Barking as Vocal Communication in Dogs
“Bark study hits popular press worldwide”

Ask any dog owner why his or her dog barks and you’ll get a plethora of answers. Because he is hungry. He wants to come inside. He can’t stand strangers on his property. With so many uses for one type of vocalization, is barking just a nuisance by-product of domestication or could barking serve some specific communication functions?

1For the answer, I initially turned to the scientific literature. Surprisingly, although there are over 52 millions dogs in the U.S. alone, there are very few studies on vocal communication in domestic dogs. The two main peer reviewed papers were published in the 1960s and 1970s and a well-known popular press article was published in the early 1990s. Most books, textbooks, and scientific articles that mention barking in dogs cite these three sources. Interestingly, the consensus among researchers has been that dogs bark so frequently, in so many contexts, and some bark for such long periods of time in the absence of obvious stimuli or receivers (animals that are listening), that barking must just be a non-specific way for dogs to get attention. More subtle communication occurs through body language and olfactory cues.

2These arguments are convincing, especially in light of the fact that barking is often considered a nuisance behavior. Many counties in the U.S. have adopted ordinances that address noise from barking dogs (Seen and Lewen, 1975), barking is a major source of noise pollution in dog kennels (Sales et. al., 1997) and inappropriate barking is one of the most common behavior problems that owners report to their veterinarian with up to 35% of owners listing this as a complaint (Beaver, 1999).

These arguments are, however, the types of arguments reserved solely for domesticated animals. It’s easy for people to assume that such animals are tainted by a soft life and artificial selection and thus have many aberrant or functionless behaviors. If, however, we take a step back and evaluate vocal communication in domesticated dogs the way we would evaluate vocal communication in a wild animal the view becomes quite different.

Vocalizations in Wild Animals

3Many wild animals use one vocalization frequently and in many different contexts and some repeat one or more vocalization for hours on end. In cases where a species uses one vocalization in many contexts, researchers have found that upon closer evaluation, the vocalization is actually comprised of subtypes. For instance, chacma baboons bark in several contexts (Fischer et al., 2000). Their alarm barks made when they see predators is acoustically different from the contact bark they make when separated from their group or from their offspring. Additionally, the alarm barks they make towards mammalian predators is different from those made towards crocodiles.

4Other animals have been known to vocalize for many hours in a row. The classic example is the song bird who sings sonorous songs in spring. These birds sing in order to protect their territory and to attract mates. Females are attracted to the males with he best songs. Even wild canids have been known to vocalize for hour upon hour. During breeding season, wolves may howl and howl without reply (Field, 1975). These vocalizations are to attract females from neighboring packs.

In light of these findings in wild animals, there are many reasons why barking could be a specific from of communication in dogs. As a first step towards exploring this topic, I embarked on a study that asked the question, “Are barks context specific?” That is, are barks in one context acoustically different from those in other contexts?

Sophia’s Research Study on Barking in Dogs

I took ten adult dogs of six different breeds and recorded barking in three different test situations-a disturbance situation where a stranger rang the doorbell, an isolation situation where the dog was locked outside isolated from its owner, and a play situation where either two dogs or human and dog played together.

Using a sound-editing program, I then converted the digital recordings of 4672 barks to visual displays called spectrograms. A computer macro written by Dr. Brenda McCowan (Department of Population Health & Reproduction, UC Davis, took 60 sequential frequency (pitch) measurements and 60 sequential amplitude measurements along the length of each bark and then recorded values such as minimum, maximum, mean frequency, start slope, finish slope, frequency range, amplitude range, bark duration and interbark interval for each bark.


Statistical analysis revealed that dog barks can be divided into different subtypes based on context and that individual dogs can be identified by their barks. Additional analysis revealed that disturbance barks tend to be harsh, low-frequency, and unmodulated whereas isolation and play barks tend to be tonal, higher-frequency, and modulated.

These findings suggest that barks may have specific functions in specific contexts; further studies should be performed to explore this idea. Additionally, because barks vary predictably with context, humans should be able to recognize the barks their dogs give in specific contexts. Doing so will give humans a better understanding of dog behavior in general as well as help in diagnosing and treating bark-related problems.

How can you learn to understand what your dog’s saying?

You can easily learn to interpret your own dog’s barks by carefully listening and noting the context, his body posture, and the response of the listener. For instance, listen to your dog’s bark when a stranger knocks on the door. Is it harsh like your voice when you have a cold or clear so that is sounds like a single note? It is low pitched or high compared to his other barks? Does it occur in clusters with many closely spaced barks or are they single barks? Are the barks so fast their fused into superbarks?

Now listen to your dog’s barks in other situations. Do the barks to strangers at the door sound the same as those to friends who come over? When Bowser’s barking at squirrels outside is it the same as his bark for his toy?

By systematically looking at the specific contexts and correlating it to what you hear, you’ll quickly be able to interpret Fidos specific barks.

What about the Bark Translator?

Read Sophia’s review in the San Francisco Chronical

Modified from the San Francisco Chronicle, August 16, 2003:

article1After months of hype, the Bow-lingual, a gadget that claims to translate dog barks into English, is finally on the market. Manufactured by Takara, a Japanese toy company, this device listens to a dog¹s barks and then categorizes the dog¹s mood as happy, sad, on guard, assertive, frustrated, or needy. Each emotion comes with one of 178 cutesy English-sentence translations.

The contraption consists of a spiffy wireless microphone that attaches to your dog’s collar and a handheld unit, which displays the bark translations. The nifty design and the fact that it’s touted to translate barks of 74 breeds makes it a seeming must for any owner who truly loves his dog. But before you rush out to buy this $120.00 toy that’s supposedly backed by sound science, read ahead. The product has already been featured on the news, and in many newspapers and magazines and has even been applauded as one of Time Magazines, “Coolest inventions of 2002” (Nov 18, 2002). Unfortunately most reporters other than Walter Mossberg of the Wall Street Journal haven’t tested it thoroughly. If they had they would have found that it doesn’t work.

As a researcher in the area of vocal communication in dogs, I’ve been waiting for over a year to get my hands on this translation device. I finally had my chance when a local television station, KTVU, sent me one to test.

article2My first step was to take pre-recorded clusters of barks collected from my earlier research and play the same bouts over and over to see whether the Bow-lingual could give the same answer to the same bouts regularly. I started with barks from my Australian cattledog Zoe, and the initial results were fair. When I played a bout of her disturbance barks, those emitted in response to the doorbell, ten times in a row, the Bow-lingual translated her mood as frustrated 80% of the time. When I played it ten more times the precision dropped to 65%. When the invention didn’t say she was frustrated though, it said she was happy, assertive, or sad.

Next I repeated the procedure with a bout of Zoe’s isolation barks which were recorded when she was locked outside and wanted to come in, as well as with a bout of play barks recorded when she was frolicking with another dog. This time the results seemed better in that they were the same 100% of the time. The problem is that according to the translation, Zoe was frustrated all of the time. She was frustrated when she was alerting me to intruders at the door, frustrated when she sat politely waiting outside for someone to let her in, and frustrated when she was playing exuberantly with another dog.

article3While it’s possible that a dog could be frustrated in all three situations, scientists determine the meaning of vocalizations by looking at the context, body posture of the barking dog, and the response of the animals or humans who hear the barks. In this case all three factors said the mood translations were wrong. To make matters worse, each mood was also accompanied by a useless English saying such as “I’ve got a funny feeling,””I want help,” and “You just don’t get it” which didn’t make any sense in the given situations.

The translations went from bad to worse when I tested the apparatus on real-live dogs. I tested it on the home alone function which records each bout of barking up to 100 for up to 12 hours. I also manually took sequential measurements in the same context. When Zoe performed her daily wild-eyed race to greet the UPS man at the door her barks were translated as, ” I’ve done it good right?” (happy), “I love to be at your side” (needy), “Why don¹t you talk to me” (sad) among others. I’m sure if, Todd, the UPS deliverer were given a list of things he thought she was saying, those phrases would not be on the list. A more appropriate translation would have been “Red Alert!” “Get off my property,” or “I’m going to kick your butt.” I never saw any of those translations listed in the Bow-lingual.

I also tested the Bow-lingual on a handful of other dogs in the same manner with the same scattered results, which brings me to another point. Say your dog’s barking at you near the dinner table because he wants you to give him a treat. If the Bow-lingual says something like, “Please be nice to me” (frustrated) versus, “You can’t beat me” (on-guard), will it make a difference as to whether you give her the treat? Or, if your dog’s barking uncontrollably at another dog and the bow-lingual says he wants to play will you let him run up and greet the other dog or head the other way? Regardless of the dog’s intention, even dogs that want to play can be extremely impolite. Their poor greeting behavior or uncertainty in this excitable situation can quickly lead to a spit and drool fight.

And then of course, if your bowser barks in front of at the dinner table every day until you feed him some scraps or stands at the sliding glass door and barks until you let him in, why do you need a translation to know what he wants anyway? Isn’t the answer right before you eyes?

Overall, my final ruling is that the Bow-lingual is fun to play with for a while if you got it for free, but it’s not very useful since the translations aren’t trustworthy and most don’t make sense. The toys is marketed as being backed by strong science carried out by respected researchers but somehow in this case, despite their accolades, they produced a dud.

Literature Cited:

Beaver, B. 1999. Canine Behavior: A Guide for Veterinarians. Philadelphia: W.B. Saunders.

Field, R. 1975. A perspective on syntactics of wolf vocalizations. In: The Wild Canids (Ed. by Fox, M. W.), pp. 183-201. New York: Van Nostrand Reinhold Company.

Fischer, J., Cheney, D. L. & Seyfarth, R. M. 2000. Discrimination of call types by free-ranging chacma baboons (Papio cynocephalus ursinus). Folia Primatology, 71, 189-248.

Sales, G., Hubrecht, R., Peyvandi, A., Milligan, S. & Shield, B. 1997. Noise in dog kennelling: Is barking a welfare problem for dogs? Applied Animal Behaviour Science, 52, 321-329.

Senn, C. L. & Lewin, J. D. 1975. Barking dogs as an environmental problem. Journal of the American Veterinary Medicine Association, 166, 1065-1068.