Posts Tagged ‘wolf behavior’

I came across a story of a wolf that learned to retrieve ducks from a water spaniel. This story appears in Durward Allen’s The Wolves of Minong (1979):

This particular story clearly suggests that wolves are capable of learning to do thing that dogs do.

However, they do them at their own pace and do not respond well to being forced.

I seriously doubt that anyone could force-fetch this wolf– and still have all of his fingers!

The truth is this animal learned through the example of Junie and the trust he had in the Smitses.

Very few wolves in captivity are kept in this fashion. Most people who keep wolves try to either keep in a way that they behave as naturally as possible.

Those studies that have tried to keep wolves exactly like dogs in urban environments have also discovered that they can’t be forced to obey in the same way.  They are also less interested in learning from people than Western dog breeds.

But I don’t think they have tested wolves that have been raised with Western dogs on their ability to learn from the dogs.

In this case, you have a dog from an easily trained Western breed. Because the Smitses were operating in Michigan’s Upper Peninsula, I am assuming that Junie was an American water spaniel, which is the regional retriever.   This breed has been bred for many generations to work closely with human handlers, and because it is a retriever, the tendency to bring back objects to its handler is a trait for which it has been selectively bred for many generations.  Although it the trait must be refined through training, it is very easy — in relative terms– to teach a water spaniel to bring back a duck without plucking it or trying to consume it.

This wolf just happens to like the water spaniel, which is likely an elder. In wolf society, pups learn from their elders. They learn which prey they are supposed to hunt, and they learn how to use their predatory motor patterns to catch prey within a cooperative pack. This wolf learned from the water spaniel that the way one uses its motor patterns is to run out, pick up a wounded or dead duck, and come running back to the humans.

I bet that one could train certain wolves to do this behavior, even without the water spaniel as the mentor. I bet that some wolves could even learn to ignore gun shots near them. However, training a wolf doesn’t mean that you just force yourself onto them, which unfortunately seems to be the paradigm in which most people operate regarding wolves.  It is unlikely that anyone was able to domesticate a wolf by dominating it, for if you actually read the wild wolf literature, extremely aggressive dominance displays between individuals ultimately lead to the dispersal of one of the wolves. In captive situations, there is no dispersal, and those that don’t get along wind up doing a lot of fighting and displaying– which is assumed to be natural behavior. If paleolithic hunter-gatherers tried this on their camp wolves, they wouldn’t stay around for very long, and a certain percentage of them tend to wander off at mating season anyway.

One of the real problems in comparing wolf and dog behavior is that researchers often use the popular Western dog breeds, which are often gun dogs and herding breeds, and northern wolves, which are actually quite specialized wolves. These two animals are going to have extreme differences in behavior.  This type of wolf is going to be naturally quite cued into other canines, while the dogs are going to be very, very cued into people.  There is a huge debate about whether wolves are smarter than dogs, which has been re-ignited when Eotvos Lorand University’s Department of Ethology began doing these comparative cognition studies with wolves and dogs.

It is often said that wolves are capable of observational learning and that dogs can learn only by association, but dogs are actually capable of learning from both people and other dogs.

I am of the view that no wolf, no matter how well-socialized, will ever be able to perform at the level of a gun dog or a herding breed when it comes to word and body language associations from humans. There will never be a wolf like Rico the border collie. Dogs are also able to get a lot more information because they are willing to learn from us– and they basically have to. There is probably a genetic basis to this difference, but we haven’t actually found it.

But there aren’t enough wolves living as intimately with people as dogs do, so we really have problem making generalizations about wolf behavior. And because we have such a relatively low n in these studies, we probably aren’t going to be able to answer the question about which animal is more intelligent– if that is even a proper scientific question to work with in the first place.

Wolves are just very hard to keep in domestic situations. They are too emotionally reactive– likely the result of  the selecton pressures on their populations that came from centuries of persecution– and they are too energetic. The Russians say that “A wolf is kept fed by its feet,” which means that wolves are meant to travel vast distances every day in search of prey. In a home, this animal will be like a  field-bred pointer, a foxhound, or Dalmatian.  It will be so full of pent up energy that it might have a hard time focusing when the person arrives home to do some training with it.

But the story of Big Jim shows that at least some wolves are capable of learning to do dog behaviors. I don’t think we’ve figured out what the big differences between dogs and wolves actually are. I certainly don’t think we’ve figured out which species is more intelligent. However, I do think the dogs from Western breeds that have been bred to work closely with handlers that live very closely with people have some traits that are very unique, and most household dogs are going to receive a wealth of information from humans that even a very socialized wolf might not be open to learning. It may be the result of nothing more than the much higher emotional reactivity and energy levels on behalf of the wolf, but it may be something fundamentally cognitive.

I don’t think we have the answers yet.


Minong is the Ojibwe word for Isle Royale, an island in Lake Superior off  the coast of Michigan’s UP.  The Smitses were operating in Rock Habor, Michigan, which is the harbor that provides access to Isle Royale. The Smitses were raising captive wolves to introduce to Isle Royale, but this proved to be a major problem.

These imprinted wolves often approached people when they came to the island to camp, and all the wolves but Big Jim wound up being shot.

Big Jim wandered the island for several years after that, but it is unlikely that he contributed any genes to the current wolf population on the island.


Read Full Post »


And get over it!

When I first heard that this was an antiquated notion, I screamed “Bullshit!” at the top of my lungs.

But I was wrong.

And I accept that I was wrong. That’s always hard.

It was hurting my understanding of dog behavior, and now that I don’t believe it, I’m open to seeing dog behavior very differently.


In 2009, an article appeared in the Journal of Veterinary Behavior: Clinical Applications and Research. No, it’s not a journal that studies the behavior of your vet! It is a peer reviewed journal of veterinary behavioral medicine. That means that the people who get things published in it are experts who know more about dog behavior than I do.

The article was a review of research on dog and wolf behavior, and its clinical application to veterinary behavior practitioners. Essentially it says that the dominance theory is bunk. Absolute bunk.

It seems the concept was on its way to being overturned through the work of Erik Zimen, a Swedish ethologist who studied captive wolf packs in Germany. He was the first to notice that some of these captive wolves wanted out  of their enclosures, which would make sense. No one wants to stay in a cage with a bully.

The authors point out that wild wolves don’t have linear hierarchies.  On page 137, the authors have a diagram of a wild wolf pack hierarchy. It’s more complex than anything you’ve read in the popular literature. There are two hierarchies: one for dogs and one for bitches. And the arrows are going in many different directions. It’s not linear at all. It is fluid. It is only the male hierarchy that has any linear quality to it. Dominance displays are very rare in the wild, because the breeding pair are able to hold their family together without much aggression. In the wild, only larger packs– mainly the result of disruption, as in the Yellowstone populations– engage in frequent aggression towards packmates. Normal wild wolves are not typically aggressive with their family members.

But the authors very quickly stop talking about wolves for a very simple reason– and a very good one.

Dogs are not big game hunting wolves from Canada, Alaska, the Rockies, or Eurasia. They are a domestic animals. Domestic animals do not always have the same social structures as wild ones. This assumption runs right through most traditional training modules. Cesar Millan relies upon this assumption almost entirely. The authors clearly state this assumption:

“Because the domestic dog Canis lupus familiaris is descended from the wolf Canis lupus, it is often assumed that its capacity to form social relationships is similar to that of the wolf.”

And that’s where it is dead wrong.

The authors explore literature on free-roaming and feral dogs, as well as the social behavior of domestic dogs that have been neutered. Feral dogs do not form pair bonds at all, so the wolf pack structure doesn’t happen. In wolves, status determines whether you breed or not. In dogs, virtually everyone gets to breed.

“Overall, it appears that domestication has radically altered the social behavior of dogs, so that when theyhave the opportunity to interact and breed freely, although they do form exclusive kin-based groups, they do not readopt a wolf-pack social system within these groups.”

Free roaming, feral, and neutered dogs don’t form hierarchies that are anything like wolves. I don’t have an answer as to why. However, the basic unit of a wolf pack is a mated pair, and if you don’t have a mated pair, you can’t have a wolf pack. It may have something to do with the fact that domestication has made dogs so socially tolerant of each other that they don’t care who mates with whom. The only time dominance displays were common is when bitches were in season, which makes sense. This is a competitive mating system.

The authors eventually conclude that this whole dominance concept is next to useless in understanding dog behavior.  Wolf packs are more cohesive and less belligerent towards each other than that model suggests, and dogs don’t form packs based upon mated pairs. The conclusion, then, is to look at dogs a from a social learning perspective.

The authors discuss very briefly how social cognition is very different in wolves and dogs. Dogs can be thought of as a tabula rasa. Wolves are much more controlled by their natural history.  Dogs are much more influence by social and associative learning. If a behavior works, the dog will keep doing it.

Dogs also use what they have learned previously to inform their social decisions.

The authors explain how this works:

Let us imagine, for example, a neutered male Afghan hound (AH) and a neutered male Jack Russell terrier (JRT). Although the 2 dogs have not met before, each will use information learned previously in similar encounters in deriving their behavioral response to the situation. The AH, for example, may have previously encountered a small, white male dog that responded to it with aggression. Because of the similar cues in this encounter, its anxiety would increase, and it would try to identify any other cues predictive of potential aggression. The  JRT may have learned to be anxious about all large dogs that show a tense body posture, because it has learned that this posture predicts aggressive behavior.

Because of previous learning experiences in other situations, therefore, the risk of aggression occurring in this encounter is relatively high, whereas if the same 2 dogs had met without any previous negative experiences, the outcome of the interaction would more likely be a friendly one. Using this learning-based model, therefore, explains the complexities of social interaction with no need toinvoke the concept of ‘‘dominance,’’ either as a goal or as an element in an overall hierarchical structure.

This model is much better than the dominance theory.  It is more consistent with Morgan’s Canon. It is easier for people to understand than the dominance-based nonsense, and it provides a clearer remedy to the situation. You retrain your dog not to be aggressive. You don’t fight him as if you are the alpha wolf!


So what does this mean for dog trainers?

Well Gun Dog Magazine has the answer.  Gun Dog is hardly a PETA rag. It’s a hunting dog magazine that has articles on gun dog breeds and training methods, as well as how to use tools like e-collars.

However, Ed Bailey writes that we need to drop this nonsense:

Unfortunately the dog behavior wannabes who love the alpha concept either haven’t read the updated literature, haven’t grasped the concept of wolf social ordering, or accepted it and are still flitting around spouting 40-year-old misconceptions.

What need is what he calls the “leader-follower” model, in which we utilize the social learning aspect of dog behavior to train our dogs. I wouldn’t call it “leader-follower,” simply because that isn’t enough of a semantic change from the dominance model. I would call “teacher-student”  or “parent-offspring.” (Remember, I think it’s very wrong to consider dogs human children.)

My other complaint about this analysis is that it does do the “dog as wolf” analysis. It’s not wrong. Dogs are wolves, but dogs are different from wolves. If you want me to explain the nuances of that contradiction, I would have to write for two or three days. I’m not going down that path right now. If you read the blog regularly, you know that it’s much more nuanced than you normally get.

For a dog trainer, the most important focus is on being a good teacher for your dog:

When the dog does the desired task it gets paid by receiving some of the desired resources. The dog is working for a living, but because he is getting paid with something positive, he is willing to repeat the desired task. When the dog does something undesirable or refuses to do the desirable thing, the pay is withheld. Gradually the dog does only the correct thing because it pays off and the undesirable drops out because there is no pay-off.

In this situation, the dog corrects itself. A mild correction such as a “no” or “ah-ah” can speed things along and even a heavier correction when required can be beneficial. But the need for any correction gets less frequent as the dog learns that “If I do this I get something good and if I do something else it’s just not worth it.”

The research by David Mech showed wolves train their young pups using this technique.

Teachers in grade school train their charges this way. Wild animal trainers and psychology rat researchers give a pay-out for every small increment of correct steps toward forming a complex behavior and have termed this process “shaping.”

There is a lot of evidence that animals that have been taught how to learn using shaping techniques are far better at solving novel problems and will work harder at it than control animals that were not taught how to learn. Hunting dogs respond the same way.

Wolf scientists moved away from the alpha concept long ago. They have re-evaluated the dominance-submissiveness model in parental care of pups and are now seeing it as a leadership model. Dog people, taking an opposite tack, seem to be getting more enthralled with alpha dominance as the way to go.

Dog people should catch up to the wolf people and drop the whole alpha thing from the vocabulary. And, like the wolf researchers, dog people need to rethink the dominance-submission model for training and realize that it can be counterproductive, cause problems, and that it may be good for making automatons but not thinking dogs.

They need to be teaching a dog to think, to put two and two together to get more than four. Why? Because for real hunting situations, a dog has to learn to think outside the proverbial box.

So for normal working dogs, it might be more useful if we adopted the learning model.

I also highly suspect that learning can also be more effective in treating dogs with aggression issues, but I’ve not had experience with dogs that are aggressive.

But when has a dog learned problem solving from being strung up on a choke chain every time it goes out? Cesar doesn’t teach dogs to think. He teaches them to obey. And that’s a major problem if his methods are applied to working dogs.


I need to clarify some things about wolf social structure.

As I stated before, the basic unit of a wolf pack is a mated pair, their pups, and grown offspring that are generally under the age of 2 or 3 depending upon the location. Some wolves remain with their natal packs until they are as old as 5 years of age. Others leave when they aren’t even a year old. All wolves eventually leave their natal packs. This includes both dogs and bitches.

Canis lupus is not the only species of dog to form packs. Ethiopian wolves, Lycaon pictus, bush dogs, and dholes form packs. Ethiopian wolves and Lycaon typically have more males in a pack than females. In fact, it is not unusual for these animals to have only a single breeding female in the pack.

Eastern coyotes also form packs, but this appears to be a relatively recent adaptation to living in deciduous forests that are full of deer. Hybridization with wolves also may have made them more likely to form packs.

It is likely that wolves and the others evolved packs from a phenomenon seen in coyotes and the three species of jackal. In those animals, it is not unusual for grown offspring to remain with their parents to help raise the next generation or two. In jackals and coyotes, these helpers also have been seen helping their parents hunt larger prey.

In the pack hunting dogs, it is likely that they used this helpers helpers to eventual establish a niche as specialists at hunting large prey. However, Ethiopian wolves do form relatively large packs, but their diet is almost exclusively rodents. So exactly why these animals formed packs in the first place is still a pretty good question. It’s much more likely that assistance in raising the young was a stronger motivator than hunting. It is also possible that finite nature of available vacant territory forces young adults in these species to want to remain with their parents. It is probably a combination of all of these factors.

All wild dogs that have been studied have a mated pair as the main social unit. Dogs are unusual among the Order Carnivora in that monogamy (or serial monogamy) is the rule. Pair bonds are very important. The only exceptions that have been found are some populations of red foxes in which harem-type systems have been observed.  No one knows what sort of social unit exists among the small-eared dog. They are so hard observe that we know very little about them.

The original studies on wolves focused on their aggression. Aggression was the thing for ethologists to study in the 1940’s.

Now the emphasis is one what makes wolf packs more cohesive, and it has been discovered that what makes them more cohesive is not dominance or aggression. It is the strong family bonds that exist between members of a pack. A pack is a family.

Nothing more. Nothing less.


It’s time to put the folk ethology away.

And open up our minds.

The dominance concept has damaged too many relationships between people and dogs. It may have hindered our ability to train good working dogs.

It’s a dinosaur.

Its time has come and gone.

Read Full Post »


Now I am not saying dogs are the same as the big game hunting wolves that Mech has studied.

However, I don’t think you can base all of dog behavior off of this dominance literature, which actually comes from the work of Rudolf Schenkel. Schenkel studied captive packs of wolves in Switzerland that, as Dr. Mech pointed out, were unrelated assemblages of individuals.

Schenkel was operating in the culture of ethology within the German-speaking world, which, of course, was heavily influenced by people like Konrad Lorenz (a National Socialist scientist) and other Germans who saw the wolf as the ideal canine. It was wild and hard, just like the society that Germans dreamed of creating with people. The dog and the golden jackal were seen as the degenerates. The golden jackal was almost like the “Jewish” wild dog, while the domestic dog was the creation of too much coddling from the welfare state and other excesses of liberalism.

Although most of these ideas were discredited after the Second World War, these studies and this culture has tinctured our understanding of what dogs and wolves are.

We incorrectly assume  that all wolves are like wild ones that live in northern tier and Canada and in northern Europe. What about the ones that live in India or the Middle East? What about  the ones that live in Ethiopia? (Yes, those Ethiopian ones are different species, but they aren’t that different from the others.)

We also incorrectly assume that wolves have always behaved the way they do. I think this is a fundamental error.

It takes real work to socialize a  modern wolf to a human, and it requires specialized formulas and taking the pups away from their mothers before they are 19 days. I don’t think that hunter-gathers were able to do those sorts of things.

That means that the wolves they domesticated had different tendencies than modern wolves do. Perhaps they had longer critical periods for socialization. Perhaps they were like virtually all other wild dogs that have never been hunted by man– extremely curious about us. If those were both true, then domesticating wolves would have been much easier. So easy that a cave man could do it.

I don’t think you’re getting very much from assuming that wolves and dogs have the same social structure, when it’s obvious that wolves don’t all have the same social structure.

And it also ignores that domestication has allowed dogs certain cognitive shortcuts that allow us to teach them rules and to give them more freedom. The average dog is capable of learning and internalizing rules that wolves would never be able to learn.

One thing I’ve always read was that socialized coyotes and wolves are nearly impossible to housebreak. They just don’t get it. And this sort of throws out the theory that dogs are den animals and won’t soil in the house because they think it’s their den.

What really happens is the dogs learn the rules from their humans where they are supposed to relieve themselves.

And the reason why dogs can do this and wolves can is that dogs learn rules. They seek social approval from people. Wolves might bond to us, but they don’t have the ability to bond with us in exactly the same fashion.

Dogs also need freedom to be dogs. This is always left out of dog behavior discussions, but dogs need to run off leash. They need to meet other dogs. They need smell them, lick them, and hump them (even if it is another dog of the same sex! Dogs do not have much use for our anthropomorphism or our puritanical sensitivities.) They need to self-anoint with disgusting substances and piss on things. They need to wallow in mud puddles and eat deer feces.

They need these things because they are dogs. What would you be like it if you never got to read a good book, listen to good music, or watch a good program on television? What if you never traveled, except to go to the doctor to get a shot or your reproductive organs removed?

I think you’d be pretty barmy.

The real cause of behavioral pathologies in dogs is not dominance. It’s an over-used term. It is that dogs are not allowed to act out parts of their nature. They are kept in suburban yards or in apartments 8, 9, 10, 12 or more hours a day. In most suburban yards, all dogs have to do is listen to the neighbors’ dogs bark at things and then eventually join right in.

The real problems that dogs are facing here is that their natures are forever being bent to suit our needs, and now their fundamental natures are reaching the breaking point.

Where I grew up, dogs were allowed to roam freely. These dogs never bit anyone. They were never in severe fights. They never tried to use aggression to control people because they had other things to think about. If you’re locked in a house and yard all day with nothing to do, at least some of you would develop some weird social pathologies, like trying to take over the house.

The real discussion about the best way to keep our dogs is trying to find some balance. Yes, give the dogs rules, but those rules should exist only to give the dog more safety and more freedom over his or her own life. And communities need to find ways of providing good off-leash spaces for dogs.

Dogs also have to be allowed to socialize as puppies. They need to grow up experiencing things. They do need some basic training here, but the really important thing is for puppies to see things like buildings, cars, strangers, horses, and all sorts of other things.

Just as wolves are far more complex than you could imagine, the relationship between man and dog is something far too complicated to get caught up in discussions of hierarchy. But just as not all human relations are based upon dominance and submission, the relations between dogs and between dogs and people cannot be so horribly reduced.

I am glad that the scholarship on wolves and dogs that was so tainted by Nazi and early twentieth century German science is now being replaced with better scholarship. Researchers at the Max Planck Institute in Lepzig and  Eötvös Loránd University in Budapest have finally begun to look at the domestic dog very differently. Their studies are shining light on the remarkable abilities that domestic dogs have evolved to live with these mercurial naked apes with nuclear weapons and God complexes.

And if we’re going to have a discussion about dog behavior and how to best keep them, that’s where we should be looking.

I personally don’t want to live with a pack of big game hunting wolves. I want to live with domestic dogs.

Read Full Post »

From the Anchorage Daily News.

Romeo was a black wolf that lived near Juneau who loved to play with dogs. (I have written about him here.)

Well, he hasn’t been seen since September, and because he is estimated to be 7 0r 8 years old, he may have died of old age.

Or a neighboring wolf pack has killed him.

Or a bear.

Or maybe it was the former governor.

Or maybe he found a bitch wolf, and they were able to form a pack.


Well, here’s some footage of Romeo from last winter.  He’s hanging out with his Labrador friends.



How Romeo came to be so friendly with the local dogs is the subject of several stories. One of which goes that Romeo’s mate was killed by a cab, and Romeo was left on his own. Not seeing other wolves with which he could form a relationship, Romeo began to see out dogs for companionship.

BTW, it has been rumored that Romeo has killed small dogs, and this photo is often offered up as evidence. That’s a pug he’s carrying in his mouth. The truth is the pug was not injured at all. Romeo was just playing. Apparently, someone took their pugs out to the Romeo’s area in hopes of enticing him out for a photo session. Romeo grabbed one pug and ran off with it, but he ran off for only about ten feet before dropping the pug.

Read Full Post »

From the Telegraph.

Apparently, this young wolf found himself without a pack, so he hooked up with the local dogs.

Events like this had to be much more common in the past. After all, wolves and dogs have been living in the same areas for thousands of years.

There had to been a gene flow between the populations, especially when wolves had little reason to fear people. It was only when we began to keep livestock that the wolf began to be persecuted.

Romeo is a good example of why we can’t make broad assumptions about dogs or wolves. In general, wolves are rather intolerant of domestic dogs, but sometimes wolves and dogs become friends.

Because Romeo needed companionship at that early age, he developed an entirely different response toward domestic dogs than one might expect from a wolf. Perhaps this suggests that the hyper-territoriality that we see in wild wolf populations is at least partially “culturally” driven.

Read Full Post »

From Apexpredator11

Believe it or not, an actual study found that shows that certain populations of wolves in British Columbia will fish for salmon before they go hunting for deer. Why?

Salmon is nutritious.  It is a good source of omega 3 fatty acids, and it is also a lot safer for a wolve to catch fish than tangle with an ungulate with sharp hooves and possibly sharp antlers.

Of course, wolves cannot live on fish alone, for salmon are a seasonal delicacy. But when the salmon are running the deer in this part of the world don’t have to worry about the wolves.

It is also possible to train a retriever to help catch fish. Two ancestors of the retrievers were known for their utility in commercial fishing. The Tweed water dog and the St. John’s water dog were both used in fishing enterprises. The St. John’s breed was known for hauling nets from the sea and bringing in fish hooked on longlines. The Tweed water dog was often seen helping fisherman net salmon near the town of Berwick-upon-Tweed. All one has to do is teach the dog to stay at a heel while both dog and human wade into a trout stream. The dog remains at a heel until the a fish is hooked, played with, and then nearly hauled in. The dog then catches the worn out fish.

Here’s “Ben of Hyde”  (the first yellow Labrador) doing this task in a British trout stream:


Read Full Post »

This is Dr. L. David Mech (“Meech”), and he knows more about wolves than ANYONE.


If you want to know more about him, well, you can check out his website and this page on his work. I’ve learned more about wolf behavior from his work than anyone else.

Read Full Post »

%d bloggers like this: