Do Animals Dream?
A new book argues that dreams are a portal to animal consciousness.
This article was featured in One Story to Read Today, a newsletter in which our editors recommend a single must-read from The Atlantic, Monday through Friday. Sign up for it here.
The octopus is still, suspended upside down on the glass wall of her tank, just below water level. Her eyes are shut, her eight arms knotted snug behind her head, her pale-gray body gently pulsing. She is sleeping. Then: a twitch, a sudden creep of yellow, and her genus’s famously mutable skin plunges deep violet. The arms writhe in place, the head bobs, and the skin morphs again, this time to a spiky, mottled green serving as camouflage. After a minute, the flashes of color calm. The dream, apparently, is over.
The video of this octopus, Heidi, aired in 2019 as part of an episode of PBS’s Nature documentary series, and subsequently went viral. In it, the narrator, a marine biologist, describes how Heidi’s changing color patterns may indicate what’s happening in her mind as she sleeps. He imagines her dream: She spots a crab, pursues and kills it, then leaps from the ocean floor—that’s when she turns dark—to a quieter location, where she disguises herself to consume her meal. The spectacle is fascinating, in part because Heidi’s visual displays are beautiful, but also because this quirk of the octopus body makes visible a drama that might otherwise play out only internally. Heidi’s skin tells a story. As human viewers, we can imagine her anticipation in the catch, her moment of vulnerability as she jets up from the bottom, and her sense of safety and satisfaction as she settles down to eat.
For many researchers, the widespread enchantment with this video was a misapprehension—how can we really know that the octopus was dreaming? For David M. Peña-Guzmán, a philosopher of science at San Francisco State University, though, the spectacle was an invitation for wonder. What surprises might animal minds be capable of? “Beneath this procession of color and texture, what was Heidi herself thinking or feeling?” he writes in his intriguing book, When Animals Dream: The Hidden World of Animal Consciousness. The book joins a recent run of popular science writing that delves into the underappreciated complexities of animal emotion and cognition. While most writing on animal cognition has focused on their waking lives, Peña-Guzmán turns his attention to what their brains might be doing in sleep. Dreams, he proposes, are more than mechanistic sparks of neural activity—they are evidence for the nebulous workings of consciousness itself. Even while stubbornly resisting close analogies with the human mind, scientists have always speculated about the inner worlds of animals; Peña-Guzmán offers a novel, and poetic, way in.
The first thing that consciousness and sleep have in common is that they are scientific mysteries. Though almost all animals sleep, what happens in the brain when it’s powered down is still not fully understood. In humans, dream states occur during REM sleep, when the brain lights up with activity. Recent research has revealed that an unexpected proportion of the animal kingdom falls into an analogous phase of active sleep: mammals, birds, fish, some reptiles, cephalopods like Heidi, even fruit flies. Among the thousands of mammal species, the only ones that lack at least the capacity to dream are dolphins and whales. (Because of the challenges of sleeping underwater, they slumber half a brain at a time.)
How can anyone know what is going on inside a sleeping animal? First, there are external clues. The behaviors of many species in sleep correspond to human behavior when dreaming, including rapid eye movement, muscle twitching, and involuntary vocalizations. Peña-Guzmán compiles some startling anecdotes of captive animals. Washoe, a chimpanzee at Central Washington University, was observed signing “coffee” in ASL while asleep. (He liked drinking coffee and would sometimes ask researchers to give him a cup.) In Kenya, a young elephant named Ndume whose family had been slaughtered by poachers was known to wake up crying with what his handlers thought were night terrors.
Peña-Guzmán surveys harder data as well, taken from ingeniously designed lab experiments. Neurological and electrophysiological measurements gathered from the brains of a variety of species have shown that they see, hear, and feel specific, identifiable scenarios in their sleep that appear to be direct replays of experiences they’ve had in waking life. The zebra finch, for example, is an Australian species of bird that learns unique songs passed down from its family. When biologists mapped the region of the brain controlling birdsong in juvenile finches while they were awake and singing, and then while they were asleep, they found that the neurological patterns matched precisely, note for note. The young finches, it seems, were practicing in their sleep.
To avoid anthropomorphization, animal-sleep researchers tend to resist the conclusion that any animal behavior, however much it looks like dreaming, actually is dreaming. Instead, Peña-Guzmán observes, study authors favor the clinical language “oneiric behavior” and “mental replay” when writing about animal subjects, putting a linguistic buffer between their findings and activities reserved for the human mind. This reluctance to call dreams what they are is, to him, an overblown concern, with roots in speciesist attitudes that emerged in the life sciences in the 20th century. Although Charles Darwin’s contemporaries were keenly interested in dreams, biologists later reversed course to adopt a largely skeptical view of animal cognition. That view still persists, particularly in animal-sleep research. “In an act of collective self-delusion,” Peña-Guzmán writes, “we convinced ourselves that [animals] could not possibly have what we have: a meaningful inner world.”
Peña-Guzmán’s survey of dream research makes a strong case for the fact that some animals do dream, though as a philosopher, not a biologist, he ultimately poses more questions than he answers. How many animals dream, and which? His book doesn’t attempt to draw lines between species that dream and those that may not. (Elephants and chickens, sure, but what about bees? Sea sponges?) Instead, the book enters the far more slippery terrain of trying to build a unique case for animal consciousness. A mind that dreams is, Peña-Guzmán argues, necessarily a conscious mind.
This is not an uncontentious statement. Consciousness presents a particularly knotty philosophical puzzle, and there is no easy consensus on how to define it, let alone prove its presence. The philosopher Thomas Nagel’s 1974 idiom, for all its casualness and imprecision, is still often cited: There is something it’s like to be conscious. There is a feeling, an experience, of being you—a feeling that a frying pan, a raindrop, or a lump of sugar does not have. Peña-Guzmán defines consciousness in its broadest sense as having awareness. You are aware of the presence of your body in the world and the moving current of your thoughts. Another way to put it: Somewhere inside, a light is on. Though, the light may not flick on all at once—many philosophers conceive of consciousness as forming progressively or in stages, a light operating by dimmer rather than an on-off switch.
In When Animals Dream, Peña-Guzmán starts his discussion of consciousness with the concept of subjectivity, which is the idea that one experiences life as a unified self, an interior “me” positioned against the world. This status is not reserved for humans. All animals that are mobile, whether flatworms or foxes, must somehow distinguish their bodies from the exterior environment. With the self comes, too, a sense of temporality (that things happen to me in linear sequence) and agency (I initiate new action in response). Here—with subjecthood, chronology, choice—we start to have the conditions for narrative. And narrative is what dreams are made of.
One of the strengths of Peña-Guzmán’s book is its evocative forays into literary terrain. Our dreams, after all, are like stories. In modern philosophy, meaning-making has been tethered closely to language, which is seen as a unique ability of Homo sapiens—but in dreams, meaning arrives outside of linguistic representation. If it seems odd to consider that a mole, a raven, or even a butterfly could dream, perhaps this resistance is due to a long-standing association between dreams and imagination, creativity, and storytelling—qualities that humans typically assume are what differentiate our species from all other animals. Our dreams, as Freud suggested, are richly symbolic. They are built from memory and desire, and laden with emotion. Neurologically, to remember that something happened to you and to project scenarios in the future are very similar cognitive functions. If animals can dream, Peña-Guzmán writes, then perhaps they can also daydream, or imagine.
At this point one might object: How can we know that animal dreams are meaningful to them, and not merely jumbles of random sensation? Animals cannot keep journals or recount their dreams to researchers. (Most cannot, anyhow—a gorilla named Michael whose mother was butchered by traffickers in the forests of Cameroon would have recurring nightmares. He described them in ASL, signing, “Bad people kill gorillas.”) Skeptics of animal cognition often describe animal actions behavioristically, as learned reactions to a stimulus. Could the same be going on in sleep? Peña-Guzmán rejects this interpretation by pointing out that a dreaming animal is not reacting to anything. A sleeping rat navigates a dream-conjured maze while its body lies still in its cage. The animal is responding only to the world that a mind—its own mind—has invented.
For the nonphilosophers among us, all this debate over consciousness might seem a bit silly. I look at my dog, I call her name, and she looks back at me. Sometimes she comes when I call, bounding over in expectation of a treat or affection; sometimes she chooses to ignore me and go about her doggy ways. That she has a mind of her own seems clear enough.
And yet most of us, pet owners included, tend not to treat animals as fellow minds. (How often have I heard the tired claim from trainers that dogs are driven purely by pack instinct.) For billions of humans on this Earth, animals are food, labor, or material. To recognize that animals have consciousness would be a first step toward considering them in possession of inviolable rights alongside our own. A court in New York earlier this year declined to grant personhood to Happy, an elephant at the Bronx Zoo, despite one judge’s admission that Happy is “an intelligent, autonomous being who should be treated with respect and dignity.” But we needn’t equate animals with humans to recognize that their subjective experience can be as vibrant as our own. Understanding that they might feel, think, and dream makes it harder to imagine 800 million years of animal life as wholly devoid of mind—a house empty and dark, with the lights switched off.
What might animal dreams look like? Peña-Guzmán leaves this question, too, unanswered. But he invites us to try to imagine. Surely the dreams of animals, even if filled with pleasure and pain like our own, don’t share the same textures. Rather than a world of image and language, theirs might be “worlds without human contours,” made of scent, sonar, the push and pull of water, of heat and cold. Though they are not ours, they are no less meaningful.
When you buy a book using a link on this page, we receive a commission. Thank you for supporting The Atlantic.