Colonel Ben H. Swett
Director of Engineering and Standardization
Defense Industrial Supply Center, Philadelphia
14 March 1981

A story of two shepherds

When I was twelve years old, my family lived in the mountains west of Denver, Colorado. I loved the mountains. Whenever I could, I spent a lot of time just tromping around, looking at things.

Once, I watched a shepherd move his flock to higher pasture, as they do every spring. He was walking behind them, prodding the stragglers along with his stick. He had two dogs with him, one on each side of the flock, to keep the sheep together and turn them at his command. The dogs seemed to be enjoying what they were doing, but the sheep did not. They were all packed together, pushing each other. They were afraid of the shepherd and his dogs. I didn't like what I saw, but I didn't know why.

Later, I saw another flock in the mountains, and another shepherd. These sheep were spread out all over a meadow, resting and grazing, while their shepherd cooked his lunch over a small fire. As I approached, his dog said, "Woof!" and came out to meet me. I stopped, and the dog stopped. The shepherd asked me what I was doing, and I told him I was just looking. He said, "Well ... come look." His dog promptly turned around and led me through the flock to the camp.

"You old enough to drink coffee?"

"Sometimes. A little."

He handed me a tin cup half full of coffee so black and hot it made my eyes water. We chatted for awhile about nothing in particular, but all the time we were talking, he was constantly doing something with one or another of his sheep. One old buck came up and half-butted at him, as if to challenge him, but he just laughed, grabbed the buck by the wool on its head and wrestled with it. I could see they were playing. Then he rolled a young sheep on the ground, and spun it around, and in the process turned each of its ears inside out and looked into them - to check for ticks, I suppose. He pulled cockleburs out of the coat of a ewe, rubbed her head, and patted the lamb she was nursing. On and on, all the time we were visiting, this shepherd was also ... visiting? ... with his sheep.

Finally he said, "Well, gotta go. Been nice talking with you." He extinguished his fire, put the cups and coffee-pot away, picked up his pack, and whistled - once. All over the meadow, the sheep stopped what they were doing, raised their heads and looked at him. He said, "So long," to me over his shoulder, and walked away up the hill with his dog beside him. And suddenly, all the sheep jumped up and trotted after him. They converged behind him, into a close but comfortably spaced flock, just as though each of them was tied to him by an invisible rubber band. As I watched them go, I thought to myself, "I have seen something here ... something that is somehow ... important." But I wasn't quite sure what it was I had seen.

Did you look at their faces?

Once, when I was a very young lieutenant, I happened to be walking through our squadron area with a very old colonel - the man must have been all of forty - when we encountered an airman who was not wearing his hat.

The colonel stopped the airman and chewed him out, up one side and down the other, took his name and his sergeant's name, and sent him on his way, while I stood back and admired the colonel's vocabulary.

We went around the corner of a building and met another airman without a hat. But this time the colonel just returned his salute, said "Good morning," and kept right on walking.

By the time we had gone another ten paces, I couldn't stand it anymore, so I said, "Sir, I notice a considerable difference in your treatment of those two airmen."

He stopped, touched my arm so I stopped and turned toward him, and asked, "Did you look at their faces?"


"Did you look at their faces? That first sonofabitch was trying to see how much he could get away with. The second guy forgot his hat."

I ran my memory tapes back, and sure enough, the colonel was right. The first man's expression was sly and arrogant; the second man was obviously embarrassed.


1. Theories

If you take a course in management, the professor will probably talk about "Management Theory X" and "Management Theory Y". Theory X says that employees are basically lazy, not self-motivated, and given to goofing off. It advocates use of the stick. Theory Y says employees want to do well, are self-motivated, and will work at their jobs. It advocates use of the carrot.

Which of these theories do you think is correct? (Response: X = 0, Y = 100%)

Okay, now take another look. Does Theory Y describe all employees? (Not all, most)

So you are saying that Theory Y applies to most people, and Theory X applies to some people? (Yes). Okay, that is what I call "Management Theory Z." I don't know if it is in any books or courses, but I am willing to bet that it is true.

A Theory X manager is a shepherd who drives the sheep. Rules, regulations, reporting requirements, reviews and reprimands are part of this management style. It works, and Theory X followers make it necessary. But it does not apply to the self-motivated: if you hit them, you lose them - or at least you lose their self-motivation.

A Theory Y manager is a shepherd who befriends the sheep. Praise, rewards, commendations, reliance on self-discipline, and few if any restraints are part of this management style. But Theory X (goof-off) followers will take advantage of a Theory Y (nice guy) manager.

So, either of these theories works sometimes, with some people, but neither of them works all the time or with everyone. A Theory Z manager is one who "looks at their faces" to see whether Theory X or Theory Y is appropriate.

2. Leadership and followership

We are all leaders, and we are all followers, in one aspect of our lives or another. Some think of themselves as leaders, and some think of themselves as followers, but in reality we are both. A woman who has no subordinates at work may lead her children and her husband at home, and be a prime mover in her church. A man who directs a vast organization may still obey his doctor, his priest, his wife, his children, and even his cat, his dog, and his goldfish. No matter how much we may fancy ourselves to be a leader, there are those we follow.

Therefore, we have ample opportunity to lead and to follow, to practice the art of leadership and the no less important art of followership. In fact, of the two, followership is probably more important, because the power of any leader ultimately resides in the obedience of his or her followers. People can choose to obey or disobey, and in the final analysis, that is what determines who is and is not their leader.

And here we run into an even deeper consideration: individuals may need Theory X in one area of their life or work and Theory Y in another area. In the same way, they may need a swat at one point in time and a pat on the back at another point in time.

How can you tell? By what they do or don't do? Not really. Remember the two airmen who were not wearing their hats. A Theory X manager would have reprimanded both of them, and a Theory Y manager would have smiled at both of them. A Theory Z manager looked into their faces to see what they were thinking, and applied both theories appropriately.

3. Habits

Both Theory X and Theory Y can be described as habits - and like other habits, they can be changed, but they persist if we don't work on them.

To become or remain a Theory X manager, keep noticing how rotten people are and how often they take advantage of kindness.

To become or remain a Theory Y manager, keep noticing how nice people are and how often they get hurt.

To become or remain a flip-flop, on-again-off-again, basket-case of a manager, jump back and forth from Theory X to Theory Y whenever your assumption that everyone conforms to either of those theories proves untrue.

To become or remain a Theory Z manager, keep on observing your people, one by one, moment by moment, and try to see where their actions are coming from.

4. Mini-max regrets strategy

Suppose you don't know. Suppose you can't tell where someone is coming from. What then? Okay, what did we say about the percentage of Theory X employees and Theory Y employees? (More people are Theory Y employees)

Okay, so when you are likely to make a mistake, which is the best mistake to make? (Go with the odds: assume Theory Y).

Yes, when you're not sure, start with Theory Y. If that assumption is wrong, a Theory X goof-off will let you know soon enough, and if that assumption is right, you have not damaged the self-motivation of a Theory Y follower. This is called a "mini-max regrets" strategy, in which you minimize your maximum regrets by going with the least damaging of your potential mistakes.

A mini-max regrets strategy also applies when someone actually has done something wrong: got another parking ticket, submitted a late trip report, caused a write-up by the inspectors - you know: committed one of the sins of your work-place - and you don't know whether to let him off easy or throw the book at him. Start by assuming it was not intentional; give him the benefit of the doubt, which is one of the greatest gifts anyone can give or receive. Do not even infer that you question his motives. Then, if you were mistaken, he will do something like that again and again, until you have more than enough evidence that he requires Theory X management. And at that point, you can hold him accountable for his past actions as well.

This approach can be described as: "Do justice but prefer mercy." Justice applies to the sinner who is not sorry for what he did and therefore is likely to do it again. Mercy applies to the sinner who is sorry for what he did and therefore is not likely to do it again - and that is the kind of sinner you can't afford to lose. So try mercy first. Then, if he takes advantage of mercy, apply justice.

Organizational considerations

1. Span of control

One person can lead a whole mob, if they will follow, but no one can directly supervise very many people. Centuries of experience have shown the maximum effective span of control is about a dozen subordinates. That is why all large organizations look like a pyramid with several levels or layers of leadership.

2. Chain of command

Most managers are not at the very top or bottom of a large organization; they are somewhere in the middle. That means they need to be very thoroughly aware of the people and mission and problems in their chain of command, at least one and two levels above them, and one and two levels below them.

3. Relationships between peers

The need to be thoroughly aware of the people and mission and problems of one's peers at the same level of organization is all too often overlooked. Relationships between peers tend to fall into one of three categories: destructive competition, constructive competition, or cooperation.

Destructive competition is like a boxing match: managers compete by hindering or damaging each others' groups. The overall organization is damaged.

Constructive competition is like a horse race: managers compete by improving the performance of their own groups. The overall organization benefits.

Co-operation is the essence of teamwork: managers co-operate by doing what they can to support their peers, each helping the other to accomplish his or her mission. The overall organization benefits greatly.

Look for signs of competition among your peers and among your subordinates. Change destructive competition to constructive competition wherever you can. Practice co-operation insofar as possible with your peers, and foster co-operation among your subordinates.

Psychological factors

1. A reasonable assumption

When you are assigned to a supervisory position, study the people you have inherited. It is not likely that any of them will meet your image of what a subordinate should be, but it is reasonable to assume you will find a normal distribution of people. That is, some will be self-motivated; most will respond to what they perceive as rewards and penalties; and some will be impossible to motivate.

2. Go with what you've got

Then remember one of the axioms of the game of poker: play the hand you are dealt. Don't waste time wishing you had better cards - or better people. Study the ones you have. See how they fit, how they might be rearranged to fit better, and what they can do as a group. If the time comes, you may have to decide whether to discard and draw or keep what you have, but in the meantime, do what you can with what you've got.

3. Personality types

Many people tend to assume that an extrovert is a natural leader and an introvert is a natural follower, but that is not necessarily so. An extrovert may be merely an attention-seeker, and an introvert may be well worth listening to.

Some people think they are leaders, and some think they are followers, but both of these views are self-limiting. It is better to be able to say to yourself, "I can either lead or follow, and I can decide to do neither."

Self-confidence and sincerity are considered leadership qualities, but they do not indicate whether the person is a good leader. In some cases self-confidence is justified by achievements; in other cases it is only boasting. Likewise, sincerity can be misleading. Mussolini and Moses, Genghis Khan and Mahatma Gandhi, Adolph Hitler and Jesus Christ all had the courage of their convictions. So it is not a good idea to follow anyone just because they seem self-confident and sincere, and it is not necessary to pump up these characteristics in yourself.

4. Security

There is an emotional security in knowing that someone is in charge. Too much democracy is scary, and anarchy is terrifying for most people. There is more security if the one in charge knows what he or she is doing: few people want to follow a dolt or someone who makes too many mistakes. And there is even more security if the one in charge not only knows what he or she is doing, but also has the followers' best interests at heart.

5. The need for disagreement

Blind obedience is not a virtue unless the leader is omniscient, and that kind of leadership is ... rare. On the other hand, to blindly accept or endorse every action of your subordinates is an abdication of leadership. Therefore, in light of the fact that we are all leaders, and all followers, and none of us is omniscient, we need to be able to disagree with our superiors and our subordinates. This is simply a form of negative feedback, and no system can exist very long without it. In systems of human beings, a continued lack of negative feedback results in the blind leading the silent into a ditch. How effectively you can disagree with a superior, or a peer, or a subordinate, largely depends on that person. How effectively your superiors and peers and subordinates can disagree with you largely depends on you.

Leadership style

1. Orientation

Look at your own loyalties:

If you identify with the people who work for you, you may find it hard to get any work out of them, or to discipline them.

If you are entirely product-oriented, you may drive your people so hard that you damage them and impair the organization.

If all you really care about is the organization, or your place in it, you may expend people without producing a meaningful product.

There is no single answer here. A flexible balance is all I know that works. It is not the static balance of a cement block on a brick, but the dynamic balance of a surf-board rider. At some point, a leader must give priority to the good of his or her people; at another point, first place must go to getting out a product, and sometimes the long range benefit to the organization may require adjustment of both people and product. All I can say is, don't lock into any one of these factors and ignore the other two: I know that does not work.

2. Democratic and autocratic

A participating or democratic leadership style is more flexible, and better able to use the minds of the followers, and thus less likely to charge off into a blind alley, than an autocratic leadership style. However, it is also less efficient and provides less emotional security.

An autocratic leadership style focuses authority and responsibility on the leader. It gets things done, and it is essential in time of emergency. There is a valuable kind of comfort in knowing that someone is in charge - provided that he or she is not leading the whole crew over a cliff.

There is a place for both democratic and autocratic leadership: democratic when the leader needs inputs; autocratic when a decision must be made: "Okay, I heard you. Now this is what we will do."

A purely democratic organization has no leader - the majority rules. A purely autocratic organization has no feedback - the leader rules. The most successful organizations seem to have some of both. This type of balance is an art, and not yet a science. However, some leaders are pretty good at it.

I worked for a squadron commander who held a staff meeting every morning. As a newly assigned officer, I noticed that some mornings everyone seemed free to disagree with him, but other mornings everyone kept their opinions to themselves and said, "Yes, Sir." After a few days, I saw what was happening. If he was not wearing his jacket, he was looking for feedback, but if he kept his jacket on, he had made a decision or was implementing orders. It was a simple and effective form of non-verbal communication to cue his subordinates as to what he expected during each meeting.


Know yourself. Practice a rigorous inner honesty. Study your own strengths and weaknesses dispassionately, as you would study the good and bad points of a dog or horse or piece of equipment. You can fool some people all of the time, and all of the people some of the time, but never ever try to fool yourself.

Know your mission. If you are assigned to a position of leadership, you are expected to accomplish something, so find out what it is. If you find yourself in a leadership role - and that happens any time someone elects to follow you - take a good look at where you are going. You are likely to be displeased with yourself if you lead them over a cliff, and it is difficult to answer someone who says, "I followed you because I thought you were going somewhere." So leadership is a responsibility that requires a workable sense of direction.

Know your people. Respect is the name of the game. Respect all of your people, and try to earn their respect, but don't get too buddy-buddy with any of them. Fight for your people when they are in the right. Do not defend them if they are in the wrong.

Clearly define responsibilities. To be a good follower, you must find out what is expected of you. Conversely, to be a good leader, you must see to it that all your subordinates know what is expected of them. It is best to do this in writing, so you won't forget what you said.


That reminds me of the time a senior officer was looking at the final draft of a regulation I had written. He thought it was too detailed and too carefully worded. He said, "You're a big boy. Just tell `em to charge!"

I said, "Yes, Sir. I'd like to do that. But in this particular subject area, if somebody says, `Charge!', one guy is going to run up a hill with a rifle, another will whip out his BankAmerica Card, and another will plug something into a wall socket."

The senior officer smiled, and then grinned. He signed the approval sheet, and as he handed it back to me, he said, "Okay ... continue to march!" So I did.


Look for anything that draws your people together as a team, and support it. Watch out for anything that turns them against each other, and squelch it.

Praise in public; reprimand in private. Do not flatter or slander. Flattery is false praise; slander is false blame, and both are lies or distortions of the truth, so they are apt to backfire on the one who uses them.

Challenge the self-motivated. Comfort the security-seekers. Boot the lazy. Nail any subordinate who lies to you.

Be slow to fix blame. First, assume the error was unintentional. Take steps to correct it and prevent its recurrence, but do so in a straightforward, businesslike manner. The habit of casting blame can paralyze an organization.

Remember that "no decision" is a decision - to wait, to let things cool off, or to let someone else decide. Deciding to not decide is useful in many cases, and it often is advisable, but if it becomes a habit, it is an abdication of leadership.

Don't go off half-cocked. Snap decisions are typical of silly little people who say, "Look at me; I'm the big cheese around here."

Make sure your people have the resources they need. Much of the work of a leader is outside the organization, interfacing with other organizations.

Do not defend bad ground, either personally, or within the organization, or on behalf of the organization. If there is a discrepancy, admit it and fix it.

Exercise judgment. Sound judgment does not deal with appearances but looks beneath the surface. It is characteristic of wisdom and wise leadership. It can be learned and practiced, so it is not just a gift that some people are born with.


The power of any leader resides in his or her followers. Authority works only where it is recognized. Command depends on obedience. In the final analysis, leadership cannot be enforced. Confidence must be earned and is easily destroyed.

The bottom line in the game of poker is, "Know when to hold `em; know when to fold `em; know when to walk away; know when to run." The same bottom line applies to programs, projects, products, proposals ... and people.

Sound judgment is the difference between a pilot and an aircraft commander. They both know how to fly the plane, but an aircraft commander must also know when to press on with the mission and when to turn back. That difference not only separates the men from the boys, it frequently separates the living from the dead.

Closing exercise: build a leader you would follow

What characteristics do you want to see in your leaders?

(The following is a compilation of responses by the attendees at this seminar.)


Listen to me. Doesn't always have to agree with me, but should consider my opinions.

Fairness. Make sure that the same rules apply to everyone. No preferential treatment, prejudice, or favoritism.

Direction. Be able to point out direction I should go, to guide, teach, help me prioritize my work.

Self-discipline. Should be a lady or a gentleman, provide basis for respect, someone to look up to.

Knowledge. Should know what he or she is doing, provide basis for confidence, don't lead me into dead ends.

Coordination. Get leaders together, so I am not told one thing by one leader and something else by another.

Honesty. Be straightforward with me. Give me a straight answer. Don't beat around the bush.

Openness. Be receptive, approachable. Listen with the heart as well as the head. Care.

Supportive. Have my best interests at heart. Encourage, guide, appraise my aspirations for progress.

Flexibility. Be able to admit you were wrong, to change, to change direction, to compromise. Don't be too rigid - but have some backbone.

Humility. Just the attitude, "I could be mistaken."

Resourcefulness. Be able to provide ideas, insights, access to other resources.

Sense of humor. Not too up-tight. Should be able to see what is and is not serious. But don't laugh at me or make light of my problems.

Keep confidences. Don't gossip or listen to gossip. It hurts everybody.


We have described a paragon of virtue, an ideal that nobody lives up to entirely, but everyone ought to pursue. Remembering that we are each leaders in some aspects of our lives and followers in others, we should realize that, in compiling this list, we have outlined the kind of person we want to be.

Home | Testimony