behind the AI models is the tooling that makes them usable. Voxaboxen from Earth Species Project is the cleanest example of how open infrastructure shapes what …
May 18, 20266 min · 1,310 wordsWhat corvid researchers wish you understood about AI bioacoustics
The disconnect between popular framing and the working-scientist position is more interesting than either alone. Here's a translation of what the people inside …
May 18, 20266 min · 1,260 wordsWhat spectrograms can't tell you
The standard visualization for sound is also misleading in specific, predictable ways. Knowing what spectrograms hide is part of using them well.
May 18, 20265 min · 1,090 wordsThe Macaulay Library and the open-research tension
The world's largest archive of wildlife sound recordings is a Cornell Lab of Ornithology asset. Its data is partially open and partially proprietary. The tensio…
May 18, 20266 min · 1,200 wordsHow eBird became a bioacoustic engine
eBird started as a checklist platform for bird sightings. It became, almost incidentally, the largest active feeder of bioacoustic data on the planet. The trans…
May 18, 20265 min · 1,080 wordsAcoustic monitoring as a conservation tool
Passive acoustic monitoring is changing how conservation biology measures what's there. The methodology is now mature enough to be a standard tool, and the impl…
May 18, 20265 min · 1,110 wordsHow birds actually produce sound: the syrinx briefly explained
Birds don't have a larynx like mammals; they have a syrinx, a sound-producing organ unique to birds. The structural differences shape what bird vocalizations ca…
May 18, 20265 min · 1,060 wordsThe history of bird recording technology
Wildlife audio recording is now cheap, automatic, and ubiquitous. Getting there took a hundred years and a half-dozen technical revolutions. The history matters…
May 18, 20265 min · 1,090 wordsWhat good bioacoustic citizen science looks like
Citizen science contributes most of the bioacoustic data the modern AI revolution runs on. The practices that distinguish useful citizen-science contributions f…
May 18, 20265 min · 1,050 wordsHow to read a bioacoustics paper
Bioacoustics papers are mostly written for specialists. The structure is consistent enough that non-specialists can read them productively with a few orientatio…
May 18, 20265 min · 1,080 wordsWhat 'open source' actually means in bioacoustic AI
BirdNET is open source. Perch 2.0 is open source. NatureLM-audio is open source. The phrase covers more variation than the marketing implies, and the variation …
May 18, 20265 min · 1,100 wordsStowell 2022: the deep-learning bioacoustics review that grounds the field
Dan Stowell's 2022 review paper in PeerJ Computer Science is the most-cited methodological reference in modern bioacoustics. What the paper actually argued is w…
May 18, 20265 min · 1,080 words
Decoding
What we can see in the calls now — and what we can't.
5 essays
Do crows have dialects? What the evidence actually shows
Family groups carry measurable acoustic signatures. Whether those signatures function as cultural dialect — versus genetics or environment — is a harder questio…
May 18, 20267 min · 1,480 wordsThe Mates 2014 paper and why it still matters
Twelve years after publication, the canonical individual-identity finding in crow bioacoustics keeps showing up as the anchor in modern AI work. Here's what it …
May 18, 20266 min · 1,290 wordsWhat we can and cannot decode about crow communication, today
An audit of the claims AI bioacoustics can actually defend in 2026, and a separate audit of the claims it can't — useful for anyone trying to read popular cover…
May 18, 20267 min · 1,480 wordsWhat we don't know about crow communication
A deliberate inventory of the open questions. The honest version of the field map includes the empty quadrants as well as the populated ones.
May 18, 20266 min · 1,290 wordsPredator-specific alarm calls: do crows have words for hawks?
Vervet monkeys famously do. Some birds appear to. Whether American crows distinguish their alarm calls by predator type is one of the most-studied and least-res…
May 18, 20266 min · 1,280 words
Pipeline
From phone recording to interpretable map.
2 essays
From caw to cluster: the eight-stage pipeline explained
How a thirty-second phone recording of a crow becomes a point in a 1,536-dimensional vocal map — in eight discrete steps, with the failure modes named at each o…
May 18, 20269 min · 1,820 wordsHow to record crows: a field guide for citizen scientists
The rig that actually works, the ethical floor that constrains where you can point it, and what to do with your recordings once you have them.
May 18, 20267 min · 1,430 words
Frontier
What's coming, and the ethics underneath.
13 essays
Wearable bioacoustic loggers and the Demartsev revolution
Why the 2026 carrion crow paper changes what's knowable about wild bird communication, and what it means for American crow research.
May 18, 20267 min · 1,490 wordsProject CETI and what we learned from sperm whales
The most ambitious public animal-language AI project of the decade has been admirably honest about its constraints. The lessons generalize to corvid research.
May 18, 20267 min · 1,480 wordsPassive acoustic monitoring: what AI ears reveal about wild populations
The deployment that's quietly remaking conservation biology — and what it can and can't tell you about birds you can't see.
May 18, 20267 min · 1,440 wordsAcoustic indices: measuring ecosystem health through sound
If you can't survey every species in a habitat, can you summarize the soundscape itself into a number that says how healthy it is? Researchers are trying. Here'…
May 18, 20266 min · 1,290 wordsCrow population trends in North America: what acoustic monitoring shows
Are there more or fewer crows than ten years ago? The answer depends on where you ask. Long-term acoustic monitoring is starting to give cleaner numbers than vi…
May 18, 20266 min · 1,230 wordsHoneybee waggle dance: the gold standard for animal communication decoding
What honeybees taught us about what 'translating' animal communication actually requires — and why nothing else has matched the standard since.
May 18, 20266 min · 1,280 wordsAlex the parrot and the comparative-cognition tradition
Irene Pepperberg's thirty-year study of one African grey parrot named Alex remains the most-cited animal cognition research in popular science. The honest versi…
May 18, 20266 min · 1,180 wordsProject CETI and the sperm whale codas
Project CETI is the most-ambitious cetacean-language research program ever launched. Its progress and limits are the closest analog to what crow-focused AI rese…
May 18, 20266 min · 1,190 wordsNicola Clayton and the scrub jay memory studies
Episodic-like memory in birds was considered impossible until Nicola Clayton's work on western scrub-jays. The findings reshaped what cognitive science thought …
May 18, 20265 min · 1,100 wordsThe avian pallium and the bird brain revolution
For most of the twentieth century, neuroscience thought bird brains lacked the cortical architecture that supports mammalian cognition. A 2020 paper showed the …
May 18, 20265 min · 1,100 wordsConvergent intelligence across distant lineages
Crows, dolphins, octopuses, and elephants all show signs of high intelligence. Their last common ancestor lived hundreds of millions of years ago. The convergen…
May 18, 20266 min · 1,180 wordsIndigenous knowledge and Western corvid science
Indigenous Pacific Northwest knowledge of raven and crow behavior predates Western ornithology by millennia. The relationship between these knowledge systems ma…
May 18, 20265 min · 1,080 wordsVervet monkey alarm calls: the other gold-standard study
Before the AI revolution, the Cheney-Seyfarth vervet monkey work was the closest thing animal communication research had to demonstrated semantic content. Under…
May 18, 20266 min · 1,180 words
Ethics
What we will and won't claim, and why.
5 essays
The ethics of playback experiments in bioacoustic research
What ethical bioacoustic research actually constrains, why those constraints exist, and what they mean for how aggressively a public site like CrowLingo can int…
May 18, 20266 min · 1,280 wordsThe replication problem in animal cognition
Several high-profile findings in animal cognition haven't replicated as well as the original publications suggested. The pattern is important context for any AI…
May 18, 20265 min · 1,110 wordsWhat machine learning probably can't decode, no matter how much audio we feed it
Some claims about AI animal-communication decoding will probably never be supported, regardless of how much data we have or how much compute we apply. Understan…
May 18, 20265 min · 1,100 wordsWhat the CrowLingo atlas is not
Boundary-setting matters for any reference work. Being honest about what the atlas is not is part of being credible about what it is.
May 18, 20265 min · 1,090 wordsCaptive corvid research: what the ethical debate involves
Most of what's been established about corvid cognition comes from captive studies. The ethical issues around captive corvid research are real, contested, and wo…
May 18, 20265 min · 1,090 words