"Guilt" Part 1 - Is Denver the dog really guilty?

Denver, Masey and a consumed bag of cat treats are the key players in the youtube video “Guilty!”. 


In this home video, Denver is deemed guilty of a ‘cat treat eating’ infraction due to what many humans perceive as her incredibly cute guilty behavior. Since its posting on March 8, the video has received over 5 million hits, and the phenomenon now includes a web site that features bandanas and t-shirts of Denver and her guiltiness. 

Last week, Denver, Masey and their human companions joined the ABC morning show. According to the dictionary, “news” is “information about recent events or happenings”. I did not see any news during that morning show. Instead, I saw a bunch of morning personalities throwing out assumptions and offering the audience pleasing banter and humorous judgments about dogs. They provide no real information or "news" about what happened to the cat treats.

The takeaway messages from the morning show are:

• Denver must have eaten the cat treats because when shown the empty bag and asked, “Did you do this?” Denver displays squinting eyes, thumping tail, looking away and presenting teeth.

• Masey, the other dog in the room, does not show such overt behaviors and is deemed innocent.

• The owners proclaim that both Denver and Masey are "sneaky".

• The owners also say that Denver displays these behaviors when she has misbehaved. According to them, she tries hard to please them and knows when she has disappointed them.

So there you have it! Solved! The story of Guilty Denver as told by ABC News and this guy:

As you can tell, I have some qualms with the reporting. This youtube video provided a great opportunity to explore what we know and don’t know about dogs, their behavioral displays and their emotional experiences. ABC News explored none of these topics.

What we know:

Research finds that dog owners ascribe guilt to dogs. There is a reason this video received more than 5 million hits. Dog owners observe particular behaviors* - such as "avoiding eye contact, lying down and rolling to the side or onto the back, dropping the tail, wagging low and quickly, holding one’s ears down or head down, moving away from the owner, raising a paw and licking" - and owners believe these behaviors correspond with a dog's feeling of guilty. In a 2008 research paper**, 74% of dog owners attribute guilt to dogs (as an FYI, the next highest attribution of guilt was made by horse owners at 36%). Additionally, dog owners believe that when dogs show “guilty” behaviors, dogs KNOW they have done something owners disapprove of. To spell it out, owners believe that “guilty” behavior corresponds with knowledge of a misdeed.

Had ABC conducted any investigatory journalism, they’d have found that scientific research to date does not support the idea that “guilty-looking” behavior necessarily corresponds with knowledge of a misdeed. 

The New York Times piece, “It’s an Owner’s Scolding That Makes a ‘Guilty’ Dog”, covered Dr. Alexandra Horowitz's research into the “guilty look”. 

This research found that if you scold a dog, you could see “guilty” behaviors. Dr. Horowitz said that the results show that dogs have learned to act in a certain way in response to their owners’ behavior: "We’ve trained them that when they see us angry, they give us that guilty look. I’m not saying they don’t feel guilt," she added. "I can’t test that yet. But we generate the context that prompts them to produce this look"

To repeat: “guilty-looking” behavior does not necessarily correspond with a dog's knowledge of a misdeed.

What we don’t know: 

Do dogs experience guilt? In humans, we know guilt is a secondary evaluative emotion that is assumed to require self-consiousness. Guilt stems from a self-perceived violation against an established rule to which one has chosen to adhere. But, as children, conformity to ethical standards lacks a moral basis and therefore is “only indicative of submission to authority rather than acceptance of it” - Ausubel***. Accordingly, it is possible for humans to display guilty behavior without possessing an internal moral code. The same could be said for dogs.

Scientists assume non-human vertebrates have primary emotions such as fear because they serve an evolutionary, fight-or-flight function. But, secondary emotions are not confirmed.

Back to the case at hand, Denver and Masey. You may be wondering (a) why do dogs show these behaviors if they are not “guilty”; and (b) in the case of Denver and Masey, why does Denver show these behaviors and Masey does not?

A) Why do dogs show these behaviors if they are not “guilty” 

We don’t know if dogs feel guilt, but we do know that they show submissive behavior. Submission serves to keep a social group together, to foster group cohesion. Depending on the context, such displays are called appeasement behavior, "behavior that inhibits or neutralizes aggression in a behavioral partner". Consider dogs and humans: could a scolding human be deemed an aggressive behavioral partner? Could dogs be trying to inhibit or neutralize aggression? There are many instances where dogs exhibit such displays, and it doesn’t have to be because they experience the secondary evaluative emotion of guilt. 

When we describe a dog as guilty, maybe we’re taking a perfectly normal dog behavior and anthropomorphizing it into guilt. In America, dogs are incredibly intertwined in our households. In many cases, we’re actively choosing, through breeding or training, dogs that have behavioral traits and morphologies that help them integrate into our society. Imagine if you scold a dog and it attacks you. That dog’s breeding potential will probably decrease. Imagine if you scold a dog and it shows submissive displays. That dog could be more successful in passing on its genes (unless of course, the dog is speutered, but that’s a whole ’nother story). 

Researcher James Serpell addresses this issue using the concept of anthropomorphic selection****, which is “selection in favor of physical and behavioral traits that facilitate the attribution of human mental states to nonhumans”. For example, we often prefer dogs with big, forward facing eyes, which also then makes it easier for us to say, "Oh, look at those guilty eyes!"

Let's consider this attribution from an evolutionary perspective. Wolves, dogs’ wild-type progenitor, show submissive behavior. As in dogs, these behaviors function in wolves to keep the group together. I’ve asked wolf experts about guilt in wolves and received blank stares (don't ask this question of wolf experts, it will not make you popular). So why do we ascribe guilt to dogs but not guilt to wolves? Is it possible that dogs evolved this secondary emotion through the domestication process? Possibly. Or is it that we make this attribution because dogs are part of our social environment and wolves are not? One has access to cuddling in our bed one minute and raiding our pantry the next, and the other, not so much. 

B) In the case of Denver and Masey, why does Denver show these behaviors and Masey does not?

First off, maybe Denver shows submissive displays because that's what she does. Every dog in the presence of a scolding owner responds differently. Some only show it in particular situations regardless whether they performed a misdeed. Just because Denver shows these behaviors, does not necessarily mean she transgressed. 

And what about Masey? Why is she so cool and collected in front of a scolding owner? Guess what! Research finds that not all dogs show the guilty look, even when they have misbehaved (I’ll share my own research in this area soon!).

In order to assess the Guilty Denver video, ABC should have asked about Denver and Masey’s history. If Masey never displays the "guilty look", regardless of what she has done, her behavior can’t be used as a reliable clue. Until you set up a video camera, you won't find out what really happened in your absence! 

Many people shared Denver's guilty video with me. It was best described as “funny, cute and sad all rolled into one!” When Denver shows these behaviors, some people think, “Aww, cute, she’s apologizing”. But we’re also watching a dog get scolded and show submissive behaviors. Is that cute? Kind of, but given everything I've talked about, maybe it’s a bit more sad.  

Now turn that frown upside down! We can be thankful this video was posted on youtube! It provides lots of food for thought and potential research questions. A major takeaway should be, who do you want to hear dog analysis from? Julie or this guy?

* Horowitz, A., 2009. Disambiguating the "guilty look": Salient prompts to a familiar dog behaviour. Behav. Process. 81, 447-452.

** Morris, P.H., Doe, C., Godsell, E., 2008. Secondary emotions in non-primate species? Behavioural reports and subjective claims by animal owners. Cognition Emotion 22, 3.

*** Ausubel, D.P, 1952. Ego Development and the Personality Disorders. A Developmental Approach to Psychopathology. Grune and Stratton, Inc., New York.

**** Serpell, J., 2003. Anthropomorphism and anthropomorphic selection - beyond the "cute response". Society and Animals 11, 83-100.