Critical Reasoning—on Aggression and Dominance

Dogs Head-To-Head

Is this head-to-head confrontation a display of aggressive or dominant behavior? We need good definitions to decide correctly for one or the other.

In the following, I will give you four definitions of aggression and dominance; and will analyze them determining whether and why they are good or bad definitions.

Behavior is tricky to study because it is highly dynamic depending on many variables. It is even more difficult to describe because nothing, or very little, is as clear cut as we would wish and our brains strive after. The difference between any two behaviors is like the difference between any two colors in the light spectrum. When is red not red but yellow is as difficult to assert as when friendly behavior turns into pacifying, pacifying into submissive, or submissive into fearful. It is, ultimately, a question of definition and convention.

Behavior and emotion are closely connected and yet we can only see the former and guess about the latter. Generalizations are troublesome and prone to error. These inherent difficulties combined with poor definitions and a sloppy use of the terms can only turn a difficult subject-matter into a nearly impossible one. My suggestion to attempt to remedy this is to formulate accurate definitions, to apply terms consistently, and to be modest in our claims and generalizations, leaving plenty of room for alternative hypothesis and improvement.

I will exemplify the problem with some representative statements about dominance and aggression, which I picked from articles on the Internet. The name of their authors is irrelevant in this context because my purpose is not to criticize single persons, only to point out fundamental errors and to help my readers to analyze what they read, to find their bearings, and develop their own educated opinions.

In all fairness, we must appreciate the problem any author faces when attempting to describe an everyday occurrence like behavior accurately, and living up to a scientific standard, an occurrence which we are used to registering, describing and conveying in daily language. It’s much more difficult than to write a mathematical sentence. “Between a rock and a hard place,” the author tries not to use too many technical terms, which require definitions on their own, putting him or herself at the mercy of everyday language and colloquialisms bound to a myriad of interpretations and misinterpretations. If you’re the author of one of the analyzed statements below, please regard my analysis as an aid. I’m definitely on your side, fully appreciating the burden of the task you so bravely undertook: to define and to explain.

Example 1: “In animal behavior, dominance is defined as a relationship between individuals that is established through force, aggression and submission in order to establish priority access to all desired resources (food, the opposite sex, preferred resting spots, etc). A relationship is not established until one animal consistently defers to another.”

The author states that dominance is one kind of relationship: (“dominance (…) is a relationship”) with the function to “(in order to) establish priority access to all desired resources.”

The statement doesn’t tell us explicitly what dominant behavior is, but leaves us guessing that since dominance “is established through force, aggression and submission,” it must be related to these behaviors; but it doesn’t define aggression and submission, which are technical terms.

The statement seems either biased or too restrictive. One individual can establish a dominant (priority access, according to the author) relationship with another through motivation, persuasion, argumentation, bluffing, bribing, all without the use of force or aggression (aggressive behavior), independently of how we exactly understand these terms.

Most social parents establish their dominance status over their offspring not by using aggressive behavior, but because they are better (older, more experienced) at solving problems (this is persuasion and argumentation) or only a limited and inhibited amount of force, which is quantitatively and functionally so different from aggressive behavior, that it deserves a new name (dominant behavior).

In some cases and for many reasons, most commonly age and experience, it is even the lower ranking (in need of protection) that bestows a higher ranking to the other (it’s a “you decide”). Aggressive behavior and dominant behavior are, thus, not the same.

hierarchy maintained by means of aggressive behavior tends to be unstable, either because some of the lower-ranking leave the group (estimating that the costs of group membership out-value its benefits), or because they constantly challenge the higher ranking. Any hierarchy maintained by means of dominant behavior tends to be more stable because the lower ranking individuals have greater benefits at relatively low costs.

The sentence “to all desired resources,” is too daring. I wouldn’t dare say all and I doubt very much that any individual ever controls all desired resources (depending on the number, of course, but all suggests many). No hierarchy is ever more consistent than it is regularly subject to changes. Even in established hierarchies, the highest ranking individuals do not always gain access to particular resources on particular encounters. Persistent dominant behavior from one individual toward the same other individual tends to turn their relationship hierarchical, but it has to be maintained by regular displays of dominant behavior.

The sentence is not a definition of dominant behavior because to be so we’d have to substitute dominance with dominant behavior and aggression with aggressive behavior, and we would get “dominant behavior is established through aggressive behavior,” which doesn’t explain anything.

If dominance and aggression are the same, there is no need to call it something else: aggression would do. The term submission seems misplaced: “dominance is established through submission” is a contradiction since one is the antonym of the other. The sentence doesn’t seem to make sense once we begin analyzing it, but I have a hunch of what the problem is: the author is confounding hierarchy with dominance. If we substitute one term by the other, then the sentence makes sense—a relationship can be hierarchical—though, it is still wrong because a hierarchical relationship does not necessarily involve force and aggressive behavior; it can involve many other aspects as we saw.

The sentence “A relationship is not established until one animal consistently defers to another,” can be misleading. It gives the impression that no relationship at all exists between two individuals unless one consistently defers to the other, which is not true. There are many examples of relationships, most as a matter of fact, be it between humans, wolves or dogs, where one of the parties does not consistently defer to the other. Quite the contrary, in most stable relationships, both parties consistently defer to one another.

In the statement above, and in general, we confound dominance with dominant behavior; we confound a defining characteristic with an attribute.

A dog is not dominant as it is black or short-legged. A dog shows dominant behavior toward (condition 1) when (condition 2).

“A knife is a cutting tool” indicates a defining characteristic (permanent). “This knife is sharp” indicates an attribute (temporary). The correct is: “This knife is sharp to cut meat (=condition 1) today (=condition 2).”

Likewise, “Bongo is dominant toward Rover (=condition 1) when they find a bone (=condition 2).” More correctly, we should write “Bongo showed today dominant behavior toward Rover when they found a bone.” We can also say, if it is the case, “Bongo shows, more often than not, dominant behavior toward Rover when they find an edible item”.

This is characterizing the behavior, not the individual. What we cannot say is “Bongo is dominant, Rover is submissive” because these are not defining characteristics of an individual.” These are attributes of their behavior under certain conditions even when these conditions seem to be rather encompassing.

Bottom line: The statement is not good because: (1) it defies observational data (aggression is not a necessary condition in a dominance relationship), (2) if aggression is a necessary condition, then we don’t need to call it dominance, we can simply call it an aggressive relationship (or antagonistic if you prefer a nicer name), (3) it is too restrictive (‘all desired resources,’ ‘a relationship is not established,’ ‘consistently defers’).

Let me give you another example of a representative sentence, this time about aggression, taken from another article on the Internet.

Example 2: “So what causes aggression? Aggression is a response to something/someone the animal perceives as a threat. Aggression is used to protect the animal through the use of aggressive displays (growling, barking, tooth displays, etc.) or protect the animal through aggressive acts (biting). Aggressive behavior is most frequently caused by fear.”

We miss the definition of “threat” to be able to analyze the sentence conclusively, but assuming that threat means “an intention of imminent harm, danger, or pain” (The Free Dictionary), the statement is false.

A threat is only a threat if it threatens, that is if it succeeds in eliciting fear behavior (apprehended as imminent danger). Otherwise, if the threatened individual does not consider itself under imminent harm, danger or pain, then it is not a threat, it is just an exclamation (information).

Animals respond to threats (not exclamations), which are aggressive behaviors, by either neutralizing the opponent’s aggressive behavior (pacifying behavior), giving up (displaying active or passive submissive behavior), or fleeing (the most effective energy saving measure when seriously threatened). When threatened, attacking would be the worst possible strategy. If attacking solves the problem, then the threat was not a threat at all (by definition) because it was not dangerous enough to qualify as such.

The only occasion when some animals respond aggressively to threats (not all, some freeze and die) is when pacifying behavior, submissive behavior, and fleeing doesn’t have the desired effect of neutralizing the threatening (aggressive) behavior of an opponent. These are exceptional exceptions.

The sentence “Aggression is used to protect the animal through the use of aggressive displays (…)” is circular. Aggression being synonymous with aggressive behavior and display and act with behavior means that the sentence reads “Aggressive behavior is used to protect the animal through the use of aggressive behavior or protect the animal through aggressive behavior.”

Furthermore, since the author writes in the passive, it leaves us with the feeling that aggression is some mysterious quality. Who uses it? Where does it come from? Re-writing it in the active, we get the even more nonsensical “The animal uses aggressive behavior to protect itself through the use of aggressive behavior or protect itself through aggressive behavior.” This is not exactly a statement likely to enlighten anyone, quite the opposite.

The sentence “Aggressive behavior is most frequently caused by fear” is the result of our use of terms without having given enough thought to their proper definitions. Fear does not elicit aggressive behavior. It would have been a lethal strategy that natural selection would have eradicated swiftly and once and for all. A cornered animal does not show aggressive behavior because it is fearful. It does so because its natural responses to a fear-eliciting stimulus (pacifying, submission, flight) don’t work. 

Let us find out why a third statement, which I also took from the Internet, is much better than the two that we have analyzed above.

Example 3: “Dominance is not a personality trait but a description of a relationship between two or more animals and is related to which animal has access to valued resources such as food, mates, etc.” We can easily improve this statement by avoiding the unclear “is related to” and the passive construction of the sentence: “Dominance is not a personality trait but a description of a relationship between two or more animals; in a dominance relationship, the dominant animal has access to valued resources such as food and mates.” (We also drop the vague and unnecessary “etc.”).

This statement is not as restrictive as the first one we analyzed. For example, it does not claim aggressive behavior as a necessary condition. It allows for more options. It also says what the term being defined is not (“not a personality trait”).

The art of making a good definition is to find the right balance between too much and too little and to state only the absolutely necessary conditions. A too restrictive definition has fewer chances to be adopted than a broader one, and a too broad definition risks losing its defining function. This definition of dominance is more likely to be accepted and adopted by a larger number of people than the first one.

A proper definition defines exactly what is supposed to define and nothing else. A good definition gives an if-and-only-if condition for when an object or a term satisfies the definition. It is conclusive and exclusive. A good definition should also involve simpler terms than what we are defining.

Sometimes, we are compelled to use working definitions. A working definition is a definition we choose for an occasion and may not fully conform to the established definition. It is a definition at a stage of being developed.—a tentative definition that in due time may turn into an established definition.

Dogs Pacifying Behavior

Believe or not, the higher ranking (ethology term) or most distinguished (in human terms) dog is the one on the right. That is the function (and the beauty) of the evolutionarily stable strategy dominant/submissive behavior (when defined correctly). It resolves a confrontation without injury to either party, spending only a manageable amount of energy. The dog on the left shows pacifying behavior, which is a voluntary display of a mild form of submissive behavior. It’s politeness, recognition, in human terms.

Example 4: “Dominant behavior is quantitative and quantifiable behavior displayed by an individual with the function of gaining or maintaining temporary access to a particular resource on a particular occasion, versus a particular opponent, without either party incurring injury. If any of the parties incur injury, then the behavior is aggressive and not dominant. Its quantitative characteristics range from slightly self-confident to overtly assertive.”

This is a good definition because it defines a term including and excluding the necessary conditions. It defines something concrete and observable, not a semantic form (it’s easier to define “dominant behavior” than “dominance relationship”).

It states a necessary condition to distinguish it from a related technical term, aggressive behavior, even explaining a defining characteristic of the latter.

It does not include other terms needing a definition. It includes enough conditions to justify the use of the term, not too few to risking being synonymous with another term, and not too many to risking losing its explanatory value by being too encompassing.

It gives an example of its main characteristic in plain words. It does not presuppose any special knowledge of a reader to understand it. It’s almost impossible not to agree with this definition unless, of course, one wants the term to describe something completely different (which is another matter altogether).

It is not likely that we’ll find the truth, the only truth and nothing but the truth about behavior anytime soon, if ever, but I believe we can come closer to it by being meticulous in our observations, strict in our argumentations, modest in our generalizations, and prudent in our conclusions.

 

"Ethology" by Roger Abrantes

If animal behavior fascinates you, you will enjoy "Ethology—The Study of Animal Behavior in the Natural Environment," the book and course by ethologist Roger Abrantes.
Enrolling for this course puts you in direct contact with the author to whom you can pose any question while you complete your coursework. Click here to read more and enroll.

Aggressive Behavior—the Making of a Definition

Contrary to what you might suppose, aggressive behavior is difficult to define. “Lack of agreement regarding definitions of aggressive behavior has been a significant impediment to the progress of research in this area,” writes Nelson in 2005 in his big book ‘Biology of Aggression.’

Why is a good definition necessary? Because only then do we know what we are discussing.

I have never been quite satisfied with my own definition, and it nags me, worse than a mosquito bite, when I can’t come up with a good definition. I have often returned to it changing a comma or two to see if it improved. Alas, to no avail, updated versions were marginally better, but not resoundingly so.

Aggression 02

Aggressive and dominant behavior (erected ears, short mouth, big eyes) (picture dogsquad).

My original definition, let me remind you, was: “Aggressiveness (or aggressive behavior) is behavior directed toward the elimination of competition. It can range from displays of intent, like growling, roaring and stamping to injuring behavior like biting, staging, kicking.”

Not bad, but could be better. I checked many other definitions to analyze their strengths and shortcomings, hoping to get the necessary inspiration to come up with a really good definition.

A— “Aggressive behavior is behavior that causes physical or emotional harm to others, or threatens to. It can range from verbal abuse to the destruction of a victim’s personal property.” (www.healthline.com/health/). Not bad, but not precise enough to use in the biological sciences.

Aggression 02

Aggressive and submissive behavior (not fearful). This is the behavior a cornered dog shows when pacifying, submission and flight don’t work (picture from doggies).

B— “Aggression is a forceful behavior, action, or attitude that is expressed physically, verbally, or symbolically. It may arise from innate drives or occur as a defense mechanism, often resulting from a threatened ego. It is manifested by either constructive or destructive acts directed toward oneself or against others (Mosby’s Medical Dictionary, 8th edition).” Not bad either, though weakened by the passive voice. It recurs to terms needing strong definitions as well, i.e. drive, defense mechanism. Finally, it is a bit too psychological for the evolutionary biologist—what is a threatened ego?

C— “Aggression is behavior that is angry and destructive and intended to be injurious, physically or emotionally, and aimed at domination of one animal by another. It may be manifested by overt attacking and destructive behavior or by covert attitudes of hostility and obstructionism. The most common behavioral problem seen in dogs.” (http://medical-dictionary.thefreedictionary.com/Aggressive+Behaviour). This one is not good. It is more a list of synonyms (angry, destructive, hostility, obstructionism) than a definition. It mixes concepts together too easily (aggression and domination?). Finally, the most common problem in dogs in our files (over 10,000 of them) is home alone problems, not aggressive (whatever that is) behavior.

D— “Aggressive people often uses angeraggressive body language […] “(http://changingminds.org/techniques/assertiveness/aggressive_behavior.htm). This one, we won’t waste any more time with. A definition that uses the definiendum is not a definition.

Aggression 03

Aggressive and submissive behavior (ears down, long mouth, smaller eyes) (picture Cesar’sWay).

E— Aggression is a response to something/someone the animal perceives as a threat. Aggression is used to protect the animal through the use of aggressive displays (growling, barking, tooth displays, etc.) or protect the animal through aggressive acts (biting). Aggressive behavior is most frequently caused by fear.” (somewhere on the Internet). This one is not good either. Again, it uses the definiendum in the definition. We miss the definition of threat to be able to analyze the sentence conclusively. More seriously, it states that aggression is caused by fear, which from an evolutionary point of view doesn’t make sense. Fear does not elicit aggressive behavior. It would have been a lethal strategy that natural selection would have eradicated swiftly and once and for all. A cornered animal does not show aggressive behavior because it is fearful. It does so because its natural responses to a fear-eliciting stimulus (pacifying, submission, flight) don’t work. 

Aggression 04

Growling and snarling are also aggressive behaviors (picture askmen)

F— “Aggression is defined as behavior which produced or was intended to produce physical injury or pain in another person.” (Nelson, R. .J. 2005. Biology of Aggression. Oxford Univ. Press). This is a much better definition, but it could be more explanatory.

So, after yet another round of deliberation, here follows my suggestion.

“Aggressive behavior is behavior directed toward the elimination of competition from an opponent, by injuring it, inflicting it pain, or giving it a reliable warning of such impending consequences if it takes no evasive action. It is distinguishable from dominant behavior in as much as the latter does not include harmful behaviors though it may require some degree of forceful measures. Aggressive behavior ranges from reliable warnings of impending damaging behavior such as growling, roaring, and stamping, to injurious behaviors such as biting, staging, and kicking. Predatory behavior is not aggressive behavior.”

This is much better than earlier versions, and it complies with all the requirements for a good definition. It defines something concrete and observable. It states a necessary condition to distinguish it from a related technical term, dominant behavior, even explaining a characteristic of the latter. It does not include other terms needing a definition. It includes enough conditions to justify the use of the term, not too few to risk being synonymous with another term, and not too many to risk losing its explanatory value by being too encompassing. It gives examples of what is and is not aggressive behavior. It does not presuppose any special knowledge of the reader to understand it.

This is a good definition because it defines the term, including and excluding the necessary conditions. Whether it will be the last word on the matter is another story. I’m sure it won’t. There is always room for improvement. A good definition must also be able to accept reviews imposed by newer discoveries. Until then, I’m happier with this one than with any earlier versions. Aren’t you?

 

"Ethology" by Roger Abrantes

If animal behavior fascinates you, you will enjoy "Ethology—The Study of Animal Behavior in the Natural Environment," the book and course by ethologist Roger Abrantes.
Enrolling for this course puts you in direct contact with the author to whom you can pose any question while you complete your coursework. Click here to read more and enroll.