This note comments on the ideas contained
HISTORY & MISCONCEPTIONS OF DOMINANCE THEORY [ABOUT THE ALPHA ROLL],
by Melissa C. Alexander, 2001. After an introduction, the appropriateness
of ethological considerations with respect to human behavior is briefly
asserted. The existence of a study and conclusions ascribed to Dr. L.
David Mech is questioned, and observations about dominance in dogs, dog
training, and Ms. Alexander's account of a book she reports reading are
made. The note ends with suggestions for readers interested in applying
seemingly complex concepts to their day-to-day life with dogs and dog
training. [Keywords: alpha role, alpha roll, dog training, dominance,
man-dog interaction, dog behavior]
Many of the concepts described in the
HISTORY & MISCONCEPTIONS OF DOMINANCE THEORY [ABOUT THE ALPHA ROLL],
by Melissa C. Alexander, 2001, have complicated underpinnings. Her summary
treatment of disparate, complex topics is marked by both brevity and ease
of reading. Factors that contribute much to the behaviors she addresses,
such as the causality (of any specific dog behavior), mechanisms of canine
learning, the manner of canine perception, and the nature of canine
cognition are addressed only tangentially, if at all. Likewise, the actual
nature of Ms. Alexander's relationship with her dogs is not described. In
light of so many intangibles, detailed commentary is neither appropriate
I agree wholeheartedly with Ms. Alexander's
condemnation of the 'alpha roll.' That said, I disagree with the some of
the rationale for the condemnation advanced by her (primarily that also
ascribed to Dr. Raymond Coppinger). In my opinion, a dog can be trained to
adopt a physical posture -- one that signals submission in other contexts.
The success of such postural training can be trivial or difficult, in the
same way that training a dog to retrieve can be trivial or difficult. One
should not confuse these activities as being anything other than training.
Dominance is not being directly established by training the adoption of a
posture. Neither is one changing anything about the fundamental nature of
the dog, a nature that seems to incorporate the concept of hierarchy in
many of its manifestations.
To reiterate, a dog's capability to exhibit a
trained posture should not be confused with a dog's submissiveness. The
exhibition by a dog of a trained 'alpha-rolled' posture has no
relationship to the deference expressed in the behavior of a subordinate
dog when it interacts with dogs (or people) it regards as superior.
Without the panoply of contextual cues associated with an 'alpha-rolled'
posture that indicates submissiveness, e.g., the behavior being exhibited
in response to forceful aggression, the posture, in and of itself, has no
relevance to a superior-subordinate relationship in the dog's
frame of reference.
Like compulsive methods in general, the compulsive
training of a posture (whether or not the dog accords the posture uniquely
canine significance in certain contexts), can contribute to the
establishment of a dominant role by a handler. Such training can also lead
to unanticipated affect. The dog could react to such training as if the
trainer is participating in canine dominance-negotiation behavior, e.g., a
challenge. In this case, the trainer should be prepared to cope with the
dog's normal behavioral response. The trainer should understand what is
required to gracefully win such a confrontation. The trainer should, as
well, understand how to behave if necessary to resign and lose the
confrontation, such that the dog understands that the confrontation is
finished. The dog's perception of human factors, e.g., the level of fear
and 'force of personality' in the trainer, contribute to the outcome of
Although she makes no such assertion, I don't doubt
that Ms. Alexander occupies an alpha role with her dogs. I do, however,
doubt any direct attribution or relation of this to her training
exercises. In this note, I will address neither the efficacy of training
dogs via the employment of deferred reward schedules in general, nor the
cognition implied by the delayed and conditional aspects of the
'sit-then-act/behave' modality Ms. Alexander describes..
QUOTATION: "So what's the truth? The truth is dogs
aren't wolves. Honestly, when you take into account the number of
generations past, saying "I want to learn how to interact with my dog so
I'll learn from the wolves" makes about as much sense as saying, 'I want
to improve my parenting - - let's see how the chimps do it!'"
Dogs aren't wolves and wolves aren't dogs. That
said, they share certain fundamental physiological traits and behaviors.
Replicable research that has been formally described in peer-reviewed
journals indicates that certain extrapolations between dogs and wolves are
appropriate. Until other research that has undergone similar scrutiny
indicates the contrary, I'll choose to believe that much about dog
behavior can be learned from the study and research of wolves' behavior.
I presume that Ms. Alexander's intent behind the
comment about parenting in the above excerpt is sarcasm. I believe,
however, that truer words were never spoken. When comparing the innate and
learned behaviors related to parenting among different primate species
(e.g., chimps and humans), striking similarities are detected. This, of
course, does not mean that human parents should raise their children as if
the children were chimpanzees. Rather, it means that there is a lot to
learn about human parenting from the scientific study of chimp parenting.
Similarly, there is a lot to learn about canine behavior from the
scientific study of lupine behavior.
QUOTATION: "Dr. L. David Mech performed a 30-year
study on dogs at Yale and UC Berkeley. 19 years of the study was devoted
to social behavior of a dog pack. (Not a wolf pack. A DOG pack.)"
I know casually of Dr. Mech's work, by virtue of its
publication in journals (and secondarily on the Web). I believe Dr. Mech
is regarded by his peers as a subject-matter-expert on wolves and their
behavior. I was troubled with some of the conclusions attributed by Ms.
Alexander to Dr. Mech in her summary statements about his alleged
'dog-pack study.' This troubled feeling was replaced with disquiet when I
Dr. Mech's personal publication list. During the almost 30 years since
1962 when Dr. Mech was first published, there is no indication of a study
that resembles Ms. Alexander's cited study. While the cited study could
well exist, and indeed might contain the conclusions related in her
article, I can not find a reference to such a study by Dr. Mech.
Note. A correspondent reported that Ms.
Alexander's referenced study could in fact be referring to a study
conducted by Alan Beck. A quick web search revealed that Alan Beck, DVM
published a study titled
The Ecology of Stray Dogs. A Study of Free Ranging Urban Animals
(1973, Baltimore). I do not know whether Dr. Beck's study was the source
of the findings apparently erroneously attributed to Dr. Mech by Ms.
Alexander. I will gratuitously observe that the stray dogs about which Ms.
Alexander writes seemed to have adopted behaviors that in many respects
mimic behaviors found in a wolf pack.
QUOTATION: "Young puppies have what's called
"puppy license." Basically, that license to do most anything. Bitches
are more tolerant of puppy license than males are.
The puppy license is revoked at approximately four months of age. At
that time, the older middle-ranked dogs literally give the puppy hell --
psychologically torturing it until it offers all of the appropriate
appeasement behaviors and takes its place at the bottom of the social
hierarchy. The top-ranked dogs ignore the whole thing."
I am wary of anthropomorphisms when discussing
animal behavior. I suggest that the reader exercise the same caution.
Uniquely human values are at work when one characterizes animal behavior
as "literally giv(ing) the puppy hell" or as "psychological torture."
While a marauding wolverine might kill every chicken in the henhouse
without eating anything, describing the wolverine as a hen-torturer or
hen-murderer is ludicrous. Dogs, for example, can in certain circumstances
exhibit mobbing behavior (characterized as some members of a group
'ganging' up on, and attacking, a single animal). This seems cruel and
torturous to humans, but I'm certain that this is not a case of dogs
'torturing' another dog. (I'm sure of little else about the behavior,
except the need to preclude the occurrence of situations where mobbing
could be expressed by my group of dogs.)
What we humans view as torturous can be part and
parcel of a normal life for a dog. A dog that kills and eats a squirrel --
or 'even worse,' begins to eat said squirrel while it is still alive -- is
horrendous and cruel to us. Some would say that such 'vicious' behavior is
torture and undoglike and unnatural. One could anthropomorphize and opine
a belief that the squirrel briefly shares our human point of view. I do
not share either point of view. I also don't view a cat's 'playing' with a
mouse before killing it as torture. I believe it's part of the animals'
innate and learned behavior, and accept it as an animal behavior that is
not easily accommodated by human sensibilities.
QUOTATION: "To be "alpha," control the resources.
I don't mean hokey stuff like not allowing dogs on beds or preceding
them through doorways. I mean making resources contingent on behavior.
Does the dog want to be fed. Great -- ask him to sit first. Does the dog
want to go outside? Sit first. Dog want to greet people? Sit first. Want
to play a game? Sit first. Or whatever. If you are proactive enough to
control the things your dogs want, *you* are alpha by definition."
To paraphrase, 'Don't do hokey stuff like A, B, and
C, do hokey stuff like X, Y, and Z.' To reprise my refrain from above:
dogs can be trained; inductive and operant methods can be employed; this
is a long and complicated topic. I believe the activities described by Ms.
Alexander have much more to do with training than with impacting any
dominant-subordinate relationship between a handler and a dog. If "you are
proactive enough" to control situational and environmental influences to
your dog's behavior, then one can prevent the expression of behavior that
would otherwise be expressed, and the converse. Nothing more, nothing
QUOTATION: Reward deferential behavior, rather
than pushy behavior. I have two dogs. If one pushes in front of the
other, the other gets the attention, the food, whatever the first dog
wanted. The first dog to sit gets treated. Pulling on lead goes nowhere.
Doors don't open until dogs are seated and I say they may go out. Reward
pushy, and you get pushy.
This is the only instance I will give utterance to
the single word that most often came to mind as I read Ms. Alexander's
article. PUHLEEEZ! Ignore the dogs' hierarchy in your own
household? Knock yourself out ... at your own and your other dogs' peril
... and then try to repair the consequences. Uniquely human concepts like
parity, equality and egalitarianism are not applicable to dogs. Likewise,
reinforcement schedules that are effective while training humans will not
necessarily work with dogs. Anthropomorphisms such as those exhibited in
the above excerpt are inappropriate, and can even be dangerous -- with the
wrong confluence of events and the wrong dog.
QUOTATION: In a recent article in the
Association of Pet Dog Trainers (APDT) newsletter, Dr. Ray Coppinger
-- a biology professor at Hampshire College, co-founder of the
Guarding Dog Project, author of several books including
Dogs : A Startling New Understanding of Canine Origin, Behavior, and
Evolution; and an extremely well- respected member of the dog
training community -- says in regards to the dominance model (and alpha
rolling) ... [Note, the HTML links are not present in the original
article. Nor is mention made that Dr. Coppinger coauthored Dogs: A
Startling New Understanding ... with his wife, Lorna,
herself an author.]
I first learned of
Dr. Coppinger through Ms. Alexander's article. While I am pleased he
finds 'alpha rolls' at least as inappropriate as I do, I do not agree with
his rationale, at least as summarized by Ms. Alexander. Reviewing a
list of Dr. Coppinger's publications (which may or may not be
exhaustive), reveals at least one peer-reviewed article related to dog
behavior, as well as other areas of study.
QUOTATION: "That pretty much sums it up, don't you
I'd make other comments if I had but the time. Not
to try to prove myself right or Ms. Alexander wrong, but to try to further
illustrate and reinforce the idea that there is no shortcut to
understanding canine behavior and training dogs. One of the easiest ways
to go wrong, is to lose sight of the fact that your companion is a dog.
Irrespective of any surrogate role that the dog has in your life, it's not
a human, a child, or a baby, even if it is loved like one. It is a dog. If
it's a working Bouvier, then it is not just a 'dog.' It is an intelligent,
assertive, and powerful dog.
It is reasonable for the reader to conclude that I
disagree with many of Ms. Alexander's opinions, and that she probably
disagrees with mine. That said, readers of her article and this note
should not be led to believe -- even inadvertently -- that either article
is authoritative in any regard. I firmly believe that an owner can be
injured by his/her working dog (to whom the owner is unknowingly
subordinate), if the dog believes the owner is attempting to press the
issue. I believe that such an injury could happen even though the dog has
been trained to sit 'politely' before eating meals or playing fetch.
What should one do, especially the novice, to ensure
one has the alpha role with respect to one's dogs? Understand that
your role is decided by the dog, and then try to
ensure that the dog makes the 'correct' decision. Start with a sound
foundation of knowledge grounded in human psychology (so you understand
the whats, whys, and wherefores of human actions) and animal behavior (so
you gain the benefit of the lifetimes of research devoted to trying to
understand the whats, whys, and wherefores of dog actions). Start with
reading and comprehending an introductory, college-level psychology
textbook and a like textbook on animal behavior.
Add other specific academic and scientific
information, the relevance of which is established by your initial studies
and your personal inclinations. In my case, my 'advanced' studies focused
on movements/schools associated with a psychologist named
Piaget, and a behaviorist named
Anyone who puts pen to paper can seem like an
expert. Buttressed by a sound academic background (which does not have to
be formally earned by attending a college or university), read articles
and treatises related to topics of interest that have been published in
academia or other circles that foster (and in many cases require) peer
review. One of the benefits of peer-reviewed publications is that wheat is
winnowed from the chaff by experts. An example of such information is Dr.
Alpha Status, Dominance, and Division of Labor in Wolf Packs, that
was originally published in the
Canadian Journal of Zoology, 2000.
Build on this broad, stable foundation of knowledge,
by branching out and critically reading ancillary published information.
Pre-armed with the relevant knowledge, one can assess information of
unknown vintage or heritage, such as the new and novel ideas ascribed to
Dr. Coppinger's book by Ms. Alexander. Adopt the credo that printed matter
has received more scrutiny than electronically published material, and
therefore could be more valuable.
Only after the reader is knowledgeable -- and to a
certain extent, jaundiced -- should the musings of authors like Ms.
Johnson and me be read and assessed for validity. Understand the concepts
that the author is trying to convey, and then see how these concepts stack
up against your understanding of the world. Then, and only then,
incorporate such musings into your thinking, or not, as is your wont.
Everyone wants to get to heaven, but no one wants to
die. There is no shortcut to understanding working dogs or their
behaviors. It's not necessarily fun or even easy, either. But don't let
the apparent complexity scare you away, or detract from the fun that dog
ownership should be. If I had to offer one single item of advice, it would
be to view with suspicion everything that an author exhorts to be the
truth -- except of course the conclusion to this note. <grin>
1 September 2001
Arlington, Virginia, USA
ABOUT THE AUTHOR. I'm a computer geek who sometimes
finds the time to train my dogs in the
German-originated sport of Schutzhund. For twenty-odd years I owned
and trained exclusively Rottweilers. Then I met
Sheila Cronan and others who know and hold the
Bouvier des Flandres
breed in high regard as a working dog. I was especially impressed when
watching Butch and his dog, Brinks, work and trial in Schutzhund.
I now own two Rottweilers and a Bouvier. I'm
regularly reminded that my Bouvier is 'smarter' than other dogs that have
enriched my life in the past. All of my dogs are intact males that joined
my household at two-year intervals when they were around ten weeks of age.
They are kept in my home unless they must be kenneled because I have to
travel. These working dogs with strong temperaments peaceably coexist --
albeit sometimes only at my insistence -- in part because I control
situational and environmental aspects such that the dogs should not 'feel
the need' to display overtly aggressive, confrontational behaviors.
People who perceive a similar basis for some dog
behaviors and comparable behaviors of wolves might comment that this
'pack' coexists peaceably for some of the same reasons as does a wolf
- Such folks might opine that the peace would
evaporate if I introduced a bitch or a bitch's influence (such as
proximity to a bitch in heat) into the mix.
- They might link the peaceableness of this group
to my controlled, gradual introduction of only puppies into the group
(and ensuring that puppy/group interactions only take place under my
- They might express the belief that the stability
and lack of overt aggression in this 'peaceable pack' could be rocked if
I tried to introduce an adult member into my group.
I'd have to agree with such folks. I'd also point
out that there are other similarities between some aspects of my group of
dogs, and corresponding aspects of a pack of wolves.
There are a number of sound reasons that this group
of dogs coexists, chief among them my management of the dogs' environment
and situations -- availing myself of that uniquely human capability to
predict possible future outcomes based on observing present events. This
should not be confused with soothsaying or necromancy. Knowledge is the
- Experiential knowledge. One must
observe and learn the manner that one's dogs can be expected to behave
in a variety of situations, and pay attention to your dogs in all their
activities and actions. The dog should be the focal point of your
observations, which can be difficult.
- Academic knowledge. One must
learn the academically accepted factors that contribute to the
expression of behavior, cognition, and learning in man and dogs and
their close relatives.
- Theoretical knowledge. One must
review and learn from the reports of the observations of others. From
such observations and their descriptions and studies, come new
hypotheses and alternative points of view. These hypotheses and
resultant theories are tested by the community (academics, researchers,
and practitioners like us dog trainers/owners). If it stands up to such
scrutiny, the resulting proven theory will eventually be incorporated
into the corpus of academic knowledge.
Popular knowledge is not mentioned above. The true
validity of hot trends and new topics are difficult for me to assess,
except perhaps retrospectively. I don't ignore these concepts, but I don't
let them be a principal determinant in establishing my actions.
The concept of the alpha role in dog ownership has
for some time been fodder for the popular press. Perhaps in part as a
result of some of this popularity, I believe my dogs view me as their
Alpha, and further, that my alpha role results from my attempts to
communicate with them in a manner that they can 'understand,' among a
variety of other reasons. It seems only logical that I might more likely
succeed learning to communicate in a manner that a dog can understand,
than the converse. After all, we humans are smarter than dogs ... right?
(To overstate the obvious, the quotations from Ms.
Alexander's article -- with its prominent copyright notice -- are
excerpted for the purpose of criticism as envisioned by the doctrine of
* * * * *