Canine Science Isn't a Soft Science: Hard Dog Data Abound

Canid science isn't a soft science and researchers have nothing to defend

In response to an essay I posted on February 9 titled "Dogs and Guilt: We Simply Don't Know" (the original title was "Dogs and Guilt: We Simply Don't Know"), I received the following three email messages.¹ I shared them anonymously with a number of colleagues, removing both the names of the person who sent them and their two male dogs, to whom I'm referring as Harry and Sam.

I've read and reread these notes a number of times and received some interesting comments from my colleagues, and I think it's worth responding to them because the sender took the time to write to me. I close down online comments for my essays because they generate an incredible amount of spam and because I receive a good number of emails about them and it's easier to have exchanges with people who take the time to write to me. Thanks to all who do so.

Let me be very clear upfront, those people who study dog behavior, cognition, and emotions have nothing to defend. Many are doing exemplary research, their results are very interesting, and what we're learning can be used to give dogs the very best lives they can have as they try to adapt to a human-dominated world.

Here are the three notes I received in response to my essay "Dogs and Guilt: We Simply Don't Know."

I received these two on February 9:

(1) I think that the field of canine cognition is one scientific discipline that is pretty approachable to commentary from ordinary people. It is definitely a soft science, unlike nutrition or fluid mechanics. There are so-called experts in this field who lack common sense. The public is not supposed to argue with their findings, even when they often contradict one another. Witness several years ago when within a span of a year it was claimed that dogs cannot remember anything for more than 30 seconds, then later that their faculties are so advanced that most dogs understand most of what is said in most households without training. Wow!

The public can see that a dog is not so dumb that it cannot understand "culpability." You can see in pictures on line that the dogs seen there know they have done something they are not allowed to do or in some sense "wrong." Whether the average dog feels any obligation for what it has done or understands in the most absolute sense that it was wrong is another matter. What maybe "WE" don't know is whether this lack of empathy for the human household has anything to do with brain function.

Harry not only knew what guilt was, he knew the words "guilt", "shame", "sorry." Harry was a highly domesticated and sensitive canine.² As an intelligent dog, Sam certainly understands what guilt is, but at least as a 3-year-old dog, seems to feel less obligation to others in his household.

It continues to concern me that the basics of my canine language instructions were thrown out entirely by you (as opposed to reserving judgment). But, speaking for many other "citizens" out there, WE do know the answers to some of the things you have written about recently. We do not need to stay tuned for the next 30 years to find out those answers.

(2) I taught Sam the word "guilt" a half hour ago, saying it meant: If you do something and are sorry for it and feel bad and like you wish you could do it over and not have done something bad.

I asked him a few minutes ago whether if he bit Mommy on the face or tummy he would feel "guilty", to which he answered "yes."

I received this note on February 10:

Today I asked Sam to answer this: If you knew a nice girl dog who had a puppy she really loved and you killed it, would you feel "guilty"? The answer to that from Sam was "yes."

Would he always know his own mind on my asking a question about that topic in other forms of the question on guilt, and also, could instinct sometimes override reason? I would think no he might not always know his own mind, and yes, sometimes animal instincts might override reason.

Thanks for listening.

Canine science is not a soft science

First, canine science is not a soft science, especially compared with nutrition. My first response to this charge was that all one has to do is turn on their TV or go to the web to find literally 100s or more diets that are claimed to be "the best." I was shocked at just how any diets there out there that claim to do everything we all want and need, and a nutritionist friend of mine found this claim to be "patently uninformed and ludicrous."

I really can't make any statement about whether or not fluid mechanics is a "harder" science than "canine science," because I'm not an expert in fluid mechanics. However, just because a scientific discipline has a lot of mathematics behind it, this doesn't necessarily make it a hard science. And, think of how many different interpretations and explanations there are for the same data sets in many of the so-called "hard sciences."

Concerning the misleading claim that canine science is a soft science, one can easily discover that there are numerous people around the world doing detailed research in both captive and "field" settings, and they are adding to an ever growing data base in many different areas of inquiry.³ There are many popular outlets for what these researchers are learning and scholarly research can be found here. In addition, numerous books on dog behavior and cognition can be found here. Surely, some of the more popular outlets differ in quality, as do some of the more formal research projects. However, there's enough information out there for interested parties to check and to assess their accuracy and quality, and a good number of more scholarly books can be found here.

There is no "the dog" because of enormous amounts of interesting and informative variation among dogs

The claim that "The public is not supposed to argue with their findings, even when they often contradict one another," surely is not consistent with what I know of my colleagues. While some choose not to get involved in these sorts of exchanges, others do. And the fact that results from different studies contradict one another is to be expected for a good number of reasons. Simply put, different dogs are being studied in different contexts by different researchers using different methods. And, sample sizes vary greatly from small to large, and this also has to be considered when assessing the robustness of results. So, it's not only interesting when there's good agreement among different studies, but it's also interesting and informative to figure out why different results emerge when people are asking the same or similar questions. Within-species variability is not noise in the system, but rather something that needs to be studied further.

Along these lines, one dog expert wrote to me in October 2016 and asked, “Who are these dogs in all of these tests?” He was referring to the fact that studies frequently treat all dogs as equivalent, but they are not. It’s just not possible to say all or even most or many dogs do this, or that all or even most or many dogs do that, or even that dogs and wolves are similar in this way and different in that way. If many of the people I meet at dog parks know this already, that’s because their dogs already act like they’re one of a kind!

Guilt Essential Reads

The Problem With Guilt

I'll Never Tell: Rethinking the Psychology of Secrecy

Because research conducted in different dog labs and in the field uniformly shows there is an incredible amount of within-species variability among dogs, when people ask me questions about “the dog,” I often say there’s no such being. Sure, there are some general trends, but what always catches my eye is the enormous amount of within-species variability among dogs, including littermates, siblings, and other members of the same breed or mix.

The role of citizen science and the emotional lives of dogs

​Let me briefly respond to the writer's claims that I threw out what they wrote me about their observations of the linguistic abilities of their dogs and also how their dogs felt guilt. First, I didn't throw them out, I simply suggested that it would be a good idea for them to write up something more formal that others could review, similar to what researchers do who then receive peer-reviews to which potential research papers are subjected. They could then submit their findings to a journal and get the word out to a wide audience. I reserved judgment on what they sent me because I'm not an expert in the field of canine linguistic abilities. However, other colleagues are and I recommended they write to them.

I'm a also staunch fan of "citizen science" and have written about how useful and important it is in various essays and books. Stories can and do lead to rigorous research projects, and anecdotes of different quality abound about guilt in dogs. And, yes, I'm sure some people "know" something that hasn't yet been studied or has been refuted by different studies, but this does not mean that dogs necessarily lack these cognitive or emotional capacities.¹ It simply means that we need more detailed studies and more details for the stories about these observations.

So, for example, I find their description of how Harry and Sam knew what guilt meant to be interesting and intriguing, but I simply want to know more details about how they determined this. Recall what they wrote:

I taught Sam the word "guilt" a half hour ago, saying it meant: If you do something and are sorry for it and feel bad and like you wish you could do it over and not have done something bad.

I asked him a few minutes ago whether if he bit Mommy on the face or tummy he would feel "guilty", to which he answered "yes."

Today I asked Sam to answer this: If you knew a nice girl dog who had a puppy she really loved and you killed it, would you feel "guilty"? The answer to that from Sam was "yes."

I'm not asking for anything more from them than I would from other citizen scientists or researchers. We just need to know more.

Let me say again that canine science researchers have nothing to defend. They're trying to do the best science they can, and different results for studies focusing on the same or similar questions are to be expected.

Please stay tuned for more on the cognitive and emotional lives of dogs

In response to the writer's saying, "We do not need to stay tuned for the next 30 years to find out those answers," I simply respond that it would be most valuable if they and others who have information that can inform the study of dog cognition and emotions take the time to write up something that clearly lays out how they did what they did and to show us what they've collected, namely, their data. Data vary in quality and quantity, and this isn't asking too much. Many citizen scientists are doing just this so that what they have done and the conclusions that offer can be reliably assessed.

So, I repeat, please stay tuned for more discussion of the cognitive and emotional lives of dogs. The more we learn the better it will be for them and for us. And, it's incumbent on us to use what we know to give our canine friends the best lives possible.

The more we learn and the more we use what we learn on their behalf makes it a win-win for all, for them and for us. However, far too often dogs get the short end of the leash, so to speak. In many -- far too many -- situations, dogs want and need far more than they get from us. Let's be sure that in the future the myriad relationships we form with dogs are far more egalitarian and both the dog and their human benefit from this wonderful relationship.

Notes

¹As I was writing this essay I received another email from the same person: "What is the point in publishing the results of a study and stating they were inconclusive because the dogs reactions could not be analyzed properly? Sometimes dogs know they are culpable of what they are accused of and sometimes they don't. On the other hand, sometimes they don't give a hoot. All of that does not mean they are not capable of feeling sorry for their actions."

²For accuracy, it's important to point out that the phrase "highly domesticated" really doesn't mean anything at all. Perhaps Harry was highly socialized because of his strong relationship with the writer, but he was no more domesticated than any other dog.

³I've written numerous essays about research on the cognitive, emotional, and moral lives of dogs that are available for free online. James Serpell's edited book The Domestic Dog: Its Evolution, Behavior and Interactions with People, 2nd Edition is a goldmine of information on all "things dog."