Skip to content
NatureHQ

species

Common raven

Corvus corax Linnaeus, 1758

The common raven is the largest songbird in the world and a separate species from the crow, with its own body of cognitive research on planning, deception and social knowledge.

Ravens are routinely folded into "crows" in popular writing, which does both birds a disservice. They are larger, longer-lived, wedge-tailed, and their research literature is different: ravens dominate work on caching, deception and future planning, where crows dominate work on face recognition and number. Two results stand out. Ravens adjust how carefully they hide food to the possibility of being watched, even with no watcher visible — which is hard to explain as reading another bird's gaze. And they will turn down an immediate snack to keep a tool they can use hours later, in tasks that ravens never encounter in the wild.

Developed record · 84% complete · reviewed 2026-08-09

What this page covers

One species, Corvus corax, with a huge range across the northern hemisphere.

Often confused with: Corvus corone (carrion crow — smaller, squared tail, higher call); Corvus brachyrhynchos (American crow — noticeably smaller, squared tail)

Quick facts

Species
Corvus corax
Size
Wingspan commonly 115–150 cm — much larger than any crow
Tail
Wedge-shaped in flight; a crow's is squared
Caching
Adjusts hiding behaviour to the possibility of being observed

Raven or crow?

The question people actually arrive with.

A raven is a different, much larger species — and neither is smarter than the other on the evidence

Established

Specialists would state this without hedging. Multiple independent lines of evidence agree.

Corvus corax is a distinct species roughly twice the mass of Corvus brachyrhynchos or Corvus corone, with a wedge-shaped tail, heavier bill, shaggy throat hackles and a deep croaking call rather than a caw. The two do not interbreed. Comparative cognition studies use different paradigms in each species and do not support ranking one above the other.

Who this applies to
common ravens against American and carrion crows
Studied in
Corvus corax, Corvus brachyrhynchos, Corvus corone

You may have heard

“Ravens are just big crows, and they are the smartest birds”

They are separate species that do not interbreed, and the intelligence ranking is an artefact of which experiments happened to be run on which bird. Nobody has given ravens and crows the same task and compared them properly.

Why we rate it this way, and what the caveats are
EstablishedHigh confidence

Species limits and morphology are settled taxonomy. The refusal to rank intelligence reflects the state of the comparative literature, which has not run matched tasks across both.

Caveats

  • Size overlap exists at the extremes; a large crow and a small raven can be hard to separate in isolation.
  • The famous corvid cognition results belong to several different species and do not transfer between them.
  • "Crow" covers many species; "raven" in common use may mean several Corvus species too.

Still unanswered

  • Would matched tasks across corvid species reveal real cognitive differences?
  • How much of the apparent difference reflects which species were convenient to study?

Last reviewed 2026-08-10

The evidence (3 studies)
Telling them apart in the field
FeatureCommon ravenCrow
SizeBuzzard-sized; roughly twice a crow’s massPigeon-to-small-gull sized
Tail in flightWedge-shaped, longest feathers in the middleSquared or fan-shaped
BillHeavy and curved, with bristles over the baseStraighter and slimmer
ThroatShaggy hackles, visible when callingSmooth
VoiceDeep croak, knocking and bell-like notesSharp caw
FlightSoars and rolls; often glidesSteady flapping, rarely soars

Voice is the most reliable single cue and the one hardest to convey in writing. A raven croaks from deep in the chest; a crow caws. If a large black bird soars on flat wings and rolls upside down for a moment, it is a raven — crows do not do that.

The intelligence ranking people ask about does not have an answer. Ravens and crows have been given different tasks in different laboratories, and nobody has run a matched comparison. "Ravens are smarter" reflects which experiments happened to be done on which bird.

Words used here
Hackles
Elongated throat feathers a raven can raise, giving a shaggy-bearded look when it calls.

What ravens eat, and who they follow

A young raven that finds a carcass yells for others — to overwhelm the pair guarding it

Well supported

Good evidence backs this, though some details remain open.

Juvenile common ravens discovering a carcass within a resident pair’s territory produce loud yell calls that recruit unrelated juveniles within one to two days. Territorial adults finding the same food stay silent. Recruitment allows non-territorial birds to gain access to a resource a resident pair would otherwise defend successfully.

Who this applies to
common ravens in temperate North American woodland in winter
Studied in
Corvus corax

You may have heard

“Ravens call their friends to share food”

Sharing is the outcome, not the reason. A juvenile calls because a crowd can overwhelm the resident pair that would otherwise drive it off, and the same bird stops calling once it holds a territory of its own — which is the observation that separates recruitment from generosity.

Why we rate it this way, and what the caveats are
Well supportedHigh confidence

Repeated baited-carcass observation across winters, with the crucial contrast between yelling juveniles and silent territorial adults, which rules out a simple "call when you find food" rule.

Caveats

  • Documented at artificially provided carcasses in one region.
  • Whether the yell carries information beyond location is unresolved.
  • Behaviour changes once a bird holds a territory, so this is a life-stage pattern rather than a species trait.

Still unanswered

  • Do recruiting calls differ acoustically from other loud calls?
  • How do juveniles assess whether a carcass is defended before calling?

Last reviewed 2026-08-10

The evidence (2 studies)

A raven will eat more or less anything, but the food that shapes its life is carrion — and specifically large carcasses, which it cannot open by itself. A raven’s bill will not get through the hide of a dead moose. Wolves will.

That dependence explains a great deal of raven behaviour, including following wolf packs, attending hunters, and the loud yelling that brings a crowd of young birds to a carcass within a day. The yelling is not sharing. A juvenile that finds food inside a resident pair’s territory would be driven off alone; twenty of them cannot be.

  • Carrion of all sizes, from roadkill to whale strandings.
  • Small mammals, eggs, nestlings, amphibians, insects and grain.
  • Human refuse, which supports high raven densities in some regions.
  • Cached surplus, recovered later — the behaviour behind the raven memory literature.

Where this connects

  • Wolves

    The animal that opens the carcass

  • Crows

    A different diet and a different bird

Can a raven talk?

Ravens can copy human words. There is no evidence they mean anything by them

Well supported

Good evidence backs this, though some details remain open.

Common ravens are vocal learners and can reproduce human speech and environmental sounds, more readily in captivity than in the wild. No evidence demonstrates that mimicked words are used referentially. Across songbirds generally, a function for vocal mimicry has been demonstrated in few species, and a by-product of general vocal learning remains the untested default explanation in most.

Who this applies to
common ravens, principally hand-reared captive birds
Studied in
Corvus corax

You may have heard

“Ravens can talk”

They can produce the sounds, often more clearly than a parrot. Talking implies using words to mean things, and nothing in the raven literature demonstrates that. The capacity follows from being a vocal learner; the content is whatever the bird has heard repeatedly.

Why we rate it this way, and what the caveats are
Well supportedModerate confidence

That ravens copy speech is not in dispute. The negative half — no demonstrated meaning — rests on absence of evidence across a literature that has largely not looked, so it is stated as unsupported rather than refuted.

How far it can be extended

Vocal mimicry of human speech occurs across corvids and parrots; the absence of demonstrated referential use is a general finding rather than a raven-specific one.

Caveats

  • Captive hand-reared birds mimic far more than wild ones; this is largely a captivity phenomenon.
  • Absence of demonstrated meaning is not proof of absence — the experiments have mostly not been done.
  • Mimicry of mechanical and animal sounds is more common in the wild than mimicry of speech.

Still unanswered

  • Does wild raven mimicry serve any function, or is it a by-product of vocal learning?
  • Do ravens use mimicked sounds differently from their own call types?

Last reviewed 2026-08-10

The evidence (3 studies)

  • Supports · primary

    Vocal mimicry in songbirds

    Kelley et al., 2008 · Animal Behaviour

    Reviews mimicry across songbirds and finds function demonstrated in few species, with a by-product explanation rarely tested against.

  • Supports · supporting

    Common Raven (Corvus corax)

    Boarman and Heinrich, 1999 · The Birds of North America

    Documents the vocal repertoire and mimetic capacity of the species.

  • Qualifies · contextual

    Grey parrot numerical competence: a review

    Pepperberg, 2006 · Animal Cognition

    The one line of work in which a bird was trained to use words referentially — showing what a demonstration would have to look like, and that nothing comparable exists for ravens.

Ravens are vocal learners with an unusually large repertoire, and captive hand-reared birds can reproduce human speech clearly — often more clearly than a parrot, because the sound is produced differently and comes out lower. Wild ravens mimic too, more often copying other birds, mammals and mechanical noises than words.

What has never been shown is that any of it means anything to the bird. Across songbirds generally, a function for mimicry has been demonstrated in very few species, and the plain explanation — a by-product of being a vocal learner, copying whatever is heard often — is rarely tested against. NatureHQ treats "ravens can talk" as a true statement about sound and an unsupported one about language.

A raven’s natural repertoire runs to dozens of call types, including knocking and bell-like notes that sound nothing like a bird.

Words used here
Vocal learner
An animal that modifies the sounds it makes based on what it hears. Rare — songbirds, parrots, hummingbirds, cetaceans, bats, seals, elephants and humans.

Pairs, nests and how long they live

Ravens pair for life and defend a territory year-round. The nest is a large stick structure on a cliff ledge, tall tree or increasingly a pylon or building, reused and added to over years. Three to seven eggs is typical, incubated by the female while the male provisions her.

Young stay with their parents for several weeks after fledging and then join the non-breeding flocks, where they may spend years before acquiring a territory. Almost every striking raven behaviour in the literature — recruitment yelling, the boldest play, the caching experiments — belongs to that floating juvenile stage rather than to settled adults.

Typical figures
MeasureCommon raven
Lifespan in the wildCommonly 10–15 years; some considerably longer
Lifespan in captivityFrequently past 30; records beyond 40
Clutch3–7 eggs, once a year
Pair bondLong-term, with a defended territory held year-round
MigrationNone — resident, though juveniles wander widely

Keeping one is a different question from whether it is possible. In most of Europe and North America native corvids are legally protected and cannot be taken or kept without specific authorisation, and a bird this intelligent and long-lived is a poor fit for a household in any case.

Rolling, sliding and harassing eagles

Ravens roll in flight, slide down snow and play with objects — mostly the young ones

Well supported

Good evidence backs this, though some details remain open.

Common ravens engage in aerial manoeuvres, object manipulation and substrate sliding that meet behavioural criteria for play. It is markedly more frequent in non-breeding sub-adults than in territorial adults. One captive study found play bouts synchronising between individuals beyond chance and proposed emotional contagion as an explanation.

Who this applies to
common ravens, principally non-breeding sub-adults
Studied in
Corvus corax
Why we rate it this way, and what the caveats are
Well supportedModerate confidence

Play is widely and repeatedly documented in this species by independent observers. Confidence is held at moderate because the interesting further claim — that synchronised play reflects emotional contagion — rests on one small captive group and has an equally good non-social explanation.

Caveats

  • Birds playing at the same time may be responding to the same cue rather than to each other.
  • The synchrony evidence comes from one small captive group of sub-adults.
  • Emotional state is inferred from behaviour and not independently measured.

Still unanswered

  • Does play serve a function — motor practice, social bonding — or is it a by-product of a long juvenile period?
  • Why is it so much rarer in territorial adults?

Last reviewed 2026-08-10

The evidence (2 studies)

  • Partly supports · primary

    Sub-adult ravens synchronize their play: a case of emotional contagion?

    Osvath and Sima, 2014 · Animal Behavior and Cognition

    Recorded synchronised object play in sub-adults and proposed emotional contagion; the synchrony is the finding, the contagion is the interpretation.

  • Supports · supporting

    Common Raven (Corvus corax)

    Boarman and Heinrich, 1999 · The Birds of North America

    Species account documenting play across contexts and its concentration in non-breeding birds.

Ravens roll upside down in flight, drop and catch objects, slide down snowbanks and hang from branches by one foot. Object play is well documented and unusually persistent in sub-adults, and one captive study found play bouts synchronising between birds beyond chance — proposed as emotional contagion, which NatureHQ records as an interpretation rather than a finding, since birds responding to the same cue would look identical.

Harassing eagles and other raptors is a different behaviour with a plainer explanation. Mobbing drives a predator out of the area and can also dislodge food; ravens are frequent kleptoparasites, and an eagle that lets go of a fish has solved a problem for the raven.

The shiny-object reputation is largely folklore. Corvids cache food and sometimes objects, and captive birds handle novel items, but experimental work on wild-type corvids has generally found them wary of shiny things rather than drawn to them.

Words used here
Mobbing
Smaller birds harassing a predator in a group to drive it away. Noisy, coordinated and usually effective.
Kleptoparasitism
Stealing food another animal has caught or found.

Not a crow

Ravens and crows are both in the genus Corvus, and that is roughly where the similarity ends for the purpose of reading research. A raven is about twice the weight of a carrion crow, with a heavier bill, shaggy throat feathers and a wedge-shaped tail. In flight the tail shape is the reliable field mark.

  • Tail: wedge-shaped in a raven, squared in a crow.
  • Call: a deep, resonant croak, against a crow's flatter caw.
  • Throat: shaggy hackles, visible when calling.
  • Flight: more soaring, with longer glides.

Related

  • Crows

    And which crow each famous finding belongs to

Hiding food from an audience that might not be there

Ravens hide food more carefully when they might be watched — even with no one in sight

Well supported

Good evidence backs this, though some details remain open.

Common ravens (Corvus corax) increased cache protection when a peephole into an adjacent room was open and auditory cues suggested a competitor, and not when the peephole was closed.

Who this applies to
common ravens, in captivity
Studied in
Corvus corax
Why we rate it this way, and what the caveats are
Well supportedModerate confidence

The peephole design removes the standard objection that the bird is reading a competitor's gaze, which makes it unusually strong. Whether the birds represent another's perspective or follow a learned rule about open peepholes remains genuinely open.

Caveats

  • Ten captive birds with extensive experience of caching experiments.
  • A learned association between open peepholes and risk could produce the same behaviour.

Still unanswered

  • Does this extend to wild ravens, where caching competition is constant?

Last reviewed 2026-08-09

The evidence (2 studies)

How we know

The peephole test

Does a raven hiding food understand that another bird might be able to see it — even when it cannot see that bird?

Ravens hid food in a room fitted with a small peephole into an adjoining room. In the key condition the peephole was open and the bird could hear sounds suggesting another raven next door, but could not see it. In the control, the peephole was closed. The birds had previously had the chance to use the peephole themselves, so they knew what it was for.

What happened

With the peephole open and a competitor apparently next door, ravens cached faster and guarded their hiding places more. With it closed, they behaved as though they were alone.

What it shows

The raven adjusted its behaviour to the *possibility* of being seen, without any visible watcher to react to. That is difficult to explain as simply responding to another animal's stare, which is the usual objection to this kind of experiment.

What it does not show

It does not establish that ravens hold beliefs about what other individuals believe. A learned rule — open peephole means risk — could produce the same behaviour, and the study cannot fully separate the two.

The controls — what makes this evidence rather than a story
  • A closed peephole with the same sounds tested whether noise alone was enough.
  • Birds had prior experience looking through the peephole, so they knew it allowed a view.
  • No competitor was actually present, removing gaze and body-language cues entirely.

From Ravens attribute visual access to unseen competitors

Words used here
Caching
Hiding food to retrieve later. Ravens do it constantly, and steal from each other constantly.

Planning for tomorrow

Ravens will turn down a treat now to keep a tool they can use tomorrow

Emerging evidence

Real findings exist, but too few or too recent to be settled.

Common ravens selected and retained a tool or a bartering token over an immediate food reward, and used it successfully when the opportunity recurred up to seventeen hours later, in tasks with no natural analogue.

Who this applies to
common ravens, in captivity
Studied in
Corvus corax
Why we rate it this way, and what the caveats are
Emerging evidenceModerate confidence

A strong and carefully designed result, but from five hand-raised birds and without independent replication. The use of tasks ravens never perform naturally is what makes it interesting and also what makes the training history hard to discount.

Caveats

  • Five hand-raised birds, extensively trained.
  • Neither tool use nor bartering occurs naturally in ravens, which cuts both ways.
  • Comparison with great apes involves different procedures and is not like-for-like.

Still unanswered

  • Does the ability appear without laboratory training?

Last reviewed 2026-08-09

The evidence (1 study)

The design matters here. Storing food for later could be a specialised instinct, so the researchers used tasks ravens have no instinct for — using a tool on an apparatus, and swapping a token with a human. Choosing a useless object over immediate food, for a payoff hours away, is difficult to explain without something future-directed going on.

Claims about this, checked

Things people have heard, and what the evidence actually supports.

The research behind this page

11 studies, newest first. Each one has a page explaining what it found and what it could not show.

You may also find this interesting

What this page is still missing

NatureHQ publishes its own gaps. This record is at 84% completeness against what we would call a finished subject.

  • no research from the last few years is attached — check for newer work
  • more experiments could be explained in plain English
  • Ravens have a rich literature on social relationships and deception that this record does not yet cover.
  • Cultural and folkloric significance is entirely absent.

Last reviewed 2026-08-09 · 8 claims · 73 search questions answered on this page