By Sam Basso
OVERVIEW: Household dogfights can involve competition, pain, social interference, learning and changing relationships. Understanding them means studying what each dog actually did, not forcing the incident into a fashionable theory.
When dogs that live together begin growling, threatening or biting one another, the answer is rarely as simple as “dominance,” “jealousy,” or “resource guarding” or “they don’t see you as the leader.”
Sometimes the dogs have a dispute that ends without contact. Sometimes both participate in a physical fight. Sometimes one dog attacks while the other tries to get away. Those events do not necessarily have the same explanation or call for the same response.
To me, this is also a larger lesson about dog training. We can change behavior through teaching and management, but we cannot make the biology of dogs, the principles of learning, the effects of pain or the history of a social relationship disappear. A theory ought to help us understand the dog. The dog should not be expected to validate your theory.
Is every disagreement between two dogs a fight?
No. Two dogs can disagree without physically fighting, and that difference is worth preserving. One may stiffen or growl as the other approaches. The approaching dog may change direction, slow down or leave, and the interaction ends. A dispute has occurred, but it would be misleading to record a reciprocal physical fight when none was observed.
A physical fight involves aggressive violent contact by both dogs during some part of an encounter. A unilateral attack involves one dog making aggressive physical contact while the other is not observed attacking back during that phase. An encounter can also change form. Dogs may initially exchange bites, after which one withdraws and the other continues to pursue it. A single label applied to the whole incident can conceal that transition.
This does not mean we should let dogs “sort it out” regardless of the danger. Communication can sometimes prevent conflict from escalating. It can also fail, be missed or be disregarded. The fact that growling is a form of communication does not make it safe to stage another confrontation or leave dogs unsupervised after a serious injury.
Why do popular training explanations sometimes make the problem worse?
Because a familiar, popular word or theory can replace an investigation. A dog growls near a sofa, so one trainer calls it dominant. Another says it is guarding the sofa. Someone else decides the owner has failed to be a leader. Before anyone has watched the interaction, the proposed solution may already be punishment, forced surrender or stricter obedience.
But what if the dog had been sleeping? What if its neck hurt when it turned? What if the other dog had followed it from room to room and ignored earlier attempts to withdraw? What if someone grabbed its collar just before the snap? Each question identifies a circumstance we could investigate. None establishes a cause on its own, but all provide more information than a personality label.
The danger applies to fashionable new language as well as old language. “Overstimulated,” “reactive,” “dysregulated”, “resource guarder”, and “traumatized” can become convenient answers if nobody defines what the words mean in that particular case. Technical vocabulary is useful only when it improves observation and reasoning. A new label is not an explanation merely because it sounds scientific.
What would it look like to study the actual encounter?
I reconstruct what happened before. Consider this constructed illustration, not a documented client case: one dog is lying on the sofa as another enters the room. The approaching dog moves closer. The resting dog turns its head away, stiffens and growls. The approaching dog leaves; the first dog remains where it was.
| Phase | Dog A, resting | Dog B, approaching | What remains unresolved |
| Before | Lies on sofa | Enters room | Past encounters, pain, quality of rest and ability to leave |
| Approach | Turns head away | Moves closer | Whether Dog B noticed or responded to the earlier signal |
| Warning | Stiffens and growls | Stops approaching | Whether response relates to intrusion, anticipated contact or previous learning |
| Ending | Remains resting | Walks away | Whether both are comfortable in other settings and on later occasions |
The record describes a noncontact dispute, not a physical fight. It also does not prove that the resting dog was guarding a possession. We have not shown that either dog wanted the sofa for itself, that Dog A intended to control Dog B, or that fear, pain or frustration was the specific mechanism.
Now change one part of the sequence. Suppose Dog B continues to approach, blocks the exit, or follows Dog A when it tries to leave. The behavioral problem is different even if the sofa has not moved an inch. Suppose instead Dog A leaves the sofa to pursue Dog B after it withdraws. That changes the assessment again. What each participant does, and what happens after an attempt to disengage, matter more than the name we give the furniture.
We also need the ordinary days when nothing happens. Does Dog B walk past the sofa safely at other times? Can both dogs rest in the same room if there is room to move away? Do the problems appear after visitors arrive, after physical exertion, or when a particular person intervenes? Comparison is part of evidence, not a distraction from the incident.
Where are dogs more likely to run into trouble at home?
It is important to learn and have a heightened awareness that some locations and routines repeatedly appear in clinical histories because they bring dogs together, create competition or restrict movement. They are situations worth watching, not a proven ranking of the most dangerous places. Without knowing how often specific dogs share each space and how many uneventful encounters occur there, we cannot calculate a room-specific probability of fighting.
A doorway does not make a dog territorial. Food nearby does not make a dog a “resource guarder.” One dog may be trying to obtain something, another may be trying to keep what it already has, and a third may only want room to leave. Those are different relationships between the animal and the situation. The word resource is not sufficiently precise to explain any of them by itself.
Why do dogs that lived peacefully together begin having trouble?
Because the dogs and the circumstances can change. A puppy grows larger and more persistent. A newly adopted dog joins an established household. An older dog becomes less mobile or hears less clearly. A previous bite changes what one dog expects when the other approaches. A household that once offered easy ways to retreat may begin forcing repeated close encounters.
Pain deserves an early place in that inquiry. Ear disease, dental pain, arthritis, spinal disease and other medical conditions may affect sleep, mobility and tolerance of physical contact. A dog that once accepted a companion jumping alongside it may respond differently when those landings hurt. A veterinary assessment is important when behavior changes suddenly or handling, movement or rest appears uncomfortable. Finding a painful condition, however, does not prove it caused every fight, and improvement after treatment may coincide with changes in supervision or exposure (Mills et al., 2020).
Development also matters, but not as a magic timetable. Scott and Fuller’s work showed that social behavior develops and differs across individuals and studied groups of dogs. Their work does not justify announcing that every household conflict begins at a particular birthday or that breed alone predicts an attack (Scott & Fuller, 1965). A dog’s actual developmental and relationship history is more informative than an attractive general rule.
Are people sometimes part of the conflict?
Certainly, and we should investigate their actions without automatically blaming them. Two dogs rush toward an arriving owner. Someone grabs one collar, pushes the other aside, raises a voice or repeatedly forces the dogs to meet after a previous confrontation. Every one of those actions may change distance, movement, frustration and the opportunity to disengage.
Sometimes the person did nothing wrong and simply happened to be present. Sometimes an owner sees more incidents because the owner is home to witness them. And sometimes human handling or a poorly planned introduction changes the conditions enough to make conflict more likely. To understand which account fits, I want the timing: what did the people do, what did the dogs do next, and what happened in comparable situations without that intervention?
This is where I agree with the value of scrutinizing human interference. The conclusion must remain specific to the event, however. The claim that human interference is the single greatest cause of all household dogfights is not established by the research. Nor is “the owners caused it” an adequate substitute for observing both dogs.
What does the science tell us, and what doesn’t it tell us?
Three influential clinical investigations examined 73 household cases within a larger 1996 series, 38 referred cohabiting pairs in 2011 and 305 referred pairs in 2020 (Sherman et al., 1996; Wrubel et al., 2011; Feltes et al., 2020). Their reports describe familiar circumstances involving excitement, human attention, food, objects, and cramped access. They also describe important injuries and varied household outcomes. But these were animals whose owners sought professional help, not a representative sample of all multidog households.
The studies cannot tell us that an object, owner or doorway was the proven cause of a particular bite. They cannot turn a frequency reported among referred cases into the probability that an arbitrary pair of dogs will fight. And when owners report improvement after several interventions, that does not establish which part of the treatment caused the change. Dogs who have had no further fights under permanent separation are not thereby proven to have recovered the ability to live freely together.
We also need research outside veterinary referral services. Ethology, learning theory, developmental science and comparative studies ask different questions. Tinbergen’s classic framework distinguishes immediate mechanisms, how a behavior develops in an individual, its evolutionary history and its possible biological function (Tinbergen, 1963). None of those questions can be answered by the simple observation that a bone lay nearby.
Pavlov’s work on learned associations, Scott and Fuller’s work on development, and comparative studies of wolves and dogs can inform hypotheses. They do not identify an unseen motive in yesterday’s kitchen incident. The way to respect science is not to recite famous names. It is to use each finding for the question that study examined, while acknowledging its limits.
What does philosophy have to do with a dogfight?
More than it first appears. The Stoics asked people to distinguish what lies within their influence from what cannot simply be wished away. For dog training, the practical lesson is not fatalism.
We can arrange environments, prevent injuries, teach new responses, and create opportunities for dogs to develop better relationships. But we cannot repeal the laws of nature, erase an animal’s history, or simply command away pain, fear, competition, and the effects of previous experiences because they are inconvenient to our intentions or preferred training methods. Nor can we guarantee that two dogs will become friends, trust one another, or even tolerate living together, regardless of how much time, effort, or expertise we bring to the problem.
A relationship involves both dogs, and neither the owner nor the trainer has unilateral control over how that relationship develops.
We may be able to influence their behavior and improve the circumstances in which they interact, but sometimes the relationship itself cannot be repaired to the degree we would like. In those cases, preventing further conflict, protecting each dog’s welfare, or arranging permanent separation may be the responsible outcome. The Stoics understood that wisdom requires recognizing the limits of our control and influence. In dog training, that means accepting that our best efforts do not entitle us to our preferred results. Success is not always getting the dogs to live together. Sometimes success is recognizing when they should not.
Karl Popper’s philosophy of science offers an equally useful challenge: what evidence would make us change our minds? Suppose I am convinced that a dog growls because it wants to control the household. What would count against that explanation? The same behavior when the dog is injured? Safe coexistence when its escape path is open? A different pattern after an unwanted approach ends? If no conceivable observation could change the diagnosis, I may be defending a belief rather than investigating a dog (Popper, 1963).
There is an important limitation to the argument about nature. Natural does not mean good, and understanding social conflict does not mean leaving animals to injure one another. Teaching is itself part of an animal’s capacity to adapt. The ethical challenge is to understand behavior well enough to change dangerous circumstances and teach safer responses without pretending that training eliminates biological needs or uncertainty.
There is another important consideration: the dog has a say in how things turn out. Like the Stoics, we would be wise to recognize the difference between what is within our control, what we can influence, and what lies beyond either. We can control many aspects of a dog’s environment, decide how we teach, manage opportunities for conflict, and influence behavior through learning and experience. But we cannot directly command a dog’s emotions, erase its history, eliminate its biological needs, or guarantee how it will respond to another animal. Nor can we dictate how two dogs will ultimately feel about living together. Each dog brings its own abilities, preferences, experiences, and responses to the relationship. Training is not something we simply do to a dog, with the animal serving as a passive recipient of our intentions. It is an interaction in which the dog participates, responds, learns, and sometimes makes choices we would prefer it did not make. Our responsibility is to create the best conditions we reasonably can, while recognizing that influence is not the same thing as control, and compliance is not necessarily the same thing as cooperation.
How should an owner respond after the dogs have fought?
First, prevent further injury. After bites, substantial wounds, pursuit of a retreating dog or repeated escalating confrontations, secure separation and prompt veterinary or qualified behavioral assessment are safer than arranging a test reunion. Bite wounds may be hidden by the coat. Never place your hands between fighting dogs or grab into an active fight.
Once the dogs are safely separated, write down the sequence while it is still remembered. Record who approached, what each dog did, what the people did, whether either animal could withdraw, when contact began, how the encounter stopped and what followed. Keep owner reports, firsthand observation and video findings distinct. Also record times the dogs shared space without problems, the amount of contact they actually had and any medical or routine changes. Do not provoke another conflict to prove a hypothesis or punish a warning simply to make the house quiet.
Progress needs more than a count of days without fighting. Are the dogs choosing to share space, resting without repeated intrusion, leaving one another alone when asked by their behavior, and moving away freely? Or are injuries prevented because the dogs are always separated? Safe long-term separation can be a responsible outcome; it is not the same as restored unrestricted compatibility.
Why should dog trainers keep studying, even after decades of experience?
Because experience can sharpen judgment, but it can also make an old explanation feel certain. If every case becomes “dominance,” every growl becomes “guarding,” and every setback becomes “the owner isn’t firm enough,” there is little reason left to examine the dog. A method that seems successful because a warning has stopped may not have relieved the animal’s discomfort or improved its safety.
I must apply the same critical standards to my own explanations that I apply to those of others, because I am certainly not omniscient. If I believe a problem is caused by “X,” I should actively look for evidence that might contradict that explanation rather than merely seek information that supports it. If I suspect pain, the dog’s medical findings, behavioral history, and the timing of any changes become important. If I suspect human interference contributed to an incident, I should examine what happened when people were involved and compare it with occasions when they were not, without deliberately creating dangerous confrontations. Complicating matters further is the fact that the information available to me is almost always incomplete. I generally rely on owners’ recollections and interpretations of events, which may be understandably imperfect. By the time I become involved, the dogs are often being kept apart, so I may never witness the interactions in question. Video recordings are frequently unavailable, behavioral histories may be incomplete or poorly documented, veterinary records may reveal only part of the picture, and information about a dog’s breeder, genetic background, relatives, or offspring may be nonexistent. Even when such information is available, it may not establish what caused a particular incident. Consequently, I am often attempting to reconstruct complex events from fragmentary evidence, some of it secondhand and some of it impossible to verify. Professional experience can help me recognize possibilities and ask better questions, but it cannot turn missing information into established facts. A sound explanation must remain provisional where the evidence is incomplete, and I must always be willing to revise or abandon it when better evidence becomes available.
Dogs do not read our training books or philosophical arguments. They respond to the animals, people, sensations and circumstances around them, shaped by their individual histories. When housemates begin having problems, the first obligation is to keep them safe and examine what happened. The moment defending the theory becomes more important than understanding the dogs, we have stopped studying behavior.
Can a dog trainer guarantee that two dogs will stop fighting?
No responsible dog trainer can guarantee that every conflict between two dogs can be resolved, that every future fight can be prevented, or that two dogs who have become incompatible will eventually live together peacefully. A relationship involves both dogs, and neither the owner nor the trainer has unilateral control over how that relationship develops. Each dog brings its own temperament, physical condition, developmental history, experiences, preferences, and responses to the other. We can influence those responses through training, management, and changes to the environment, but we cannot dictate them. Nor can we guarantee that every factor influencing the relationship has been identified or can be changed.
This is an important distinction for people seeking professional help. Hiring a dog trainer is not purchasing a guaranteed outcome. The trainer may identify problems the owner has overlooked, improve communication, teach useful behaviors, recommend veterinary examinations, reduce opportunities for confrontation, and help establish safer household routines. Those efforts may substantially improve the situation. But improvement is not necessarily recovery, and even an extended period without fighting does not prove that another fight cannot occur. Sometimes the dogs are getting along better. Sometimes they are simply being kept out of situations in which they previously fought. Sometimes the relationship remains unsafe despite conscientious efforts by everyone involved.
A trainer’s responsibility is to provide competent assessment and guidance, communicate what they perceive to be significant risks and uncertainties, recommend reasonable precautions, and adjust the plan as new information becomes available. The owner’s responsibility includes understanding those precautions, providing accurate information, implementing agreed management, and reporting changes or additional incidents. Neither party should mistake effort, expense, good intentions, or professional credentials for a promise of success. Equally, uncertainty about outcomes does not excuse careless handling, unreasonable advice, or avoidable exposure to serious danger.
This is where expectations can become particularly troublesome. An owner may reasonably hope that professional assistance will restore the relationship, while the trainer may only be able to offer an opportunity for improvement. If that distinction is not explained at the outset, the owner may interpret an unsuccessful outcome as a failure to deliver what was promised. There is a difference between promising to use one’s professional knowledge and reasonable care to address a problem and promising that the problem will be fixed. Dog owners should seek to understand that distinction before they commit their time, money, and hopes to a course of training.
Sometimes the most important professional judgment is recognizing that the outcome everyone wants may not be achievable with acceptable safety and welfare. Two dogs may never become friends. They may require permanent management or separation, and occasionally one may need to live elsewhere. Such outcomes are not necessarily evidence that the trainer failed or that the owner did not try hard enough. They may reflect the limits of what can reasonably be accomplished in that particular relationship. Success is not always making two dogs live together. Sometimes success is recognizing that they should not, before another animal or person is seriously injured.
Glossary
Dispute: An interaction involving incompatible actions or interests that may end without aggressive physical contact.
Reciprocal physical fight: An observed phase in which both dogs engage in aggressive physical contact, without implying equal initiation or motive.
Unilateral attack: Aggressive physical contact by one dog during a specified phase when the recipient is not observed attacking back.
Antecedent: A circumstance preceding a behavior. Timing by itself does not prove causation.
Disengagement: Ending or reducing participation in an interaction, such as moving away.
Competition: Two individuals seeking incompatible access to something under given conditions, not a diagnosis of emotion or aggression.
Possession retention: Continuing to hold or use something already possessed, which differs from attempting to acquire it.
Resource guarding: A source-dependent term used in dog behavior literature. It is not, on its own, an explanation of cause, motivation or treatment.
Generalization: A learned response occurring in settings different from those where it was originally trained.
Management: Changes to environment, routine or access that reduce risk without necessarily changing the underlying relationship.
Falsifiability: The capacity of a proposition to be challenged by observations that could show it is mistaken.
Scholarly bibliography
Cafazzo, S., Marshall-Pescini, S., Lazzaroni, M., Virányi, Z., & Range, F. (2018). The effect of domestication on post-conflict management: Wolves reconcile while dogs avoid each other. Royal Society Open Science, 5, 171553. https://doi.org/10.1098/rsos.171553
Feltes, E. S., Stull, J. W., Herron, M. E., & Haug, L. I. (2020). Characteristics of intrahousehold interdog aggression and dog and pair factors associated with a poor outcome. Journal of the American Veterinary Medical Association, 256(3), 349–361. https://doi.org/10.2460/javma.256.3.349
Mills, D. S., Demontigny-Bédard, I., Gruen, M., Klinck, M. P., McPeake, K. J., Barcelos, A. M., Hewison, L., Van Haevermaet, H., Denenberg, S., Hauser, H., Koch, C., Ballantyne, K., Wilson, C., Mathkari, C. V., Pounder, J., Garcia, E., Darder, P., Fatjó, J., & Levine, E. (2020). Pain and problem behavior in cats and dogs. Animals, 10(2), 318. https://doi.org/10.3390/ani10020318
Pavlov, I. P. (1927). Conditioned reflexes: An investigation of the physiological activity of the cerebral cortex.Oxford University Press.
Popper, K. R. (1963). Conjectures and refutations: The growth of scientific knowledge. Routledge & Kegan Paul.
Scott, J. P., & Fuller, J. L. (1965). Genetics and the social behavior of the dog. University of Chicago Press.
Sherman, C. K., Reisner, I. R., Taliaferro, L. A., & Houpt, K. A. (1996). Characteristics, treatment, and outcome of 99 cases of aggression between dogs. Applied Animal Behaviour Science, 47(1–2), 91–108. https://doi.org/10.1016/0168-1591(95)01013-0
Tinbergen, N. (1963). On aims and methods of ethology. Zeitschrift für Tierpsychologie, 20(4), 410–433. https://doi.org/10.1111/j.1439-0310.1963.tb01161.x
Wrubel, K. M., Moon-Fanelli, A. A., Maranda, L. S., & Dodman, N. H. (2011). Interdog household aggression: 38 cases (2006–2007). Journal of the American Veterinary Medical Association, 238(6), 731–740. https://doi.org/10.2460/javma.238.6.731