Behavior & Training
Print|Email|Text Size: ||
Dog Speak: The Sounds of Dogs
More than just noise


A friend suggested that one of the reasons we love dogs so much is that they can’t talk back. But I wonder whether that’s true. Sure, a dog won’t tell you, “You really shouldn’t have that second cookie,” but does that mean dogs are not talking back?

Dogs are anything but mute, and while we usually focus on wagging tails and beguiling eyes, vocalizations—among them, barks and growls—provide us with another window into dogs’ everyday experiences.

Social species are known to be much noisier than animals who lead solitary lives. Snow leopards roam the mountains of central Asia in near silence, but groups of monkeys do a lot of highvolume chattering. So, given that dogs and their wild progenitor, the wolf, are über-social, it’s no surprise that both produce a wide range of vocalizations: they bark, whine, whimper, howl, huff, growl, yelp and yip (among other nuanced sounds). From the earliest moments of their lives, dogs and their canid relatives produce tonal yelps and whines, and atonal barks and grunts appear in the fi rst few weeks of life in conjunction with the onset of social behavior.

There’s a big difference between the bark of an adult dog and that of an adult wolf, however. Dogs seem to play every instrument in the orchestra, hitting the highs of the flute and the lows of the tuba, sometimes with the duration of a Wagnerian opera. Plus, there seems to be no context in which a dog won’t bark: They bark when alone and with other dogs. Some bark before, during and even after a ball is thrown. A car goes by or the doorbell rings and barking ensues. In contrast, wolves bark less frequently and in fewer contexts, primarily for warning or defense.

Meanings Behind the Message
What do canine vocalizations mean? Animal behavior researchers have only recently begun to chip away at this question. As Monique Udell, PhD, who is currently a faculty fellow at the University of Oregon, refl ects, “Vocal behavior in other species has received a lot of detailed attention. In birds, we’ve looked down to the note sequence and explored tiny variations. Vocalizations are such a prominent feature of dogs, and there is a lot to learn.” To date, dog vocalizations have not received comparable scrutiny.

That being said, research that has been conducted on the subject is incredibly insightful. Take growls, which, it has been shown, dogs use to accurately judge another dog’s size. How in the world do we know that? Tamás Faragó, PhD, and his colleagues at the Family Dog Project at Eötvös Loránd University in Budapest (familydogproject.elte.hu) presented dogs with two images of the same dog: one was true to size and another was 30 percent larger or smaller. Dogs then listened to a pre-recorded growl, and most dogs looked at the image of the full-size dog rather than the altered image.

Growls appear to be meaningful in other ways as well. In another study, Faragó and his colleagues used some clever trickery to explore how dogs respond to growls recorded in different situations. In an apparently empty room, a dog was allowed to approach a bone. Unbeknownst to the dog, there was a speaker hidden behind the bone, and as the dog approached, the sound of a “play growl,” a “stranger-approaching” growl or a “food-guarding” growl was transmitted through it. Dogs were likely to take the bone when hearing the “stranger-approaching” or “play” growl, but the food-guarding “my bone” growl deterred them. Even though the foodguarding and stranger-approaching growls sound quite similar (at least, to our ears), they prompted different behavior.

Many studies investigating vocalizations are based on prerecorded samples, but it is important to remember that vocalizations and visual signals usually go hand-in-hand. In the strangerapproaching context, dogs growled with closed mouths, whereas in fooddefense situations, they showed their teeth and pulled back their lips.

While we tend to take notice when we hear a growl, we often dismiss barking as meaningless noise, as though it is simply an item on a dog’s daily checklist: “Take a walk, have breakfast, bark.” Before the turn of the century, that was the prevailing view among researchers and theorists. At most, barking was thought to result from social facilitation— one dog barking prompts other dogs to bark—or maybe attentionseeking, or even rivalry or defense.



CommentsPost a Comment
Please note comments are moderated. After being approved your comment will appear below.
Submitted by Mona Lindau-Webb | March 9 2013 |

Humans tend to be sooo human oriented, looking for "meaning" in animal vocalizations. Dogs bark, growl, whine, grunt, do soft mmh-mmhs, etc., etc. but the standard communication model for human speech with a speaker at one end, and a listener at the other and a "message" in between just does not fit animal communication. Animals probably do not convey information to each other. Why would they? It would not confer any survival advantage to the speaker. So instead, look at what the dog has to say from the dog's vantage point.

The frequency code as described by E.Morton has been shown to fit close communication between animals. There is now evidence from several hundred species on this, including fish and humans. The medium is the message, an animal uses ritualized communication to manipulate other animals into some sort of action. High pitch and harmonic spectrum communicate smallness, "don't hurt me", submission, friendliness, not willing to go into conflict. Low pitch and harmonic spectrum like a dog's growl is a warning. Low pitch and a spectrum with lots of noise is a dire threat that unless the listener moves away, something bad will happen.

What we call pitch is produced by the vibrating vocal cords, and animals reinforce this by also changing the resonances in the vocal tract size so that they sound smaller or larger. This is done by pulling the lips back to shorten the vocal tract so that the animal sounds smaller together with the high pitch, or by pulling the corners of the lips forward to increase the length of the vocal tract to sound larger. Look at a serious threat from a dog, the corners of the lips is the sure sign that tells you how close the dog is to follow through on the threat. In addition, dogs lower the larynx as well in a threat to sound even larger. Humans of course are totally used to decipher this, we constantly calibrate the speech of various size people to normalize the vowel space, for example.

Human speech uses the same frequency code where for example questions in many many languages are signalled by some sort of high pitch, a question meaning uncertainty, appeal to the listener. Lower or falling pitch over a sentence signals assertiveness, dominance (?), etc. So of course humans should be good at deciphering friendly and threatening canine vocalizations.

Humans also manipulate the size of the vocal tract in the same way as dogs and other animals by retracting the corners of the lips. John Ohala demonstrates that this is the answer to the eternal "riddle of the smile": why would you show all those teeth when you intend to be friendly? Well, retracting the commissure of the lips shortens the vocal tract so you sound smaller, and friendlier, and non-threatening.

Of course dogs have a lot more to say other than manipulating other beings. They certainly have the "alert alert alert" bark. There is a special high pitch of the play bark.

Some dogs use a type of harmonic spectra soft mutterings when communicating with people, never with other dogs, as if to imitate human speech to a small extent. I have observed these signals when dogs attempt to manipulate me into waking up, or getting my attention.

Canine vocalizations should really be described much more from the dogs' point of view, not from the usual anthropocentric angle. This article is quite correct that dogs have a lot to say. I also observe that dogs that have been brought up to communicate with their person
will produce fresh types of sounds that are vaguely similar to some human sounds when "talking" to their special person.

Evolution in the making???

Eugene Morton and Jake Page, Animal Talk. 1992.
John J. Ohala, The acoustic origin of the smile Journal of the Acoustical. Society of America 68.S33.1980
Mona Lindau, Testing a model of intonation in a tone language. Journal of the Acoustical Society of America. 80(3):757-64.1986

Submitted by J | June 23 2014 |

Humans are animals. Animal vocalizations, whether primate, canine, or anything else do have meaning. If anything, your definition of 'conveying information' is more human-oriented than the author's. Do you not think that a warning bark is trying to convey something?

You seem to be generally agreeing with the article, while your objections focus more on the language the author uses. It should be obvious that the author chooses that approach because they are communicating the information to humans, not to dogs

More in Behavior & Training:
Dogs Take to the New
Zack's Amazing Transformation
Bringing Home a Second Dog
Cautious Canines
Aggression in Dogs
Q&A with Denise Fenzi
Training Dog Trainers
Is Your Dog a Southpaw?
Run for Your Quality of Your Dog's Life
Learn How to Train Dogs at ClickerExpo 2014