Decoding Your Dog's Barks

Canine communication is more sophisticated than it sounds.

the year of the puppy

In her new book ‘The Year of the Puppy: How Dogs Become Themselves,’ Alexandra Horowitz (a professor of canine cognition at Barnard and Columbia University) explores canine development through moments in her own puppy’s first year of life. 

In the excerpt below, 5-month-old Quid is settling into becoming a certified “teenager,” with all the typically destructive habits that entails. Amidst the din of Quid’s seemingly unrelenting howls, Horowitz marvels at the sophistication of her young dog’s developing vocabulary. 

Barks are considered “noisy” in the language of audio, where “noisy” means there is a lot of broadband sound—full of different frequencies but without a clear tone or note. To amateur ears, they are just noise. But “noisy” does not mean meaningless. Barking is communication, and our reaction to this noise—to shush it or punish it—is more or less like slapping a child for talking unbidden. What are dogs talking about?

Barks convey a lot of emotional information—there are barks to induce play and barks barked when alone; there are barks to announce a stranger and barks used as a request—and also have information about the age and sex of the barker, for anyone listening carefully. Indeed, while most people are unenthusiastic about any barks their pups make, it is notable that barking is right smack in the frequency range of human speech—and may be an attempt to talk with us in a way we could understand. Wolves, dogs’ nearest relatives, rarely bark. Dog barks are certainly pitched to get our attention. Some authors see a homology between baby crying and some kinds of high-pitched barking: both are considered “annoying ” and trigger a response from people; especially afflicted are young adults of childbearing age.

The dog person’s concern—will she bark all the time?—is a real one, given the societal view on barking. And yet that quiet puppy in the other room: watch out. With a puppy, as with a baby, silence is desired and yet disquieting. One of Quid’s early lesson plans for us had her being perfectly quiet. After ten minutes we realized that we could neither hear nor see the puppy. We found her in O.’s room, methodically disemboweling a Sharpie. Often, quietness indicates miscreance.

Many people’s instinct when a dog barks loudly is to yell, delivering a stern No! or Stop it! But when we do this, we are responding with a kind of bark right back at the dog. It is funny, and a little sad, that we do this, since their barks are not angry yells, but messages with content. Now, the No! might work for a moment (what is she barking about? their cocked heads seem to say), but this is not a long-term way of getting a dog not to bark. The dog’s behavior indicates that they do not understand “no” as meaning “no barking, please,” but rather take it as general opprobrium. In the face of barking, the truly savvy keep quiet.

Babies appear to begin to recognize their own names by four and a half months; Quiddity knew both her name and her most common nickname within, conservatively, two days of hearing them.

This is not to say that dogs do not have a sense of what “no” means: any dog who has been on the receiving end of a human’s angry negation surely does. They understand the anger of it—if not exactly which thing they are doing we might be angry about. And it is non obvious even to humans: it is eight or so months before a child understands the word “no”; a puppy raised among humans will have picked up the tone of it five months earlier.

A comparison of child and puppy language comprehension in their first year is informative—about what we assume “comprehension” involves in each case. Babies appear to begin to recognize their own names by four and a half months; Quiddity knew both her name and her most common nickname within, conservatively, two days of hearing them. I should clarify what researchers mean by “recognize” or “know.” Those four-and-a-half-month-old children turned and listened longer to their own names than to those of others— e.g., a child called Emily listened three to four seconds (about 25 percent) longer to the sound of her own name than to someone saying the names Christopher, Marissa, or Samantha. When I say Quid “knows” her name, I mean: she looks at me when I say it, stops moving when I say it, and (a big ask) comes closer to me when I say it. And she fails to look at all when I say “Samantha.”

While dogs will never become fluent language users, we weirdly downplay their own communications. Children, it is widely touted, have a few words in their arsenal—in addition to the early mama or dada—by twelve months; Quid, at five months, will never say mama to me, but she has different whines and barks for needing to go outside, wanting attention, disturbance at the cat’s taking her bed, and requesting more tickling-of-ears; she knows the spatiotemporal difference between “wait” and “stay”; she knows not just her name but the names of Finn and Upton. She lies “down” when I ask her to get “down off the bed”—not so much a misunderstanding, I would say, as a wry commentary on the equivocation inherent in our requests to dogs.

When a researcher (or parent) gushes over the creativity of a two-year-old child’s ability to put together words in a novel way, to express a new meaning—water plus bird for a duck, say—I smile, nod my head agreeably, and remember the moment that Quid put together two actions—walking over to the stairs to our bedroom, and looking at us—to express a new meaning: I need to pee. I’m just sayin’.

In a parallel universe, where everything is identical but for our urgent need of a skilled locator of and alerter to squirrels, small dogs, and passing cars, Quid would be a hero. I feel a little sympathy for her that she, so clearly a master bark-alerter, was destined to live in a family where that skill would not be celebrated.

Barks, whines, and growls aside, most of dogs’ talk is with actions. None of these action words has been taught to her. A paw on my hand: a request to keep petting her. Turning her head away: a refusal, expression of distaste or disgust. Head resting on my lap: somewhere between possession and affection.

There are canid-sapiens overlaps in language learning too, though. Dogs have the same problems with reference that children learning language do. Imagine a child walking with her parents, who, spotting a bird in the distance, directs her attention to it: “Look at that bird!”— or even, “Look at that woodpecker!” Assuming that the child has not already learned about birds, this is a significantly confusing statement. There are any number of things that might be meant by “bird”: the trees, leaves, sky, clouds, something between the tree and the parents’ gaze, or the bird itself. It could be the redness on the bird, the fact of flight, the small thing, the big thing. It could be the heat on their face, the brightness of the light, the smell of summer.

Eventually, kids do get it. We have all eventually gotten it “within the first few years of life,” one paper exults. “How do [we] accomplish this task?” It is indeed quite a feat of cognition—but the caveat here is important for us to remember. It takes years of life for a species being explicitly taught, talked to near constantly, and already genetically predisposed to learn languages, to understand references like “Look at that bird.” And yet with puppies, we say “no” to their barks and expect that they will understand our meaning out of the gate. 

From The Year of the Puppy by Alexandra Horowitz, published by Viking, an imprint of Penguin Publishing Group, a division of Penguin Random House, LLC. Copyright © 2022 by Alexandra Horowitz.

Alexandra Horowitz

Alexandra Horowitz observes dogs for a living. Her research began more than two decades ago, studying dogs at play, and continues today at her Dog Cognition Lab at Barnard College. She is the author of Inside of a Dog: What Dogs See, Smell, and Know and three other books: On Looking; Being a Dog; and Our Dogs, Ourselves. She lives with her family of Homo sapiensCanis familiaris, and Felis catus in New York City.


Please note: Lemonade articles and other editorial content are meant for educational purposes only, and should not be relied upon instead of professional legal, insurance or financial advice. The content of these educational articles does not alter the terms, conditions, exclusions, or limitations of policies issued by Lemonade, which differ according to your state of residence. While we regularly review previously published content to ensure it is accurate and up-to-date, there may be instances in which legal conditions or policy details have changed since publication. Any hypothetical examples used in Lemonade editorial content are purely expositional. Hypothetical examples do not alter or bind Lemonade to any application of your insurance policy to the particular facts and circumstances of any actual claim.