Mr and Mrs Dog: Our Travels, Trials, Adventures and Epiphanies

Luke and June each emerge as individuals with something to say. Few people can write about communication between human and dog as well as McCaig. He has learnt to listen to his dogs, and respect their  points of view. 

Interwoven with this story are reminiscences of McCaig's early days with dogs, an account of the development of pet dog training, and observations on different approaches to training. McCaig's first dog was Rascal, a Cocker. Rascal was mannerly, though his owner doesn't remember training him. McCaig's home town didn't need a trainer, he tells us, because the streets were safe, and mothers stayed at home. If there's a gap in the logic here, it's because McCaig, like his hero, Konrad Lorenz, is an intuitive rather than an analytical thinker, as becomes apparent in his discussions of pet dog training.

 It's only quite recently that people thought it worthwhile to train pet dogs. For centuries, trainers focused on working dogs, like hunting dogs. McCaig notes that obedience training for pets was developed from exercises used to train dogs for the military, and only pedigree dogs were allowed to enter early obedience competitions. Obedience training, then, wasn't initially geared to making pet dogs mannerly. 

McCaig travelled to observe classes representing three approaches to training, Koehler, drive training (developed from ethology), and 'positive' training (developed from behaviourism). He sketches in a background to each approach, with anecdotes about the founding fathers, Bill Koehler, who trained for Walt Disney and wrote 'The Koehler Method of Dog Training', Konrad Lorenz, and B.F. Skinner respectively. McCaig also observed a class where owners were taught how to use e-collars. He contrasts these four classes with Nicholas Dodman's pharmacological remedies for behavioural problems.

The Marketing of Behaviourism

McCaig starts his chapter on 'Behaviorism' with a quote from Noam Chomsky on Skinner's odd conception of science. McCaig then explains how a theory of learning was developed from lab experiments recording how animals responded to rewards and punishments. Skinner believed that the animal's feelings, understanding of the situation, and inbuilt desires weren't relevant. Speculating about what went on inside the animal was unscientific. However, for a sheepdogger like McCaig, most of what's important is going on inside the dog and the shepherd. Sheepdogging depends on a dog's inbuilt desires to work sheep, his understanding of the ways of sheep, and communication between shepherd and dog, plus a feeling of mutual trust if each is to take what the other says seriously. It's little wonder that McCaig sees behaviourism as deficient. 

Karen Pryor and others adapted behaviourist methods used in training wild animals and applied them to dogs, claiming their 'positive' approach was kind, because it aimed to avoid punishment. Though McCaig doesn't mention this, 'positive' trainers try to avoid what in operant conditioning (a method derived from behaviourism) is called 'positive punishment', which is adding something, like a shock, to make a behaviour less likely to happen. It's a bit confusing, because, while 'positive' trainers don't like positive punishment, by using rewards to encourage a behaviour ('positive reinforcement', for example, caressing a dog for sitting) logically, they sometimes withold rewards to discourage a behaviour (eg turning away when a dog jumps up). This is 'negative punishment'. So calling their approach 'positive' is misleading and confusing. 'Positive' is a marketing term rather than a precise, scientific label.  

Are 'positive' methods kind and safe? McCaig notes that 'positive' trainers see their methods as 'less likely to cause problems than over-harsh or ill-applied punishments' (p80), though you don't have to call yourself 'positive' to agree that such 'punishments' can be counter-productive. McCaig later says that 'positive' methods are probably safer, citing examples of two sheepdog handlers losing their tempers, upsetting sheepdogs so seriously that their later work was impaired (p182). However, 'punishment' in an everyday sense isn't 'punishment' in Skinnerian terms, if it fails to stop the targeted behaviour. For Skinner, 'punishment' was simply a stimulus that made a behaviour less likely, and the feelings of the animal were irrelevant.

 For most pet owners, the dog's feelings are important, so we try to guess what they are. When Skinner fed or shocked lab rats, it was simpler, food was clearly nice, and shocks were clearly nasty. But outside the lab there's a lot more going on, so we can't always tell which particular stimulus stopped a dog from doing something, or if we can, we can't always be sure why it worked. Maybe through upset, or surprise, or perhaps by simply providing information. Owners sometimes give messages meaning 'stop, don't go there/do that, it's not safe'. Dogs can give similar warnings to humans, often by blocking our paths, or pulling on our clothes. We may understand one another and stop, or misinterpret the information, or override the warning.  

We have relationships with dogs, so trust, and our ability to communicate are important. Relationships didn't enter Skinner's model It's difficult to fit the richness of our everyday lives with dogs into a Skinner box. Behaviourism may help to explain simple situations, but if we try too hard to fit a theory based on the recorded behaviour of lab animals, a theory where feelings and relationships were irrelevant, to our much more complex relationships with dogs, we can end up with confusion rather than clarity.  

McCaig also stresses that the 'why' of training is important, as well as the 'how'. He quotes an eloquent passage from Heather Houlahan on training a dolphin with 'positive' methods. The dolphin, wrenched from the sea and his family, stuck in a small enclosure, unable to hunt, is fed dead food to encourage him to perform tricks to amuse humans. The methods may not involve positive punishment, but the training is hardly 'kind'. Conversely, if owners are afraid of telling their dogs by whatever means 'don't do that', in case it upsets them, they may be putting their dogs' lives at risk. That isn't 'kind' either. The same point about the 'why' of training was made by Steven Lindsay in his science-based work on applied dog behaviour and training (1).  

Although McCaig expresses doubts that 'positive' methods are either as 'scientific' or as 'kind' as believers claim, he concludes that they 'may appeal to those with an intellectual or scientific bent', as well as to people who want to see themselves as 'kind' (p108). This conclusion doesn't fit his previous arguments. Furthermore, it's people with a 'scientific bent', like Steven Lindsay and Adam Miklosi (2) who spell out some of the limitations of behaviourism. It's non-scientist owners who tend to swallow the marketing language of 'positive' and 'kind', now fashionable in pet dog training. The words appeal to feelings which, coupled with a desire to conform, can act as barriers to enquiry.  

Behaviourism emerged from psychology. For a long time it was difficult for psychology researchers to obtain funding to study what might be going on inside an animal. However, this changed with the growth of research into canine cognition - how dogs perceive and understand the world. Psychologists have moved away from assessing dogs as deficient humans, and have become interested in canine skills. Ethologists are also researching canine cognition and specifically canine skills. This shared interest in cognition makes for more interesting dialogues between psychologists and biologists.  

Training as Practical Ethology 

Ethologists study animal behaviour by observing animals. McCaig describes behaviourists as interested in nurture, and ethologists as interested in nature. However, as dogs change their behaviour according to their environment, ethologists are also interested in the impact of environment, and are constantly reassessing what it means to be a dog. Feral dogs in Italy show less pack-like behaviour than those in West Bengal, for example, so neither environment alone is enough to draw confident conclusions about canine social life (3) (4).  

As McCaig notes, hunters were the first practical ethologists. In a way, dogs are intuitive ethologists. They study us, and can understand our body language better than we understand theirs. McCaig describes how Luke and June can do what he cannot. They can immediately grasp the social order in a flock. 'They know whether the sheep are ready to fight, split up, or break for the tall timber, because the sheep tell them what they mean to do' (p103) 

Humans may write about animal behaviour, but dogs aren't distracted by words, they just perceive. A good trainer can instantly size up a dog and owner using intuitive intelligence. In other words a good trainer is more like a dog than is the average human. This intuitive intelligence comes both from experience, and from being willing to listen to dogs, to be sure of our perceptions. McCaig quotes Behesha Doan on a difficult dog:'I had to stop trying to be like, think like and act like all the trainers I had sought so hard to emulate. I would have to have total confidence that this dog wanted to communicate with me, and I had to have confidence in my ability to hear him' (p57)  

A key difference between dogs and wolves is that dogs want to communicate with us. As Vicki Hearne (5), and after her, Miklosi noted, you can tame a wolf, but this does not make it domesticated. A tame wolf doesn't have the same inbuilt desire to interact with humans that a dog has. Miklosi argues that 'training should preferably be part of a daily interaction and should be based on the rich social tool set that is available for both the human and dog' (6). Success 'depends also on the genetic constitution of the dog, and the human's social skills' (6). He notes that tame wolves raised with humans aren't as socially competent as dogs (7). It isn't possible to use this 'rich social tool set' with wolves because they aren't as motivated to interact with us, and aren't as good at understanding us as dogs are.  

McCaig, as a practical ethologist is particularly concerned about learning to read his dogs better. All trainers need to understand simple messages, especially 'is this dog going to bite me?'. McCaig wants to understand more. He's aware that his dogs know more than he does about the ways of sheep. He wants a two-way conversation, to learn from them what he doesn't know and they do. Ethology, then permeates this book, though only two chapters are explicitly devoted to the subject, the chapter called 'Ethology', and the account of training in drive. 

Miklosi also notes that dogs are unique in their ability to learn the social norms of different species (8), a key concern of a sheepdogger. Sheepdogs learn about sheep both by being with sheep, and by being with dogs that are mannerly with sheep. Human social skills affect our ability to train dogs, as do the social skills of the other species we let our dogs meet. McCaig mentions that his sheepguarding dogs protect 'their' sheep from unmannerly dogs. And savvy old ewes are better teachers for pups than are yearlings that run mindlessly from any possible danger. 

This book is also mostly about training, if one takes Miklosi's broad view of training as teaching that's embedded in our everyday interactions with our dogs. In that sense, McCaig did in fact train his first dog, Rascal, through their everyday interactions, though he wasn't aware that he was 'training'. If McCaig thought that dogs didn't need training in his home town, because mothers stayed at home, very likely the mothers put most of the effort into training those dogs, by telling them what was OK, and what wasn't OK. Dogs that could come and go freely  when roads were safe also tended to be less demanding of their owners. In some ways, training was easier then. 

What Can Professional Trainers Do? 

McCaig argues that dogs become mannerly as a byproduct of the bonding that happens during training - teaching heel, sit and fetch don't make a dog mannerly  (p185). This tells us that he defines 'training' narrowly, as 'what happens in training sessions'. Now, many people share this narrow view, Cesar Millan and Karen Overall, for example. Millan commented that obedience-trained dogs often behave badly, and saw training as separate from teaching a dog how to behave well (9). Overall saw trainers as ill-equipped to deal with behavioural problems (10). Yet competent trainers are aware that manners training is about teaching dogs social rules, not just commands. The commands can help you to enforce those rules. Teaching social rules is a form of behaviour modification. If McCaig's dogs are mannerly, it's because he taught them manners. He stresses that a pup 'needs to know the household's rules' (p179). This is part of training.  

Finding a good professional trainer then, is more complicated than McCaig's simple prescription of finding someone who shares your goals, and choosing between two labels, 'traditionalist' and 'positive' (p181). Traditionalists, he argues, help you win obedience trials, while 'positive' trainers just want dogs to stay with owners, rather than being given to shelters.  

Firstly, his classification of trainers into two camps, 'traditionalist' or 'positive' is misleading. The label 'positive' tells you that the trainer is likely to use 'hands off' methods. However 'positive' trainers vary in terms of methods, and disagree among themselves about what sorts of punishments are acceptable. I recently observed a 'positive' trainer with a puppy. He used hands-off methods with frankfurter pieces to encourage sitting, then a modified version of Koehler's changing direction leash technique to discourage pulling, and then deterred the pup from chewing his rug by prising the the pup's mouth open. The pup yelped at the pressure on his gums.  

If trainers prefer not to market themselves as 'positive', are they by definition, 'traditional'? Koehler's book was published in 1964, and has been very influential among trainers as a whole. Though the book´s description of extreme corrections is well-known, it's less well known that many positive training methods are based on Koehler (McCaig doesn't mention this). 'Traditional' might accurately apply to someone who uses pure Koehler, but not to most trainers who prefer not to call themselves 'positive'. One might expect older trainers to be more 'traditional', but many are keen on learning more, both from other trainers, and from the latest academic research.  Perhaps the most influential modern trainer is Steven Lindsay, who does not wear the 'positive' label. He argues for building trust between dog and owner, using the dogs desire for pleasure by training with rewards, and minimising punishment. Reward and punishment 'flow from the dog's ability  to produce outcomes' (11). The aim of training is to enable a dog to adapt, to cope. Lindsay drew on both his own ample experience and scientific research to develop his ideas.

 It may be a human trait to divide people into two camps, usually 'our group', and outsiders, but it's not helpful for choosing a dog trainer. The labels oversimplify, and a label is no guarantee that the trainer is competent. As Vicki Hearne emphasises 'No method or impersonal theory relieves the trainer of the burden of judgement' (12).   

So what can competent professional trainers do, whatever their label? Firstly, help owners to apply what's learnt in class to everyday life. McCaig implies that obedience training can help improve a dog's behaviour when he says that trainers who focus on obedience want 'real-world off-lead reliability in the presence of real-world distractions' (p181). However, he's ambivalent about the link between obedience-training and everyday behaviour, also arguing that it doesn't make dogs mannerly. This may be because many owners are inconsiderate, rather than because the trainers are incompetent. McCaig's first 'magic principle' for having well-behaved dogs is 'don't be nuts', because dogs take their cues from us (p179). Being considerate is part of not being nuts. A considerate owner prevents his dogs peeing on the hubcaps of his neighbour's cars, so the dog learns not to. Some owners don't realise that not everyone wants a dog peeing on their cars, or a  large-breed adolescent dog 'being friendly' by jumping up. We need social skills in order to teach them to our dogs. Trainers can teach dog walking etiquette, but it's more difficult to teach thinking of others to people who haven't developed the habit.  

McCaig' second 'magic principle' is that 'puppies are babies', so we need to be patient with them (p179). Unfortunately, as Patricia McConnell notes, owners sometimes perceive adult dogs as 'needy infants', thereby creating brattish adults (13). A competent trainer can both persuade a demanding owner to ease up on a pup, and persuade a doting owner to treat their dog like an adult. His third principle is 'exercise your dog' (p180). Dogs vary a lot in terms of the amount and type of exercise that suits them. A competent trainer can advise on a particular dog needs. Lastly he advises 'give your dog a job' (p181). Again, a competent trainer can suggest activities that suit a particular dog. 

Competent trainers can do much more, such as help owners teach dogs how to behave with other species. They can also size up and communicate with dogs and owners, and if necessary, vary their techniques in line with what they feel would work with a particular dog-owner combination. They can structure long-term plans of action, progressing from easier to more difficult skills. Being methodical and using checklists are also useful skills, especially for detecting dogs which may need medical treatment. Experienced trainers can add to this list, and there are many areas where experienced trainers tend to agree, for example, that pups learn canine social skills better from mannerly adult dogs than from brattish adolescents.  

So, though McCaig stresses that training is important, perhaps he underestimates how helpful professional trainers can be, and how much agreement there can be between experienced trainers, whatever their label. A key test of competence is the trainer's track record in tackling the kinds of challenges you face, whether it's basic manners, or learning two-way communication with dogs. It's also worth observing a trainer in action before making a commitment - unfortunately, incompetent trainers can do more harm than good, whatever their label.   

Taking the Tablets 

McCaig comments of Dodman, 'I like dog training ideas, but absent real live dogs, I can't evaluate them' (p94), a curious comment, because Dodman isn't really interested in training. Dodman has written several books on raising well-behaved dogs, though he doesn't actually own a dog, and sadly comments that he never meets well-behaved dogs at his consultations.  

Dodman sees pet dogs as family members, who don't need to learn to heel. He sees dogs as like small children, deficient adult humans. McCaig, like Miklosi, is aware that dogs can do many things humans can't do, so it's a mistake to evaluate them as though they were human. Their superior skills make for interesting conversations, which are part of training. 

Dodman prescribes pharmaceuticals rather than training for behavioural problems. McCaig notes that dogs can be stressed by being isolated all day, and not having enough exercise or interaction with their owners, so Prozac may help them (p182). So McCaig implies that it's a choice between putting effort into your dog, or dosing it with pills. True, as with humans, exercise and social contacts tend to make for happier lives, but they don't cure all ills. Sometimes medication is necessary, for example for thyroid problems. Dogs may also benefit from temporary medication, to relax enough for us to teach them better ways to cope with stressful situations. Pharmaceutical and training approaches to behavioural problems aren't necessarily mutually exclusive, though veterinary behaviourists and trainers often have trouble communicating with one another.  

E-collars, the Devil's Tool? 

Lastly, the vexed question of e-collars! McCaig notes that the American Veterinary Society of Animal Behaviorists classes them as a 'method of physical punishment', and to be avoided (p79). McCaig himself finds e-collars scarily powerful, and argues that they 'should be restricted to trainers with experience, good timing and dog sense' (p183). It's too easy for someone who's incompetent, or unable to control their temper, to do a lot of damage with one. Both he and Behesha Doan, who teaches handlers how to use e-collars, believe they shouldn't be freely available in pet stores. So it's a surprise that McCaig also advises 'if you are training retrievers, bird dogs or coon dogs, buy an e-collar' (p179). That bald imperative contradicts his more thoughtful comments. What does a novice do with the new e-collar if there's no-one competent to show them how to use it? And what if they don't trust their temper, or simply prefer the dog to understand where a  stimulus is coming from? 

Sheepdoggers don't favour e-collars, which can interfere with the intense communication between dog and shepherd. It can confuse a sheepdog not to know where a message is coming from. One problem with his imperative is that McCaig assumes standardised 'working dog' cultures for other types of work, like retrieving. E-collars are often used by retriever trainers, so he assumes they're the best tool for the job. Yet there's a big difference between a teaching a pet dog to retrieve, so the owner and dog can do something enjoyable together, and training a dog kept in kennels, and used solely for competition retrieving. 

Both the AVSA and McCaig are also confusing a training tool with a training approach. McCaig calls e-collars 'pre-theoretical' (p72), but the theory comes from the human using the tool, rather than from the tool itself. Obviously, e-collars can be used at high levels as powerful aversives, but McCaig describes Behesha Doan carefully adjusting the level until the dog notices it, looking alert rather than upset. A sound may be painful at high volume, but not once the volume goes down. At low levels, e-collars may work by simply reminding a dog that the owner exists.   

McCaig notes that we use many means of communication with dogs, such as voices and whistles, or ways to control dogs, like leashes and head halters. These can fail us. We may lose a whistle or our voice, or run out of treats. It helps to be versatile. Snoozing beside me is Tilly, almost 15, almost totally blind, and very deaf. I can no longer communicate with her through hearing or sight, so I use touch, taste and smell. Touch helps orient her. She's also happy when something tasty arrives in her bowl, and dances excitedly when it's time to go for a walk, to check everything outside intensely for new smells. Tilly's near the end of her life. That's enough for us. What if she were young and deaf? A distance communication device could allow her to have more freedom. True, e-collars are probably riskier than alternatives like vibrating collars. True, there are good reasons for making it difficult to buy them. However, it's all too easy to make assumptions  on an emotive topic like e-collars, without checking to see if our assumptions are correct. 

British pet owners tend to have a visceral dislike of e-collars, coupled with lack of information, since our trainers don't use them. They're simply not an option for most of us, so most information on their usage comes from Americans. Steven Lindsay sees their effect at low levels as resulting from a startle rather than pain (14). He argues that 'they should not be used in the absence of reward-based training' or as a crutch (15). Dogs which come back after a recall prompted with an e-collar should be praised and rewarded (16). He's also critical of the low quality of products and instructions from many manufacturers, and argues for limiting the power available to inexperienced owners. Though McCaig touches on some of the issues raised in Lindsay's lengthy examination, McCaig certainly doesn't give enough information to draw the conclusions he reaches.  

Intuitive and Rational Intelligence 

It's perhaps churlish to pick so many holes in this book, when the sheepdog stories and the training anecdotes are so well told. So, there are gaps in McCaig's logic, but then most of us think we're more rational than we actually are. As Scott Adams says 'Most people wouldn't know the difference between a logical argument and a porcupine stuck to their forehead' (13) 'Positive training' has feelgood appeal, even if it's an illogical name to give an approach that sees negative, rather than positive punishment as OK. 'A professional salesman will avoid negative phrases and use positive-sounding words' advises Adams (17). So what if it's a confusing term, and the operant conditioning term 'positive punishment' itself tends to confuse. For McCaig isn't the only person who believes it includes simply losing your temper with a dog. As Adams advises, 'confusion is your friend in sales' (18). 

Besides, who wants to be rational? It's handy for finding plausible reasons to justify our more stupid decisions, but in everyday life we often have to act fast, and our intuitive intelligence allows us to take quick decisions. Then again, working on analytical skills is more boring, and can make your brain hurt. Well, here's why it's worth trying to stand back and balance intuition with reason. If we don't check whether our perceptions and beliefs are true, we're condemned to endlessly repeating mistakes. Working out whether arguments are logical and fit the facts, being methodical and using checklists, can all help us take better decisions in the long term.  

Humans and dogs are both very adaptable, and since dogs adapt to us, their behaviour tends to reflect the environment we provide for them, including our expectations of them. This helps feed our beliefs that our perceptions of dogs are correct even when they're at odds with other people's perceptions. Often, people from different dog cultures simply don't talk to one another, or if they do, they try to shout one another down. So it's a great strength of this book that McCaig is willing to listen to people from different dog cultures, and report back to those of us who prefer to lead more comfortable lives. 

Review by Alison Lever, 2014

 Acknowledgement: Thank you to Tiffani Howell for comments on an earlier draft of this review.

 Notes

 1) Lindsay, S. (2005) Applied Dog Behaviour and Training vol 3, Procedures and Protocols. Blackwell, London, p709

 2) Miklosi, A. (2007) Dog: Behaviour, Evolution and Cognition. Oxford University Press, Oxford 

 3) Boitani, L., Franscisci P., Cuicci, P,and Andeoli, G. (1995) Population biology and ecology of feral dogs in central Italy. In Serpell J. (ed) The Domestic Dog pp 217-244. Cambridge University Press, Cambridge

 4) Pal, S.K. (2004) Parental care in free-ranging dogs, Canis familiaris. Applied Animal Behaviour Science, 90, 31-47 

5) Hearne, V. (2007) Adam's Task. Skyhorse Publishing, New York pp22-23 

6) Miklosi, A. (2007) Dog: Behaviour, Evolution and Cognition. Oxford University Press, Oxford, p240  

7) Miklosi, A. (2007) ibid, p239 

8) Miklosi, A. (2007) ibid, p22 

9) Millan, C. (2011) Cesar's Rules. Chapter 3 'Dog Training vs Dog Psychology'. Hodder and Stoughton, London 

10) Overall, K. (1997) Clinical Behavioral Medicine for Small Animals. Mosby, St Louis, p101 

11) Lindsay, S. (2005) Applied Dog Behaviour and Training vol 3, Procedures and Protocols. Blackwell, London, p727 

12) Hearne, V. (2007) Adam's Task. Skyhorse Publishing, New York, p132 

13) McConnell, P. (2002) The Other End of the Leash. Random House, New York, p107 

14) Lindsay, S. (2005) Applied Dog Behaviour and Training vol 3, Procedures and Protocols. Blackwell, London, p585 

15) Lindsay, S. (2005) ibid, p570 

16) Lindsay, S. (2005) ibid, p595 

16) Adams, S. (1996) The Dilbert Principle, Boxtree, London, p214 

 

17) Adams, S. (1996) ibid, p215