Aggression starts from inherited traits
SEATTLE, Washington––The dogma that dog aggression, especially pit bull aggression, is “all in how you raise them” is again refuted by science, but the average reader will have a hard time deciphering the message from the highly technical report “Highly heritable and functionally relevant breed differences in dog behavior,” published on October 2, 2019 by Proceedings of the Royal Society.
Co-authors Evan L. MacLean of the University of Arizona, Noah Snyder-Mackler of the University of Washington, Bridgett M. vonHoldt of Princeton University, and James A. Serpell of the University of Pennsylvania “integrated behavioral data from more than 14,000 dogs from 101 breeds with breed-averaged genotypic data from over 100,000 loci in the dog genome,” their Royal Society paper reported.
“We found high levels of among-breed heritability for 14 behavioral traits,” MacLean, Snyder-Mackler, vonHoldt and Serpell wrote, in one of their few sentences not stated in technical jargon. “We next identified 131 single nucleotide polymorphisms associated with breed differences in behavior, which were found in genes that are highly expressed in the brain and enriched for neurobiological functions and developmental processes, suggesting that they may be functionally associated with behavioral differences.”
Line-bred dogs show line-bred traits
In other words, line-bred dogs such as pit bulls exhibit line-bred traits, exactly as dogfighters have understood since the earliest documentation that breeding “game” dogs existed as an occupation.
While fighting abilities may be whetted by training, a pit bull does not need to be trained to have the instinct to rush at another dog, bite the body part most vulnerable at that moment, clamp down, and shake.
Comparably, greyhounds do not need to be trained to run, retrievers to retrieve, pointers to point, or border collies to herd. For each breed, “Highly heritable and functionally relevant breed differences in dog behavior” suggests, training builds upon innate behavioral tendencies evolved parallel to physical adaptations to do specialized work.
The science of how
This is exactly as behaviorist Alexandra Semyonova explained much more intelligibly, but without the detailed genetic analysis, for ANIMALS 24-7 in The science of how behavior is inherited in aggressive dogs (2014, updated 2015.)
“The majority of variance among modern breeds,” MacLean, Snyder-Mackler, vonHoldt and Serpell recognized, “has probably resulted from the repeated crossing of novel phenotypes, which—originating from a limited pool of genetic variation—has nonetheless given rise to extraordinary phenotypic diversity.
“Although dogs have been bred by humans for millennia,” MacLean, Snyder-Mackler, vonHoldt and Serpell continued, “the formalization of modern breeds, as defined by closed genetic pools, occurred only 200–300 years ago. Consequently, most modern breeds are characterized by limited genetic diversity, and members of the same breed can be reliably assigned to a single breed-specific clade,” a clade being any group descended from a common ancestor not shared with other groups.
Dog & human evolution
“From an evolutionary perspective,” MacLean, Snyder-Mackler, vonHoldt and Serpell explained, “variance across breeds reflects diversifying selection, which has occurred rapidly through selective breeding by humans.”
Measured by the average interval between birth and first successful reproduction, a dog generation tends to be just one to two years. Demographers define a human generation as 18 years.
A dog breed category developed over 750 years or longer, including herding dogs and fighting dogs, would thus have gone through at least as many generations as humans have since the most recent common ancestor of all humans alive today.
A dog breed category such as retrievers, developed since the advent of shotguns circa 500 years ago, would have gone through approximately as many generations as have humans since the dawn of civilization.
“Attributable to genetic factors”
“We found that a large proportion of behavioral variance across breeds (among-breed heritability) is attributable to genetic factors,” MacLean, Snyder-Mackler, vonHoldt and Serpell concluded.”
Translated, this means that the behavioral differences that distinguish greyhounds from pit bulls from golden retrievers are chiefly inherited, not taught––again, exactly as practically anyone familiar with dogs has believed since breeding dogs for specialized tasks began, with the sole exception of pit bull apologists who insist that a tendency to kill and maim other dogs and humans is “all in how you raise them.”
“Interestingly,” MacLean, Snyder-Mackler, vonHoldt and Serpell confirmed, “the traits with the highest among-breed heritability were trainability, stranger-directed aggression, chasing, and attachment and attention-seeking, which is consistent with the hypothesis that these behaviors have been important targets of selection during the formation of modern breeds.
Genetic heritage vs. drift
“Indeed,” MacLean, Snyder-Mackler, vonHoldt and Serpell found, “the overall patterns of breed differences align closely with genetic clades corresponding to functional breed groups. Analysis revealed that all 14 behavioral traits [that they identified as highly heritable] were under strong positive selection, and could not be explained by genetic drift alone.”
Translation: if conditions in a particular habitat favor dogs with long red coats, over time a dog breed will emerge in that habitat characterized by a long red coat, since more dogs with long red coats will survive to breed successfully. This would exemplify genetic drift. Short brown coats might persist, but only as a recessive trait.
However, if someone were to begin deliberately mating dogs with long red coats, culling the rest, the dog population would come to be characterized by long red coats much sooner, and dogs with short brown coats might entirely disappear.
Behavior influenced by many genes
“The researchers were able to narrow down the list of possible candidate genes responsible for several types of behavioral traits,” assessed science writer Bob Yirka for Phys.org.
“They found that some were more easily identifiable than others. For traits such as chasing, aggression or trainability, for example, the researchers found genes that contributed to approximately 60 to 70% of variability between breeds. Other genes related to traits such as fearfulness or energy levels were not as evident.
“The researchers suggest this is likely,” Yirka wrote, “because such traits are more heavily influenced by environmental factors or training. The researchers note that no single gene was found to be responsible for any particular behavior type, suggesting that it is likely that behavior is influenced by many genes,” acting in combinations which may be more pronounced in some dog breeds than in others.
“Didn’t find gene that makes tails wag”
Molly Slann interviewed the “Highly heritable and functionally relevant breed differences in dog behavior” co-authors for the University of Washington newspaper The Daily Washingtonian in January 2019, soon after a preliminary version of the paper was posted by Biorxiv: The Preprint Server for Biology, a project of Cold Spring Harbor Laboratory on Long Island, New York.
“The goal was to try and understand what component of [dogs’] behavioral suite is controlled by their genes,” Snyder-Mackler told Slann. “By that I mean sort of [whether variations in behavior] can be explained by how genetically similar [the dogs are] to one another. For questions like, ‘Did you find the gene that makes dogs’ tails wag?’, the answer is no, we didn’t. We probably won’t ever,” since most likely multiple genes shared by all dogs contribute to tail-wagging, a universal behavior.
What MacLean, Snyder-Mackler, vonHoldt and Serpell hoped to do was to isolate what can be ascribed to genetics from what may be ascribed to environment, including conditioning, by looking at the most distinctive characteristics of a selection of dogs believed to have been bred for many generations to produce those characteristics.
Their starting point was behavioral data for about 14,000 dogs from 101 breeds collected by the Canine Behavioral Assessment & Research Questionnaire (C-BARQ), developed by Serpell and Yuying Hsu in 2003, made available online since 2005.
“C-BARQ asks questions like, ‘What does your dog do when a stranger comes to the door?’ to allow owners to objectively characterize 14 aspects of their pet’s personalities,” wrote Elizabeth Pennisi for Science, also in January 2019. “Since the survey was developed in 2003, more than 50,000 owners have participated,” but MacLean, Snyder-Mackler, vonHoldt and Serpell used only the data from dogs of clearly defined breed.
“The team matched up these behavioral data for each breed with genetic data about breeds from different sets of dogs,” Pennisi continued. “They didn’t look at genetic and behavioral data for individual dogs,” which would have produced a stronger data set, but at prohibitive cost.
Instead, of necessity, MacLean, Snyder-Mackler, vonHoldt and Serpell looked at the averages of behavioral assessments for each specific breed––which is all that most people considering acquiring a dog could do, unless personally acquainted with the canine parents.
Developed to assess behavioral problems
“Although sometimes described as a form of canine personality assessment,” the C-BARQ web site itself explains, “C-BARQ was originally designed to measure the prevalence and severity of behavioral problems in privately-owned and working dogs, and that remains its primary value and purpose.”
“C-BARQ is a survey,” the C-BARQ web site continues, “aimed at dog owners/guardians and handlers that provides a set of numerical scores for the following fourteen different categories of dog behavior.”
These 14 categories are:
- Stranger-directed aggression: Threatening or hostile responses to strangers approaching or invading the dog’s or owner’s personal space, territory, or home range.
- Owner-directed aggression: Threatening or hostile responses to the owner or other members of the household when challenged, manhandled, stared at, stepped over, or when approached while in possession of food or objects.
- Dog-directed aggression: Threatening or hostile responses when approached by unfamiliar dogs.
- Dog rivalry: Threatening or hostile responses to other familiar dogs in the same household.
- Stranger-directed fear: Fearful or wary responses when approached by strangers.
- Nonsocial fear: Fearful or wary responses to sudden or loud noises, traffic, and unfamiliar objects and situations.
- Dog-directed fear: Fearful or wary responses when approached by unfamiliar dogs.
- Separation-related behavior: Vocalizing and/or destructiveness when separated from the owner, often accompanied or preceded by behavioral and autonomic signs of anxiety including restlessness, loss of appetite, trembling, and excessive salivation.
- Attachment and attention-seeking: Maintaining close proximity to the owner or other members of the household, soliciting affection or attention, and displaying agitation when the owner gives attention to third parties.
- Trainability: Willingness to attend to the owner, obey simple commands, learn quickly, fetch objects, respond positively to correction, and ignore distracting stimuli.
- Chasing: Chasing cats, birds, and/or other small animals, given the opportunity.
- Excitability: Displaying strong reactions to potentially exciting or arousing events, such as going for walks or car trips, doorbells, arrival of visitors, and the owner arriving home; has difficulty settling down after such events.
- Touch sensitivity: Fearful or wary responses to potentially painful procedures, including bathing, grooming, nail-clipping, and veterinary examinations.
- Energy level: Energetic, ‘always on the go,’ and/or playful.
“15% of a dog breed’s personality”
“In all, the team identified 131 places in a dog’s DNA that may help shape [the above] 14 key personality traits,” Pennisi wrote for Science. “Together, these DNA regions explain about 15% of a dog breed’s personality, with each exerting only a small effect.”
But that may be somewhat like saying that the steering wheel explains only about 15% of where a car goes. The tires, brakes, and gas pedal also have a great deal to do with it, and make up much more of the bulk of the car. Nonetheless, the steering wheel has the dominant influence.
Pennisi argued that since “Serpell’s behavioral work has shown that pit bulls are aggressive toward other dogs but not people, this new analysis can’t lead to a DNA test of that behavior. However,” Pennisi added, “Serpell and his colleagues are starting more studies looking at the DNA linked to within-breed variation in behavior, a step in that direction.
BARQing up the wrong tree
But ANIMALS 24-7 has pointed out for more than 25 years that Serpell’s behavioral work started out by barking up the wrong tree. Though the genetic findings explained in “Highly heritable and functionally relevant breed differences in dog behavior” may have some corrective effect, Serpell’s original hypothesis that pit bulls are only dog-aggressive and not human-aggressive has not become any more accurate through frequent amplification by pit bull advocates and some academic colleagues.
The first and most fundamental flaw in the work in question is that when an academic study of anything produces results that fail to match real-world experience, it is the terms of the study that need to be adjusted.
Reality is that pit bulls, less than 5% of the U.S. and Canadian dog population now and under 1% until the past three decades, have since 1982 accounted for more than 60% of the dog attack fatalities and 75% of the dog attack disfigurements.
More than a third of the fatal attacks have been found to involve victims from the pit bull’s own household, and have been inflicted by pit bulls who have never been neglected or abused.
Theories must explain the data
These findings, first tabulated and distributed by ANIMALS 24-7, have now been verified by similar data collected and verified by a multitude of other sources, including the authors of 16 different studies published in major peer-reviewed medical journals since 2016, insurance companies, law enforcement agencies, the Centers for Disease Control & Prevention (until it discontinued doing dog attack research in 1989), and, since 2007, by Dogsbite.org.
Any study, such as Serpell’s work, which denies that pit bulls are “owner-aggressive” or “human-aggressive,” must produce some other plausible explanation for the observed phenomena: why are pit bulls inflicting this much damage on this many human victims, when other dog breeds who are just as big and strong, raised in comparable conditions and comparably numerous, are not?
A flying start misses rate of acceleration
A related issue is that C-BARQ itself is not a measuring tool that starts from a neutral point.
If a dog does not already have behavioral issues that the owners, trainers, and handlers recognize and seek a diagnosis, the data pertaining to the dog is relatively unlikely to be entered into C-BARQ.
Thus, instead of being an objective assessment of normal dog behavior, C-BARQ is skewed toward abnormal dogs. Since aggression is among the most frequently problematic aspects of dog behavior, using C-BARQ to measure aggression relative to normalcy may be a bit like measuring vehicular acceleration not from zero to sixty miles per hour, but from twenty to sixty miles per hour.
Starting the measurement from twenty miles per hour would have the effect of missing how rapidly a car overcomes inertia.
Starting the C-BARQ measurement with dogs who are already skewed toward aggression may overlook the difference between how rapidly aggressive dogs become aroused as compared to normal dogs, who are never subjected to behavioral diagnostic questionnaires.
Threshold of perception
In addition, behavioral diagnosis usually begins at the point where the owner deems a particular behavior to be problematic. Owners who have acquired dogs precisely because they are of breeds with aggressive reputations may not recognize abnormally aggressive behavior, or consider it problematic, until that behavior goes so far beyond what other people would find acceptable as to result in legal issues.
A further problem is that C-BARQ accepts arbitrary breed definitions, which tend to split dogs into smaller categories than those among which there may be actual genetic differences of behavioral significance. This may tend to dilute the significance of shared traits, relative to other dogs.
For example, “Highly heritable and functionally relevant breed differences in dog behavior” splits the data for pit bulls among 288 dogs defined as American pit bull terriers, 101 dogs defined as American Staffordshire terriers, 103 dogs defined as Staffordshire bull terriers, and 40 dogs defined as just bull terriers.
Only among pedigree snobs do these distinctions have any meaning. Reality is that all four of these “breeds” share a common history of breeding to fight, and many pit bulls could be enrolled with different registries as two, three, or perhaps all four of these “breeds.”
Behavioral data from the total pit bull sample size of 532 would therefore be more representative of pit bull traits than the more limited data from the four “breeds” counted separately––especially when offset by comparison to the data from 392 Rottweilers, a much narrower breed, historically outnumbered by pit bulls in the general dog population by a ratio of four or five to one.
Better dogs through chemistry?
Among the four co-authors of “Highly heritable and functionally relevant breed differences in dog behavior,” MacLean, unlike Serpell, has previously studied dangerous dog behavior from a perspective oriented toward prevention and treatment, rather than denial.
Reported McLean in a 2017 special edition of the journal Frontiers in Psychology, as summarized by a University of Arizona media release, “Existing interventions for dogs’ behavioral problems often target testosterone and serotonin. Pet owners commonly neuter male dogs to help manage levels of testosterone, which has been linked to aggression. And to manage levels of serotonin, which is believed to reduce aggression, some dogs are prescribed selective serotonin reuptake inhibitors, the most common type of antidepressants.”
MacLean, however, suggested that “Future research might consider new interventions focused on vasopressin and oxytocin,” two hormones that he believes to have more to do with dangerous dog behavior.
Elaborated MacLean himself, “It would be reasonable to think that if vasopressin facilitates aggression, you could develop pharmaceuticals that could target the vasopressin system to help in cases where dogs are really aggressive.”
But why not just stop breeding pit bulls?
Much simpler, however, would be to just stop breeding dogs with a genetic predisposition toward violence.
“Highly heritable and functionally relevant breed differences in dog behavior,” though it is not easy reading, helps to establish that this predisposition exists.
Of course this was already known long ago to John P. Colby, Charles Werner, J.D. Johnson, Earl Tudor, and the other 19th and 20th century dogfighters, a few dozen in all, who bred some of the ancestors of practically every pit bull alive today.
That is very interesting, and is something I have been saying for 50 years. However, dogs were used not just to fight each other, and not just in the past 300 years. There are many drawings of dogs fighting with bears, lions, bulls and even humans. Google it and you will find some. Some present day dogs come from these lines bred for aggression for thousands of years.
Merritt Clifton says
No one is disputing this. What MacLean, Snyder-Mackler, vonHoldt and Serpell wrote is that, “Although dogs have been bred by humans for millennia, the formalization of modern breeds, as defined by closed genetic pools, occurred only 200–300 years ago.” ANIMALS 24-7 then pointed out that, “A dog breed category developed over 750 years or longer, including herding dogs and fighting dogs, would thus have gone through at least as many generations as humans have since the most recent common ancestor of all humans alive today.”
The look of many varieties of dogs were created simply by geographical isolation, People breeding dogs to guard their herds did not set out to make them look certain way; they chose those best suited to the task. Looking at ancient art works you can see dogs came in many sizes and shapes for a long time. There are the molossus type war dogs and sighthounds as well as small pet dogs. Of-course, they look bit different from present day relatives; all breeds have changed significantly even since 1873 when the stud books were closed. Everything evolves.
Merritt Clifton says
“Molossus-type war dogs” were scarcely created by geographical isolation. On the contrary, they were deployed and transported by the Romans from one end to the other of an empire covering most of Europe and the Mediterranean region, and were bred specifically for the attributes of fighting dogs, including large jaws, heavy musculature, the grip-and-shake biting style, and willingness to attack and fight even much larger humans and animals (such as war horses) to the death.
I did not say that Molossus were created in geographical isolation. I said they existed for very long time. Obviously they were selectively bred for their size and aggressive nature, and as the Romans advanced, their dogs, no doubt, left behind some of their progeny, which then in some cases was bred on, and some then ended up as foundation stock for large fierce guard dogs, others may have lost some of the fierceness and were used more for their size. Why do you think that so many breed names are geographical?
Anyway, the point I was making was that pit bull type of dogs were not used just to fight other dogs. My point was a response to the question why nowadays are pit bulls “also” people aggressive.
James H. Mundy IV says
Outstanding research and presentation………………………
Joyce Grindstaff Howell says
Thank you for the research and assistance on this disturbing issue.