Dogs Who Lead

[ELLIOTT HUMPHREY, now in his forties, has spent most of his young and middle years working with animals. He has herded cattle in Wyoming, driven mules in California, curbed horses in Kansas City and Camp Lee, and helped build up a famous stable of Arabs in New Hampshire. He is equally wise in the ways of a dog, particularly the German Shepherd, the breed with which he has had most success in training the dog guides for the blind. For the past eight years he has been associated with The Seeing Eye, an organization with headquarters at Morristown, New Jersey. We have put to Mr. Humphrey a series of questions which will, we think, kindle the interest of every dog lover. — THE EDITORS]

What type of dog do you use and why?

First we want a dog that is large enough to handle its master. In a thoughtless moment, the master may not stop when his dog gives the signal and it becomes the duty of the dog to pull his master back to safety. Inasmuch as the dog is not to be a plaything and will be worked day in and day out, it must have an all-weather coat that permits work winter and summer, rain or shine. It must have sound feet to enable it to work all day, every day, on every kind of pavement and roadway, and must be so built that it can work easily and quickly with a minimum of effort and a minimum of fatigue.

Another point, and one which has a psychological influence on the blind master, is that the dog must be goodlooking, so that the master will be proud of it rather than feel that he must apologize for it.

Naturally the dog must have the brain to take our course of instruction, and especially must he have what we term ‘a sense of responsibility,’ for that is what makes the dog wish to guide his blind master and impels him to make decisions if the master is uncertain or in error.

These several requirements naturally limit our choice of breed. We have used, and do use, many breeds, but about 95 per cent of the dogs which best meet all the specifications are German Shepherds, or what are more commonly called e police dogs.’ These dogs are particularly well fitted for the work because they are of a good medium size, capable of handling a man, have a weather-resisting coat, good feet, good structure, a teachable brain, and, in many cases, a splendid sense of responsibility. It is possible that their sense of responsibility comes from hundreds of years of herding sheep in Germany, where the herding system allows the dog to use its head instead of requiring it to be constantly obedient to a word or gesture from the shepherd.

The remaining 5 per cent are animals which are not typical of their breeds, I have seen on the Continent an excellent working dog that was an Airedale. Now the Airedale as a breed would not be suitable because it has been bred for generations as a fighting dog, and a fighting dog does not make a safe guide for the blind. The dog that I saw was an exception to the breed. In the same way, we have used, or seen used, dogs from many breeds which were excellent guides, but which were equally exceptions. These have included an Irish Setter, a Pointer, Dobermanns, a Boxer, and Labradors. The Labradors have been quite high in the average of suitability— especially when they lacked the good nose of the hunting dog.

Is it true that you use only females?

No, it is not true. We use both sexes. We use about 60 per cent females and the rest males. Many questioners would prefer to have us say that the female is smarter. Actually, we see no difference between the sexes in this respect. In general, the female is slightly easier to handle, yet there are certain temperaments among the blind that require the use of a ‘harder dog’ — which is to say, a male.

Do you find it very hard to get dogs that have enough intelligence to do the work?

After a rather short lifetime spent almost entirely in work with animals, I should say no, it is not difficult to find dogs with enough intelligence to do guide work. The mere learning is not difficult. Most dogs have the ability to do it. I think that much of the misunderstanding concerning dog intelligence comes from a misuse of the word. Most people, when they speak of a dog or animal as being intelligent, mean that the animal does what they want it to do. But it may not take a great deal of gray matter for a dog or for an individual to follow instructions. On the other hand, it may take a great deal to study out a way to avoid doing what has been asked.

In our own rating of canine intelligence, we give an animal just as high a grade if he uses his brain to avoid doing what we ask as we give him if he uses his brain to obey our command. The matter of whether he does what we want him to do or what he prefers to do comes under an entirely different heading in the rating sheet — a heading which we characterize as willingness; and willingness may be quite devoid of intelligence.

Intelligence is not only the ability to learn, but the ability to use wisely what has been learned. There are certain individuals, and they are not at all limited to the species we speak of as ‘dumb animals,’ who may learn quickly, by rote, but later act without thinking of what they have learned. From that standpoint there is probably no more alert animal in the dog family than the French Poodle. He will learn the ordinary obedience exercises, in which exact obedience is demanded, in much less time than the German Shepherd. But he has simply learned by rote. He has learned to perform those exercises as a trick. It never occurs to him not to obey what his master commands.

Now at the other end of the scale we find the Dachshund, regarded by all Poodle lovers as rather dumb. In his peculiar way, the Dachshund is probably the most intelligent of all breeds, if he is typical of his breed. His seeming dumbness lies in the fact that he uses his brain to avoid doing what has been asked. A really good Dachshund simply cannot be taught to obey, whereas a good Poodle obeys naturally — at least the owner will tell you he does.

In teaching dogs to guide the blind, we can’t use the Poodle type of thoughtless obedience, nor, aside from its size, can we use the Dachshund type of intelligence that never obeys. In our work we have to have a special type of intelligence — a type which can learn with reasonable quickness, but which never obeys without thought. We must have a dog that remembers lessons and is able to use past experience in present action. He must obey when it is safe to obey, and disobey if the command does not seem safe from the dog’s point of view. This is necessary because of the fact that the future master will not be able to see danger and may give commands which would be dangerous to obey in Poodle manner. On the other hand, the blind man may give many commands which are on the border line — which may be partially obeyed but must be partially disobeyed.

How do you teach the dog to obey sometimes and to disobey at other times?

I want to emphasize the fact that there is a vast difference in exercises. Some may seem simple to the master but are extremely complicated to the dog. When you throw something and say ‘Fetch,’ what does the dog have to do? First he has to see what, it is that you threw. Second, he must leave master, and if he really loves his master he does not like that. When he reaches the object (and he must remember where it went) he must drop his head. No dog likes to drop his head. He must open his mouth and pick up something that is apt to be distasteful to him; hold it, although it may put his teeth on edge; return to master balancing it; take his position at master’s side and hold it while master fumbles around, and then — and this is often hardest if the dog likes to chew on the object — open his mouth and give up what he has on the command, ‘Out.’ It is really rather complicated, and it is still more complicated if we make it a ‘simple little thing like bringing my slippers’ — in other words, distinguishing the slippers from the shoes or pyjamas or cane, or whatever else is in the closet.

The only general rule we could give as to how we teach these exercises is that every dog pupil must be considered as an individual. The instructor who has spent years in the study of the animal mind and animal reactions considers his pupil as an individual and strives to give the lesson so that that particular individual will understand it and respond with the proper reactions. For correct responses the dog, or pupil, is given a reward, generally a reward of word alone. For improper responses the dog again receives a word, but it is a corrective or shaming word. In this way the animal is taught what is wanted and what is not wanted. The success of any instruction is almost entirely dependent on the trainer’s ability to see the slight individual differences between different pupils and so change his manner of presenting each lesson that each pupil will understand it and react as the instructor wishes.

How do you teach a dog to judge heightfor instance,to know what awnings to go under and what awnings not to go under?

When we start the dogs on overhead obstacles, we don’t start with something that is just the right height. We start with something that is very low, something the master could n’t possibly go under. When a dog has the idea that there are certain things he can go under but master cannot, it becomes a matter of simply varying the height of the obstruction. Once he has the idea, and heights commence to vary, he very quickly learns what he and master can go under, and what he and master cannot.

Now this leads to one question that may not be asked but which I should like to suggest. If the dog learns what awnings he and master can go under, it would naturally imply that the future blind master must be of approximately the same height as the instructor. This is often not the case. What happens then? If the instructor is short and the blind master tall, what happens? For the first two or three days the blind man bumps his head, but it does not take more than two or three bumps for the dog to readjust himself to the height of his new master. From that time on, bumps cease — that is, they cease if the dog has learned the first lesson, that ‘there are some things I, as a dog, can pass under which my master, as a man, cannot pass under.’

This ability to decide what dog and master can pass under brings us to that quality which we spoke of as ‘a sense of responsibility.’ By that we mean the ability to decide whether a certain movement is safe or unsafe, whether a command given by the master can be safely obeyed or whether it must be disobeyed. From the standpoint of responsibility, we find that the German Shepherd is particularly adapted to the work, and probably this is the deciding factor which keeps us more closely to the German Shepherd than to any other breed.

I remember two dogs, working with two blindfolded trainers, who were presented with exactly the same problem. One dog was a German Shepherd, the other was not. These two blindfolded men had just walked along a street and made a crossing in perfect safety. They were to go one block farther and then turn around and come back. This meant that they would be crossing the same street not more than four minutes later. At this particular street crossing there was a manhole cover, and when these two men were sent down that street there was no thought on the part of the supervising instructor concerning the manhole. The street was chosen because a lot of garages opened on to it, and by going two blocks down and two blocks back you were almost sure to have an automobile shoot out of a garage and across a sidewalk in front of you to occasion what we speak of as a ‘traffic check.’

It so happened that after the men had crossed the street the first time some workmen removed the cover of the manhole, and were a little slow in putting up a railing around it. The men and dogs turned around, came back, and started to recross that street. The masters gave the command, ‘Forward.’ Both dogs hesitated, because there was now a gaping manhole in the street which they had crossed safely before. Both masters were absolutely sure that the street was clear, because they had just crossed it going the other way. When neither dog started, both masters said, ‘Phui,’ which is a corrective word, and ‘Forward.’ Each of them, with a little irritation, shook the harness and insisted that the dog start.

Now for the difference in reaction. The German Shepherd knew that master was wrong; he knew that master was insisting on obedience, but in spite of that he pulled back as if to say, ‘Now, listen, you’re wrong — you will have to come with me.’ The second dog seemed to say, ‘Well, if you’re so insistent we’ll try.’ He took a step forward and then jumped the manhole. It was only because of the alertness of the supervising instructor that the man did not step into the manhole and fall to the bottom. Let me add that the dog which failed was eventually discarded.

What did the German Shepherd do then?

Well, he pulled back, and then very easily but very insistently pulled his master to the right. He placed his own body between his master and the hole and went around it, straightened out, and then went to the curb on the opposite side. That illustrates what we call ‘a sense of responsibility,’ or the willingness to disobey an unsafe command even though master may be extremely confident that the command he has given should be obeyed.

Personally, I like to consider our system of educating dogs in comparison with the present system of educating children. Our system is laid out so that each step is a normal passage from the last lesson and a normal link to the next lesson. The first step is learning a set of obedience exercises, which might be compared with the multiplication tables for children. When these are well in hand, we put a harness on the dog and step him up, as it were, by offering him problems in harness. These problems, while he has teacher’s help, will be solved through the use of the obedience exercises which he has already learned. Once this grade is successfully passed, problems are given the dog to be solved without help. Then we go a step further to the abstract problems in which the master’s command may be absolutely dangerous and the dog must obey the intent but not the letter of the command. Through willful disobedience he must accomplish the original intent of the master. This gradation of exercises is directly comparable to the step-up system in human education.

Naturally we must have competent instructors, and so far we have had to teach all our own instructors. Preferably we have taken men who, while they like dogs, have had no previous experience in dog training. This means that we have less to unteach before we start to teach. In working with them we have emphasized the fact that a dog never can think like a human being, so the trainers will have to learn to think like a dog. But that will be only half of their problem; they will also have to learn to think like a sightless man, because the problems of a blind man are distinctly different from those of the seeing. We explain that the work will be hard, even dangerous, and sometimes dirty; the hours will be long, and at no time will the wages be exceptionally high. Instruct ors get a good living wage, but, as in most professions, the satisfaction of doing something for someone less fortunate must be a part of the pay.

If the applicant is accepted, the first part of his apprenticeship is to learn what it is like to be blind. Perhaps he arrives on a Sunday: during the afternoon and evening he is allowed to look over the house and grounds and to familiarize himself with his room. On Monday morning work starts, and for the next four weeks the apprentice instructors are, to all intents and purposes, perfectly blind. After a preliminary lecture, they are all given a sleepshade, which is an absolute blindfold, not permitting the least ray of light to enter. They wear these blindfolds day and night. They eat blind, they walk the house blind, they shave blind, they are given dogs and are handled as a blind class. They are taught to use those dogs in traffic and are carried to exactly the same point that a class of blind students would be carried, to the point where the student is told that he is safe to go home with his dog guide.

During this period, the apprentice is just as sightless as though he had been suddenly blinded by accident, the only point of difference being that he knows the time will come when he can see again. The work during this period is hard, mentally hard and physically hard. For the first few days the boys do not get a great deal to eat, for they can’t find it on their plates. They don’t get very good shaves, because they cannot see into the mirror. They bump into walls and posts, are hesitant in traffic just as a blind master would be while learning to use his dog. It is a difficult course, and some apprentices quit. We want them to quit if there is any quitting in them. We prefer to have them stop right then, because the full course is going to take about four years and it is less expensive for us and for them if we find out right at the start that they have the quitting streak. If they survive this month of blindness, they are ready to start as real apprentices, and their work for the next four years will be almost a college course.

Daily they will work with dogs and hear lectures. These lectures teach theory, maintained on an even basis with actual practice. They will learn during the first year the mechanical parts of the instruction of the dogs. The second year they will learn more of the theory behind the mechanical instruction of dogs and will begin to learn the mechanical part of the instruction of the blind. The third year they will have almost finished with their dog work and will start to learn the theory behind the instruction of the blind.

The fourth year is given over to working out very definite problems. Each instructor has his problem, something in the form of a thesis that might be required to show a general knowledge of all that he has been taught the first three years. During these three years many animals are given to him with the secret knowledge that they will never become leaders of the blind. They are faulty types, but the instructor must work with them to learn the types of dogs that can do the work and the types that are not usable, and why.

This period of apprenticeship is a period of difficult and nervous work.

Are not the instructors under a good deal of nervous strain?

They certainly are. At the present time all our apprentices are men. This does not mean that we have never tried women as trainers. We have, but after experience we decided not to employ women trainers any longer. Girls seem to break down under the strain more frequently than boys — even the boys break down occasionally. It is a nerveracking job, not so much in the teaching of the dogs as in the teaching of the sightless to use them. No dog is permitted to start with a blind master until it has worked safely with its instructor, under blindfold, in heavy traffic. This means that the instructor has implicit confidence in his dog. He knows the dog is safe, but when it works with a blind master he does not know what the reactions of the new master will be.

To see a man step off the curb in swift traffic in the early part of the adjustment of the blind, when the trainer knows that the man does not yet react properly to the dog’s signals, and when he also knows that the man should not be touched or spoken to unless safety demands it, is always a nervous shock to the bystander. In this light it can readily be understood that every day the instructor is working under a constant nervous strain, doing things that have to be done, but unable to do them until the last split second, and then controlling himself if the move is not necessary.

Do many dogs fail in their training?

In the period of instruction for the apprentice instructors, we find a good many dogs that we know will not make the grade. We also have some dogs that may surprise us, dogs that looked all right when we purchased them but, as time goes on, show that something has happened in their past life which makes them unsafe as guides. The men must learn to see the difference between the dog that is working its way out of a fault and the dog that seems to be all right but, as we find after a time, has had some experience that unfits it for the work.

If the dog makes good, however, and graduates, goes out with a sightless master, there are very few failures. Years ago the head instructor in one of the German schools made this statement: —

‘If a dog fails in the first six months, it is generally because the blind man and his family have been unable to adjust themselves to a dog in their home and in their lives. If something comes up after six months, that is something that the dog has learned, a fault that has been developed in him. If the dog comes back in the first six months, there is no hesitation in putting him with another blind man; it is not the dog’s fault. After six months, be careful — the dog has learned a fault.’

I would not imply that the dog never learns anything after it leaves the school. It learns every day, but most of the things it learns are useful, or what the blind man calls ‘handy.’ Other things the dog learns are simply comical.

One blind man and his dog lived with his sister and her husband. It was the custom in this family for the brother and sister-in-law to come to the house almost every evening. The four sighted people would play bridge, and if the sightless man was at home he usually rested on the couch during these games and took part in the conversation. The dog remained on the rug beside the couch.

On one particular evening, while the bridge game was going on, the players noticed that the blind man had fallen asleep and was breathing deeply. As they looked, the dog got up, turned toward her master, and cocked her head first on one side and then on the other in a puzzled manner. Finally she left the room and was gone about four minutes.

At last she came back with something in her mouth — it was the blind man’s pyjamas! She walked over and put her paw on his arm and dropped his pyjamas on the couch as much as to say, ‘All my life with you, when you went to sleep, you wore pyjamas, and now you’ve made a mistake!’

Tell us something about your famous dog graduates, about their courage and individuality. What spectacular feats have they performed?

This question strikes me as a little peculiar, for every dog that goes out is famous in his own particular way. The work itself would make him famous, because it is something that dogs have not been thought capable of. If any dog works less well than another, from the public standpoint, it is pretty sure to be a limitation of the human element rather than a limitation of the ability of the animal. To us the outstanding performance of any animal is reflected in the new courage of its master more than in any spectacular feat of the dog.

If I were pressed, I would say that the dog that always comes to my mind is a little gray female. You probably would not think much of her if she passed you on the street without a harness, but she has done wonders, not only for her master, but for his family as well.

Before we consider the dog, we must consider the man she took in hand. He had been a worker all his life, but not a ‘brain worker.’ He came from a Pennsylvania Dutch family in a little Pennsylvania Dutch town where everything was spick-and-span. When we first heard of him he was easy to find. As the blind father of a family he had only one place — in an armchair by the stove. You could find him there anytime, day or night. After a member of our organization had seen him, one of our sightless members with a dog went to pay a call. The wife, in her good Pennsylvania Dutch way, met them at the door and said the man could come in, but the dog would have to stay out; she would n’t tolerate any dog hairs in her house, to mar its spotless cleanliness. It took weeks of work to persuade the man that he could come and get one of our dogs even though his family might not approve. Finally he came to The Seeing Eye, slow, groping, hands out in front of him. There was nothing ahead but darkness. I wish you could have seen that little gray dog take that man in hand. I wish you could have seen day by day the way she put confidence into his heart, took the shuffle out of his feet; the way the hands came down to his sides, the head went up, the shoulders back. I wish you could have seen him a month later when he went home. His wife was n’t sure whether she would let him come in with that hairy creature. Purposely we did not visit him for several months, and when we did he was n’t at home; he was out walking with his dog. A week later we visited him again and he was not at home, and again a week after that.

Finally the visitor went in anyhow and met two grown children of our sightless graduate — a boy and a girl of about twenty and twenty-two. The visitor said, ‘Well, what do you think of your father’s dog?’

They said, ‘Fine! You know, Daddy is just like he used to be when he was a seeing man.’ That’s what the dog had done for them.

Then came the wife, tears in her eyes.

She said, ‘I want to apologize for all the objections I put up against John having a dog. I knew, in a way, that it would free him. I knew, too, that it would probably put dog hairs in the house. But what I did n’t know,’ and here the tears rolled down her face, ‘was that the dog was going to free us all.’

That’s what the dog had done for the family. She had taken the man out of his armchair and away from his daily routine of walking around the same block without ever stepping off the curb — tap, tap, tap — with a cane. She put him into a selling job, where he sold not only the things the community needed, but also the idea of dog guides for the blind. Just one little gray dog.