Guard Dog Blog

on Livestock Guardian Dogs and small farm life…



This is a long post, out of necessity. I need to be as clear and comprehensive as possible, so that you not only understand the reasoning behind the corrections but also how and when to implement them. Please read it as many times as necessary to feel confident about these things and feel free to comment or message me with any questions, comments or stories on this topic. 

One of the most important things to learn to do when raising and training LGDs of all ages is the art of binary feedback. Binary in this case refers to two distinct parts of information or input: positive and negative. For those who read the previous post about Skinner’s four quadrants of operant conditioning, and who want to put the binary feedback under them – what I’m referring mainly to is positive reinforcement and positive punishment. Remember, if you will, that positive in this case doesn’t mean emotionally positive, instead it means “adding”. Positive reinforcement means to add reinforcement (or reward) and positive punishment means to add punishment (or correction).

Before we start into the hows and whens of corrections, let me touch on the opposite side of the binary feedback system – that of reinforcement. The word reinforcement is often used interchangeably with rewards, which is both helpful and unhelpful. For many people, the word reward brings to mind treats or “dog cookies” and not much else. For some people it also makes them think of dogs who are demanding or out of control, unable to behave unless the owner gives them a treat. I won’t go into too much detail today on the subject of reinforcement/rewards, but I will say that there are many kinds of reinforcement/rewards available to us to use when communicating with our dogs. Essentially, anything that a dog finds satisfying, desires or otherwise will work to attain is inherently rewarding/reinforcing for them. This can be anything from food to affection to freedom.  Watch your dog and see what they seek out apart from basic needs (we don’t use shelter, water or access to minimal amounts of food as reinforcement). These are their reinforcers/rewards. Observe your pup when you bring them a small scrap of meat and toss it to them – do they perk up and pounce on it? That’s a good indicator that it’s what we call a high value item (something they really, really want). Do they seek you out for a pet or to put their paw on you (Great Pyrs, especially, are known for this)? This indicates that they enjoy affection and close connection. LGDs quite commonly find praise from their owners to be very rewarding as well – does yours turn towards you and look interested when you speak to them kindly? Listen to your dog.


As a general rule, I am a strong proponent of the 80/20 feedback system. This rule was developed for human parenting purposes and states that 80% of the input you give your children should be positive in nature, while the negative input should be restricted to 20%. It has proven itself to be very helpful and as I’ve found, equally as beneficial for raising and keeping dogs. It keeps the balance of the relationship solidly on the positive side while still being effective at stopping inappropriate behavior. This means that the bulk of the communication from human to dog should be positive, while the remainder can be negative if necessary. Personally, I work to get the 80% of positive interaction/input higher and higher over time, and it is good to note that this should happen naturally as the dog matures. It’s true that there will be times when the balance seems really off, times when dogs are testing the boundaries or just generally acting up – adolescence can be one of these times. Sometimes, during teaching phases, the negative input will feel very strong and difficult to do. As long as the dog has a fairly clear understanding of what should be done instead or guidance is given in the moment, this feeling needs to be overcome. We often make more of a mess by anthropomorphizing (placing human feels and motivations on animals) and over compensating for dogs than we do by being clear and firm with the boundaries. Allowing a small pup to behave in a way that we wouldn’t want an adolescent to, for instance, by jumping on people or chasing/mouthing stock leads to confusion on the part of the dog as they grow and the rules suddenly change. The amount of force/effort needed to correct a an older dog is typically much more than is needed with a pup. This is also why LGD pups are wired to learn well from early experiences. No one has time or energy to spend getting an adolescent in line and teaching them about being a working dog from square one. It’s frustrating, ripe for side effects and requires nerves of steel – not for the amateur.

The sooner the lessons are learned, the better. That said, lessons must be age appropriate. Tailor your expectations to the life stage of the dog; quite simply, an 8 wk old pup cannot be expected to never make a mistake. They just don’t know any better and have a low capacity for self control. Notice that I don’t say that they have NO self control, just that it’s less in both amount and length. This is also why it’s vitally important that pups grow up with good, fair mothers and in litters that help them learn the basics of impulse control. The breeder in the developed world plays a large part in the development of self control as well, especially if the pups do not have access to real world experience. Objectively assessing your pup(s) after they arrive at your home (or before!) at 8+ wks is very important – how much of this training have they received? Are they displaying an ability to handle frustration or to control their behavior when required? Do you have older dogs who can help them learn proper social language and how to grow self control? I have actually used my teacher house dogs in a controlled way to teach pups I’ve gotten in who lack age-appropriate behavior control. I would venture to say that if this is not accomplished early on, pain and frustration inevitably follow. It is one of the non-negotiables that all pups MUST learn. Learning self control is not a panacea that will ensure all LGD pups are successful, but it certainly weights the odds in that direction. Proper self control and social behavior (not playing with livestock, deferring to older LGDs through submissive behavior, soliciting peer play away from the stock or being careful around them, lowering body language when interacting with stock, not fighting picking fights over food/resources, refraining from endlessly seeking out the stimulation of play or physical connection, learning to “turn off” and relax) are vital skills for LGDs. Waiting until they mature to insist on these behaviors in any consistent way is not only counter intuitive and a complete waste of time, but also leads to selecting for dogs who cannot function as LGDs until they are grown.

The other important thing I adhere to when providing feedback of any kind is to ensure I make my feedback very distinct and clear. I have a rather calm baseline that I’ve developed extensively over time living with dogs. Most people I know who live with a lot of dogs find that this is the best way to ensure they don’t accidentally feed excited energy into their dogs’ behavior. It also has the added benefit of lending itself well to provide clear lines of voice control and body language for feedback/input. My dogs know well when I’m happy with them, my voice becomes a bit higher and more soothing in nature. My body language is inviting, open and positive – I smile so that the smile comes through in my voice. I offer my hand to them to smell, I pat their head or stroke their side. I will often hold their heads, look into their eyes and tell them just how wonderful I think they are (please do not do this with a dog you do not know well or one that has proven themselves to be aggressive with humans – they can take direct eye contact as a threat). This positive input is not the same as being excited. That is much more animated, both in voice and body language and reserved for when I’m inviting them to play or I want to motivate them – something that is typically not done around the stock or in the case of my house dogs, in the house.

This clear distinction in voice pitch and body language means that the first steps in conditioning correction are easy to achieve. LGDs especially understand the change in tone and intent of tense body language. Their social language with each other, as is the case with many working dog types, is very overt and clear-cut. A content LGD will have a low, sweeping tail, and a relaxed body that is balanced equally over their four legs. Their faces will also be relaxed, often with a hint of a wrinkled smile in the corners of their mouths and eyes. Their ears will be neither pricked forward or held back, instead, they will be floppy and relaxed. They will move neither backward or forward intently when approached. When they lie down, they will often “flop” down and sigh deeply.

An excited LGD may be so for various reasons, but their body language is fairly consistent. Their ears will be pushed forward, their eyebrows lifted up. If they are happy and excited, they will often dip into a “play bow”, their bums up in the air and their fronts in a lying down position. Their bodies will be bouncy, as will their tails, even going into a helicopter blade motion – around and around. They may vocalize with a higher pitch or whine. If they are excited due to the presence of a predator or because they are anticipating something, the balance of their weight will shift to being over their front legs. This makes them look like they are leaning forward and could leap into motion in a split second. It also serves to make them look bigger and more intimidating, which means that they use this pose when correcting each other or as a way to signal intent for conflict. In this case, their bodies will stiffen and their eyes will widen significantly. The sides of their mouths will pull back, but in a different way than if they are happy. The corners of their mouths and eyes will be very tight and hard. Depending on the severity of the intent, a growl may precede or follow these body changes. The growl may be slight and come from higher up in the throat with a significant pause to see if the offender will change their behavior or it may start low, come from the chest and rise in tone and strength. If physical conflict follows, it usually is swift and fierce; it is not uncommon for people who are not aware of the other signs to claim that the physical attack came “out of the blue”. Generally, the attacks are very loud and full bodied – slamming into each other, working to push each other over or going over the back of the other dog so they can gain the advantage of having higher purchase. It is not often that any real damage is done before the conflict is resolved, commonly it is confined to a maximum of tooth punctures. Ears and throats are favorite targets, as is the back of the neck. If a LGD goes for the throat or belly, they are very serious about their attack and fully intend to cause a lot of harm or even to kill their opponent. This is not a routine course of behavior for a LGD, but can appear when mature males or females fight for rank and territory. A dog who pretends that all is well, sidles up beside another dog (or much less commonly, a human) and then attacks with lightening speed is a dog who is not displaying any appropriate social behavior whatsoever – we call this predatory aggression, something commonly selected for with dogs bred to fight.

Living socially within a group of dogs (as is common with LGDs kept for larger operations, in breeding operations and in their countries of origin) and with prey animals requires a high level of social competency. Two sets of interactive behavior are required – one with their charges and one with their fellow working dogs. There is a lot of overlap, however and the one consistent factor is the ability to read tone and body language appropriately. We can harness this ability to read tone and body language and to learn from single events when correcting them effectively. We are not trying to replicate LGD behavior when we do so (as some may claim), but we are doing our best to speak their language in order to communicate clearly with them. LGDs typically start out their corrections with plenty of warning, which means that the one being corrected has a lot of time to change their behavior before things escalate. This is especially true with very young pups, who are given what we call a “puppy pass” by older LGDs for much of their overly enthusiastic behavior. They will be corrected for going too far, though.

The correction will typically start low and slow and “ratchet” up if the pup doesn’t listen (this is the warning system). Depending on the dog, the start could be very quiet, almost under their breath.  Most pups only need things to escalate once or twice before they learn that the early warning signs will be followed by physical correction. Typically this portion culminates in a loud roar alongside a leap over top of the pup, standing over them. Often times this is combined with the older dog placing their mouth directly and widely over the muzzle of the pup and holding it to the ground. The pup is meant to hold still and “give” to their elder. Quite often, the older LGD will not remove themselves from the pup or allow them to get up until they are satisfied that enough submissive behavior has been displayed and that the message is solidly received. Unlike the mythical “alpha rolling” technique, the correction is wholly controlled by the one being corrected. They hold themselves to the ground (no LGD lies on another to keep them down) and either offer appropriate apologetic behavior or don’t. If they don’t, as is sometimes the case when the dog being corrected is an adolescent or young adult, the older dog will repeat their attack until they do.  It is important to know that as young pups mature, the tolerance portion or “puppy pass” for their silly or excited behavior disappears. If it doesn’t, it is quite likely that the older LGD isn’t up for the job of sole disciplinarian or needs some guidance.

It is also very important to understand that LGDs have a system of fairness for corrections. Small infractions result in smaller corrections or a longer warning phase. Large, serious infractions result in no or a very short warning phase. The dogs inherently understand this from the time they are little, which is why it does not destroy relationships between dogs or even with humans when it’s applied. Random, very harsh corrections as some people are fond of doing with their dogs (often also inappropriate for life stages) result in a confused dog who loses trust in their leader. This is what people talk about most when they talk about corrections going awry. Using corrections/negative feedback as the bulk of the interactions with your LGD (flipping the 80/20 rule around, for instance) will also result in a frustrated dog who learns that humans are difficult to please. The working partnership requires that we teach with benevolence – and correcting appropriately is a part of that kindness.

When we humans correct, we must have in our minds what the “minor” infractions are, what the “major” ones are and what life stage our dogs are in. We must construct the environment for as much success as possible (don’t put dogs in no win situations with stock), feed our dogs well and care well for their health. We must ensure they know where we want them to be (don’t have the dog on your bed one night and in with the stock the next, don’t move them randomly from one stock group to another) and what their jobs are. If you want them in with stock when mature, have them in with stock from a young age. This doesn’t mean they can’t be kept separate at times, but ensure that they understand why it is or at least keep it short and work it out to integrate them back in a reasonable period of time. The corrections should start in the way that an LGD elder would do, low and slow for minor infractions and high and swift for large ones. The amount of force and effort required will be determined by the dog themselves – some give easily to a verbal correction (a sharp, low tone with an angry inflection) such as “NO” or “HEY” or “UH” while others require a higher level just to initially acknowledge you. The easiest way to condition this tone is to ensure that you always use the same one, in the same tone (vary in intensity for intent) and follow with body language. Shift your body forward in the upper torso (lean forward) and step into the pup while saying it or immediately after. Raise your eyebrows and look directly at them. If the pup doesn’t stop what they are doing or give ground to you, narrow your eyes and tense your mouth/facial expression. Increase the volume of your chosen correction word/sound and make it deeper. Step towards them. If this doesn’t result in the desired behavior, stride toward the pup and physically stop them/remove them. Express your disapproval of their choice not to listen by shaking your head at them and muttering disapproving words under your breath. If they immediately return to the behavior, remove them without a word and either place them in a pen, on a chain or outside the environment for a short period. Remember that a short time can feel like a very long time to a small pup or adolescent.

Alternately, if it’s an inanimate object they are being inappropriate with, remove it from them. Put it where they can’t access it, or block access to it with your body. Cross your arms, plant your feet solidly and lean forward with the aforementioned facial expression and tone if the pup doesn’t sit back or leave. This is helpful for situations such as where pups are beginning to refuse stock access to water or food. If they try to go around you, move in front of them. If they keep trying, drive them off with large arm motions and by increasing your tone to a yell. In a pinch, and when you can’t get to a dog in time to stop a behavior, throw something beside them that makes a noise or bang on the bottom of a pail to get their attention. Do not hit the dog with the object or with anything else, including your hand.

When the pup shows that they are no longer trying to repeat the behavior, when they sit/lie back on the ground, or when they show other submissive behavior such as curving their bodies around to you in a semi circle, lift one front paw toward you cautiously, lick their lips, look toward the ground – relent. Relax your body language. Soften your facial expression and raise the tone of your voice back to the calm middle ground. If you had a hard time getting through to them about the behavior or you were unable to, you can choose not to speak to them or acknowledge them at all for a while. This sort of social isolation is very effective with LGDs. If you’ve cultivated a bond with them based on companionship and positive reinforcement, they will take your refusal to acknowledge them very seriously.  Forgive them by using your friendly body language and tone – very quickly for young pups and within a reasonable amount of time for older dogs. Don’t be surprise that if, when you choose to interact with them again, they respond with great enthusiasm. Falling out of favor is a very strong correction for many LGDs – and often is enough to stop them from repeating the behavior again.

Once a pup knows that you will follow through from the starting point of tone to stepping into them, removing them or blocking them – and especially with removing your affection, they will begin to respond quickly and well to the tone of your voice and your verbal correction word/sound. This may require a few times of following through and refreshers as they go through adolescence, but it will make your life much easier overall. When you can stop behavior from a pup or dog with a word from the house or across the yard, it lowers your work load and keeps them more reliable. If you are in closer proximity to them, start by speaking as softly and coolly as possible (don’t make it personal), allowing your facial expressions and body language to do the work of convincing them of your seriousness. Eventually this will result in being able to just speak your correction word/sound in a soft tone without any of the rest.

Some infractions and just some dogs in general, require harsher corrections. This is partly because the nature of the behavior is so dangerous (or could become extremely dangerous over time), but also partly because the harshness of the correction tells the dog that this behavior will never be tolerated. There is never a time, for example, that chasing stock is acceptable. Some people will tell you that any “inappropriate” glance at stock should be corrected in this way, but I strongly disagree. Others will tell you that every behavior should be treated equally, with the same rote low level corrections, but I equally disagree with that. Both approaches have a high likelihood of failure. How do you know if your corrections are too harsh or are ineffective? That is easily sorted out. Is the behavior disappearing? Then you’re being effective. Does your dog seem confused or frustrated or are they starting to avoid you entirely or suck up to you obsessively? Then you are being too unfair and too harsh. The dog will always tell you.

Examples of harsh, higher level corrections are scruffing (grabbing the skin on either side of the dog’s face and getting very loud and angry in their face), yanking swiftly and strongly on a prong or slip collar (I very much prefer the prong for this if it’s used), using stim or vibrate on an e collar (use a reliable, very multi-level collar and know what level your dog finds aversive – LGDs typically hate vibrate more than stim as vibrate can actually send them into a panic-  they will respond well to very low levels of stim), pushing them over onto their side and refusing to allow them up (do not lie on them, that can also cause them to panic and it is not well understood by dogs) until they “give”, isolating them by placing them in a pen or on a line alone for long periods, and getting very loud and angry with them, driving them back over a large area or up against a vertical object.

I cannot emphasize the following enough:


The only exception to the second warning, that of dogs who have displayed aggression toward humans, is to do with young pups. Serious corrections for these pups can make the difference between life and death, provided they are done fairly, swiftly, and they are able to make amends afterwards. This is not an exaggeration.


This is also why penning or isolating for long periods of time doesn’t work as some people claim it does.

Finally, for minor infractions, remember that pups and dogs who are learning will make mistakes. Pups will make more of them, simply because their memories can be short, their self control easily depleted and their desires very strong. They are the mental equivalent of small children – and we would not expect them to do high level reasoning or sit through a board meeting without something to amuse them. Luckily for us, pups mature much more quickly than human children do, so the requirements can keep pace on an accelerated schedule. Pups over 4 months of age should be able to be in with stock (the exception may be with chickens) for long periods without major issues. They may regress at certain times as they mature (9 months and 18 months are particular hotspots), but reminders should get them back on track.

Keeping this all in mind, start from low on the scale and ratchet up over again when a young pup displays a minor infraction, even if you just corrected for that behavior this morning and yesterday. If it was just a few moments ago, consider being more effective and using management more – separate and supervise for short periods perhaps. As they get older, start a little higher on the scale and consider a harsher correction if the behavior hasn’t fallen out of their repertoire. Again, ensure that they have their basic needs met – if that doesn’t happen, it will increase the likelihood of inappropriate behavior.

Sometimes, high levels of correction will require “set-ups”. By that, I mean, setting up interactions and the environment so that the behavior will appear. Correcting a pup for chasing stock will require them to be in close proximity to that stock. Consider doing this for behaviors that you don’t see in routine life, but you see evidence of (dead poultry, frazzled stock, pulled wool, etc). It is also helpful when training to “hot” or electric wire fences – ensuring the dog gets a good zap by tying food to the wire is an example of this. The best way to handle these things is directly head on. Don’t let time pass without addressing it, especially if you see continual evidence of the behavior.


These are lists of minor and major infractions that in my opinion require correction. These lists are not by any means exhaustive, but they are some of the most common inappropriate behaviors.


  • jumping up to greet people/jumping on people
  • being pushy at feeding time
  • demanding constant attention
  • beginning to guard the stock’s food, shelter or water from them
  • soliciting play from stock
  • licking stock intently
  • staring hard at stock
  • pushing into stock
  • being rowdy around stock
  • picking a fight with another pup around stock
  • refusing to wait at gates after being taught how to
  • pushing in at the home door
  • refusing to wait calmly when tied for short times
  • breaking out of a pen
  • running away (should not be corrected at all if recall has not been taught)


  • chasing stock (of any kind)
  • mouthing (placing their mouths on) the stock
  • biting the stock
  • mounting the stock
  • not allowing an individual or small group of stock freedom of movement
  • escaping the pasture fence to roam or muck around (my Ivy has escaped the pasture to come to the house and tell me of an injured or sick animal)
  • knocking a person or stock over
  • being enthusiastic around stock that leads to injury or the potential for injury
  • obsessively harassing elderly or weaker LGDs
  • “stealing” babies from mothers aka “over-mothering” (do not put immature dogs in with birthing mothers)
  • attacking stock who show curiosity over their food items
  • attacking other dogs for their food items
  • responding to appropriate corrections from stock by becoming angry and coming back at them


To correct, or not to correct, that is the question.

I’ve put off making this post for a multitude of reasons, not the least of which is the fact that our current political climate does not support the judicious use of positive punishment, more commonly referred to in the dog world as “corrections”. My third and final puppy raising post relies on the clear understanding of this topic, so it’s come time to stop putting it off and jump in. I’ve done this over and over privately for people who seek out my advice, so hopefully this will be as clear and helpful for you as it has been for them.

As someone who cut her teeth in the dog world on training that consisted mainly of positive punishment (adding punishment, aka adding something that is is aversive or painful), and then moved to positive reinforcement (adding reinforcement, aka adding something that is rewarding or pleasurable) only training, finally ending up somewhat in the middle or what we call “balanced” training, I feel that I have a fairly even amount of experience to draw from. I also bring a certain amount of objectivity to the conversation, as I haven’t been married to one form of training over another. I see the undeniable benefits of using reward based training. I also see how stopping behavior through the use of punishment is critical as well. As an aside, I also tend to believe that making an animal do something for either our comfort or amusement only is a fairly unethical approach to living with our beautiful friends.

I won’t talk here about how positive punishment relates to the training of the average pet dog, as that is beyond the scope of our focus on LGDs. That said, I do believe that there is a larger picture that ought to be addressed, specifically in terms of how the obsession with rewarding our human children has been taken up wholeheartedly by dog people, much to the detriment of our furry companions. This is a conversation that I show up to have in person or online, time and again – that’s how strongly I feel about it.

The main reason I’ve wanted to write on this topic relates specifically to training LGDs, simply because I believe that these dogs are one of the types suffering the most due to the evolution of the positive only movement. This current trend started with good intentions, as a badly needed push back against the heavy handed training of yesteryear. It was, quite frankly, at the time quite radical to say that dogs could be trained without the use of “alpha” dominance – both radical and timely. Dogs, including our LGDs, continue to benefit from the work of the brave men and women who stepped out of their comfort zones to ask the hard questions. What does a dog feel? How does a dog perceive the world? What do they need? How can we change what we do to meet those needs better? WHY do we have to do things as they have always been done?

Like so many reactionary movements, though, certain people in it were not satisfied with the general acceptance that training needed a kinder approach. They wanted it all to be what they perceive as the best form of communication with other species: reward them. The works of B.F. Skinner, (a psychologist who spent time researching the effects of different types of input on animals) became the basis for what these people call “science based training”. B.F. Skinner is most famous for developing the four quadrants of operant conditioning, a structure that developed from his work attempting to effect behavioral change with animals kept in a perpetual state of neediness. These animals were then placed in an isolation chamber, known as a “Skinner Box” and subjected to experiments that measured their responsiveness to positive and negative stimuli. It is with the information that Dr. Skinner recorded from observing the response of these animals that he formed the diagram that we know of as the four quadrants of operant behavior or “Skinner’s Quadrants”.


The four quadrants of operant conditioning

In 1938, Skinner’s first book, “The Behavior of Organisms” was published. This book furthered the idea of behaviorism: a concept built in the early 1900’s by a Dr. Watson onto Pavlov’s classical conditioning (remember the experiments showing that a dog would salivate at the sound of a bell if that bell had always preceded the appearance of food?). Behaviorism is explained by Psychology Today as:

“Behaviorism seeks to identify observable, measurable laws that could explain all of human behavior. Although psychology now pays more attention to the inner landscape of emotions and thought, behaviorism has had a durable influence on everything from animal training to parenting techniques to the bonuses financial managers receive.”

In other words, the strict approach of behaviorism views people and animals as “input/output” machines; apply one of the quadrants to them and get the desired behavior result. It also relies heavily on the notion that if the desired behavioral result is not achieved, it is the fault of the person applying the quadrant. They are not doing it appropriately: fast enough, well enough, effectively enough or creatively enough. There is a lot of truth to the fact that animals (ourselves included) respond to input in predictable ways. You only have to offer to give a pig a tasty treat or observe a child’s behavior after they’ve been injured to know that certain input = certain output for many things. Overall, however, life is not that simple.

Much of Skinner’s work involved the observation of animals under sterile laboratory conditions, where the animal had no competing stimuli to deal with. These animals were kept undernourished, in a state of perpetual need. It was the combination of these two factors that led directly to the strength and consistency of the behaviors that Skinner was able to observe. It wasn’t until a couple, both psychologists who were students of Dr. Skinner’s, took his methods out into the real world in the form of training performing animals that more of the picture emerged. Their paper, “The Misbehavior of Organisms” was published in American Psychologist in 1961. In it, they spoke about the limits to the theory of behaviorism that they were able to see even in their more controlled training environments:

“When we began this work, it was our aim to see if the science would work beyond the laboratory, to determine if animal psychology could stand on its own feet as an engineering discipline. These aims have been realized. We have controlled a wide range of animal behavior and have made use of the great popular appeal of animals to make it an economically feasible project. Conditioned behavior has been exhibited at various municipal zoos and museums of natural history and has been used for department store displays, for fair and trade convention exhibits, for entertainment at tourist attractions, on television shows, and in the production of television commercials. Thirty-eight species, totaling over 6,000 individual animals, have been conditioned, and we have dared to tackle such unlikely subjects as reindeer, cockatoos, raccoons, porpoises, and whales.

Emboldened by this consistent reinforcement, we have ventured further and further from the security of the Skinner box. However, in this cavalier extrapolation, we have run afoul of a persistent pattern of discomforting failures. These failures, although disconcertingly frequent and seemingly diverse, fall into a very interesting pattern. They all represent breakdowns of conditioned operant behavior.”

marian breland walt disney

Marian Breland with Walt Disney, 1965

The Brelands informed the world in 1961 in no uncertain terms that operant behavior could not explain all that animals do, nor could it control or change it. Animals are complex, far from the input/output machines behaviorism would have us believe they are.

Why does this matter? Why am I telling you all this? Well, it’s quite simple really. The people who are the rabid proponents of positive reinforcement training to the exclusion of positive punishment build their arguments for this type of training on Skinner’s work, which proved that behavior strengthens best through the use of rewards. In other words, the Skinner Box experiments proved that under those circumstances, animals learned to do behaviors and refrain from doing certain behaviors when people gave and took away rewards.  For dog people who think this translates the same way to real life outside of the laboratory, who have ignored the Breland’s sobering conclusions, this means that they stubbornly stick to demanding that everyone give or take rewards away as the only humane methods needed to train dogs of every type, under every circumstance.

These are the people who are most active on social media dog groups. These are the people who are lobbying to change our animal welfare rules. These are the people who are trying to get tools like prong collars and e collars banned, even though more injuries happen due to flat collars, harnesses and head halters than with the “bad” tools. These are the people who label anyone who uses such tools or physically corrects their dogs as “abusive”. You’ll find them being very outspoken about the possibilities of psychological harm from the use of adding punishment to training regimes, a very real possibility, but not in the way these people would have you believe when it comes to LGDs. These are the people who are taking over the LGD training conversations online and in written print.

And it’s costing LGDs, their owners and the working dog relationship that should be evolving a lot. In many cases, it’s causing the complete breakdown of the relationship and the dog to behave in ways that would not happen were corrections better understood and more appropriately applied. It is my firm opinion, and that of many people who have owned these dogs for generations that corrections are an integral part of training LGDs successfully, especially pups.

The reason you’re not being told to apply corrections when training LGDs or if you are, the instructions are vague or largely unhelpful (extended “time outs”, tackling dogs to the ground, isolation, etc.) is because there are very few people in LGDs who both understand the benefits of using positive reinforcement and have the balls to talk about corrections appropriately. I’m no shrinking violet when it comes to dog world politics, and even I struggle with how much to say on social media or in print for fear that I will be labeled an “abuser” or harassed endlessly. The other reason is because some people believe that dog owners are generally unable to understand the limitations of correcting a dog and will therefore use any advice to correct as a green light to be cruel and harsh with their dogs. I have never had this happen when I take the time to explain the thought process behind using both rewards and punishment judiciously, which ironically, relies on an understanding of how they affect behavior, Skinner-style.

LGDs have been bred for centuries to be quick learners and to rely on binary feedback (both positive and negative) for their learning process. When a certain behavior brings them a reward, whether it’s internal (the satisfaction of driving off a threat, nurturing a lamb or playing with a canine playmate) or external (praise or food from their owner), it easily becomes something they want to do. When a certain behavior brings them a punishment (a shock from an electric fence or a head butt from an angry goat), it easily becomes something they don’t want to do again. These are best seen in younger pups, who are like little learning sponges. They soak up the binary input and learn solid lessons that they adhere to for their lifetimes. Sure, even the best of LGDs will color outside the lines during their teenage phases, but a small reminder of the lesson will generally bring them back in line. What people who loudly say that we have to prevent LGD pups from practicing behavior we don’t want aren’t telling you is that there is often a better, more efficient way to achieve that: corrections.

When our dogs do not receive the input they need, ideally as pups, they don’t learn the lessons that are necessary for them to be successful. Conveying this to people requires the clear communication that corrections will be necessary. These are dogs who are living with vulnerable creatures, and they are required to be with them full time, outside of human input a fair bit of the time. The western people who think they know something about how to raise a LGD will tell you ridiculous things like “a LGD shouldn’t be in with animals unsupervised until they are 2 years old” and “LGDs were never meant to be good with poultry” instead of telling you the information needed to guide your LGD to being a successful guardian. I’m terribly sorry to be the one to have to tell you this, but there are no for profit producers who are going to feed and house an extra large dog for 2 years before they can be trusted to work. No one has the time to babysit their dogs for years in the hopes that they will be appropriate guardians in the end. Further, there are no working LGDs in their countries of origins who aren’t learning on the job from day one. This is not to say that dogs who have not been raised as mainly full time guardians can’t become one, but it is to say that that isn’t the most efficient way to do it. It’s not even the most effective way to do it.

LGDs need clear, effective communication. They need to know the boundaries of the job early on. They need to know that mouthing stock, of any shape or size, is inappropriate. They need to know that escaping fences because they feel like it is not ok. They need to know that fighting over food, jumping on people, chasing stock or enthusiastically licking a newborn while its mother cries is wrong. They need to know that being polite is a requirement for living well with people and stock. While it’s important to note that a dog must be taught what TO DO, it is still very important to correct them effectively and efficiently in order to teach them what NOT TO DO.

young kuv learning to herd sheep 3 of 3

Some dogs will need very few corrections in their lifetime. Some dogs will need many, many more. This does not mean that the former are better dogs, although, it is highly likely they will be perceived that way. This also does not mean that the latter are less apt to be good LGDs – in fact, it can mean the opposite. These can be dogs who are the most tenacious with predators, the best and most creative types who don’t let anything get in their way and who can stay ahead of the wiliest of predators.

Don’t be fooled into thinking that straight forward corrections are inherently abusive or unnecessary. Judiciously applied, they provide an integral part of the basis for understanding the “do nots” of life as a working dog. Training highly aggressive dogs to live with prey animals, to protect and nurture them in the face of threats, to live together well with other dogs and people requires some form of active, timely correction. In tomorrow’s post, I will outline exactly what that looks like.


P.S. So far I have refrained from putting forth the most obvious argument, that LGDs use corrections with each other.  This is something that people who own older LGDs or keep familial groups often forget when talking about raising/correcting pups. As humans, we cannot “replicate” well what older LGDs do with younger ones to teach them, but we can take a page out of their book regarding the fact that proper corrections are an appropriate and highly effective form of communication.