Great Speeches: Creativity by Richard Hamming

Intro: Creativity, originality, novelty, and such words are regarded as "good things," and we often fail to distinguish between them - indeed we find them hard to define. Surely we do not need three words with exactly the same meaning; hence we should try to differentiate somewhat between them as we try to define them.

Creativity, originality, novelty are words that our society tends to value. I'm going to have to distinguish to some except between the meanings of those words before I go too far.

Primitive tribes don't appreciate originality, creativity, and novelty. You are supposed to do what the elders in the tribe always did. That is not too different in a good many large organizations. Management knows how things should be done and young people who come around wanna do it differently. “I'm not always appreciated.” It's got the same reasons in primitive tribes as in corporations. But it's obvious also that we do value these things and we do expect to have people be creative.

One the purposes of this talk is to increase the chance you'll be a creative person. Long ago friend of mine said he'd like to use a computing machine to do something no one had ever done. I thought a moment and said, “hmm why don't you take a ten-digit number multiplied by another ten-digit number?” It's almost certain that product has never been calculated, and you do subtract the envelope calculation, and decide that I'm right. A random product and two, ten-digit numbers has probably not been calculated. It was new. But it wasn't what he wanted. He said he'd rather do something like find the biggest prime number so he would go in Guinness Book of Records for a little while.

But what's the difference? One thing is we expect that something getting in a book of records is either very hard to do or to remarkable coincidence. Easy things to do don't usually rate as a record. So picking a random product is not exactly hard to do. Finding the biggest prime number is rather hard to do so there is a difference between them. Some difficulty in doing it seems me one of the things we talked about.

Evidently not done before doesn't really count. It's got to have more than that. This is a problem that modern art has grappled with. I assume you're all familiar with some modern art and you've seen it. The art galleries have grappled with it. I can clearly paint a picture and it will be novel and new and different but it won't be great art. Whether what they produce in the art galleries for modern art now is art or not is up to your opinion. It's a difficult question. There used to be a time when our agreed upon cannons of art. But there always been changes and it's very difficult to cope with particularly in art where you know the history. You are well aware that various artists died miserably in poverty. They couldn't sell their pictures and now the pictures will sell for millions of dollars. It's a common phenomenon. It's so common as being discouraging.

A great many artists use this and what is called a Robert Fulton complex. Fulton was trying to build a steamboat, and everyone called it Fulton’s Folley. The argument goes they laughed at Fulton and they were wrong. They're laughing at me, and therefore they're wrong. I'm a great artist. It's a very common argument given. But obviously it's not legitimate reasoning at all. You're going to say if it is going to be something important, it must be widely appreciated.

Let me take new theorem in some branch of mathematics, an important theorem. It maybe only ten people in the world can appreciate that moment. I can't go around saying it's got to be understood by many people. Significant contribution may be made at a time when very few people to understand it. In fact you remember for a while was maintained that only ten people the world who understood relativity. Yet it was regarded as important contribution. So breadth of understanding will not do the trick of telling you one good thing from a bad thing.

In women's fashions I’ve decided after watching Vogue magazine for many years that fashion should be different, but not too different. This is what we often mean it should be different but not too different. I cannot tell you the differences. You see in art alone the great problem of the difference. Now in ballet you will see that ballet still retains some classical ballet and some very modern quite different ballet. One form has grown from another but it's not discarded the old version. In modern art you will find very few people painting in the style of the Renaissance.

In 1838 a guy named Dick wrote a book in which continental drift occurs. In 1905 or something like that Wagner wrote a book on continental drift. These were ignored by the experts until after the Second World War when they can find a mechanism and could literally see the oceans splitting. They can see the magnetic ripples on both sides to convince themselves that in fact the continental drift occurred. They had plenty of other evidence earlier. The biologist has had to make GunWanda land so they can account for why various species on widely remote continents resemble each other. Then some people invoked land bridges. But the land bridges weren't visible so who can believe in those? But it's true that the scientists have not done so much better than have the artists.

Mendel. He did the work. It was published and it was ignored. In 1900, three different people found the same sort of phenomena but connected it with fruit flies and other topics. Then they found out Mendel had done it first. In Mendel's case we generally give Mendel credit for having started genetics even if it didn't but we don't generally give Dick credit for having started the continental drift. You can't tell how it's going to come out. We all have our failings recognizing new things very badly.

Discussing creativity one time a friend of mine said if he took results from three different fields which were well developed and put them together, that would be a great contribution. After thinking over a while, I had agree it would be. I can give you a simple example from my own experience.

At Bell Labs the expert who written a very thick book on magnetics came to me with a problem. One thing another, we worked on it back and forth and finally due to the contributions of my own we got the thing worked out. He wrote the whole thing up and put his name and my name on it and he came down to me to sign because you couldn't get a release without my signature. I went down to a very shrewd physicist friend of mine who I had great respect for and said look I cannot put my name to a paper which my contribution was simply applying least-squares. That's all I really did. He said Hamming my most requested reprints in solid-state physics was for a paper which I simply applied network circuit theory the problem. Sign and let it go ahead. It’s new in their field, they aren't used to it, so just sign and let it go. So I did. I merely applied well-known techniques in one area to a different area and apparently an expert of the field though it was worthwhile. Other people thought it was worthwhile. I still l have my reservations about the matter.

Creativity should be useful to us. But the problem of creativity also involves what I will call the psychological distance. How difficulty is to associate two things together? How difficult was it for me in error correcting codes to say I really want the math likes one plus one is zero? how difficult is it? Once told is very easy. Your creativity is in fact bringing things which are psychologically far apart and bringing them together and saying “look these are related.” This is all we're really doing. Often it is the basis of a great contribution.

I don't know what creativity is. I can't say it I'm convinced I cannot put into words. In fact if you want the whole book is devoted to creativity or as you want style. I became convinced that people are not taught about how to do engineering. There's lots of able engineers out there and they never really do anything great. Why can't they? Why can't they be trained to do great things? So I sell to try and do something about these courses. Reading it in an attempt to get you to think in more creative ways, make bigger contributions than you would if you just go along the same old rut.

If I can't say it why am I writing about it or talking about it, well because you'll find lots of books author and the subject and I'm convinced the bulk of them who write the book never did a creative act. I figured it's better I should try to tell you what creativity is about because you know I’ve done it.

I think frequently creativity is something like sex. You would tell a boy of 14 all about sex. You can give any number of books. He can read all he wants to. But you have a suspicion he isn’t going to understand sex until he's been income involved in himself.  Even then he may still not know it. There are things which you can talk about and read but you have to experience. I think creativity is one of those things you really need experience. The best I've been able to do is to tell you stories about myself so you will be vicariously experience creativity. This is why I've talked so much about myself. The attempt to get you to see and  experience to some extent what creativity is so you've got a better chance of doing it.

Introspection which was frowned on for a long while in psychology but has come back into favor is one of the things you can do in studying creativity. I learned somewhere in the early years of Bell Labs, when I'd done something creative, to stop. Look. Ask myself immediately why. Ask how did it happen.

As in sports you don't keep that in mind while you're doing it. You don't keep that in mind “just how I had baseball bat” while you're batting. You practice. When you're up at bat you just swing. In the same way you're doing creative work you just do it. Then you look back and see. You ask yourself, how could I have done it? how should I have done? but you can't really think about that at the same time you're doing creative. I've only got memory back from what happens. Nevertheless, all the reports say, “it suddenly appeared my mind.”

I recall telling you that the error correcting codes. I couldn't tell you why suddenly thought about a triangle code. frequently you cannot say what made you think of this little thing but there it is. Suddenly you can rationalize it afterwards and say well I should have thought because so and so. But it may not be true. It comes down to the subconscious and by and large I would say all the people who talked about creativity, except a few people think you can manage it and just practice it mechanically, most of us say it comes out of the subconscious somehow or other.

What you know about your subconscious? Not much. The main clue you have is your own dreams. They're clearly what you mean by the subconscious. One thing you notice is that particularly you're young the dreams elements come out of events that happen recently. Not exactly the same, they’re distorted. But many things are somewhat related to episodes from the past couple of days. Once the subconscious works on that, what can you do? A very simple thing which I've advocated to you several times already. You think about the problem you want to solve. And you think about it, and you think about it, and your subconscious hasn’t got anything else to think about when it goes to bed and go sleep. You saturate your subconscious with the problem. You haven't let your subconscious get much else. So while you were happily sleeping it's got to work on your problem. Maybe in the morning, there it is. One of the few ways I know to manage my subconscious is simply saturation.

This is what Newton said “if other people minded it but as I did they would do similar things.” This is really why I say Pasteur’s remark, “luck favors the prepared mind.” The person who's thought about it constantly is more likely to find the answer than the person who has not. The person who's thought about it a lot very intently finds his subconscious delivering the goods later on. If subconscious couldn't do anything else. That's one way about you. Of course, you have to eat sleep and so on but you keep your mind more or less intently involved on one thing. You keep the high emotional content of the day on that problem. You don't let other things get in the way. Then you can manage sometimes to come up with it.

If you can't, and the persistence doesn't work after a reasonable length of time, and I can't tell you just what reasonable is at the moment, you set it aside and go do something else. Sometimes left alone, the subconscious will still deliver the answer weeks later when you haven't been thinking about it. Sometimes you have to bring a problem again and go at it. But these mono-maniacal pursuit of an idea, intently week after week, after week, seldom results.

Now a classic example of the dream business helping you out is a chemist. He dreamed about snakes biting their own tails or rolling in hoops down the hill. He thought about a little while and so yes that's really the carbon atoms in a cycle. That's where the idea that carbon atoms form a ring, came from Kelly's dream of the snakes biting their tail and rolling down a hill. He saw a connection between his problem of what he dreamed of. This is the characteristic of Dreams. They don't give it to you right square out. Sometimes you have to see what the dream is telling you.

I want to delve into the importance of emotional content. If you don't have emotional content it doesn't happen. I dwell on it I was talking about error crafting codes. I've said other times. It’s partly because I have a friend at Bell Labs who was a very very good mathematician. Very skillful. Very able. But he would go home play the guitar. He did very good work but not the very best. Yet he could have I think. But he was never emotionally involved heavily the way some of us were. By and large those who care greatly are more likely to do something than those who don't care greatly. Emotional involvement is one of the things necessary. Its the same as I told you, keeping your mind saturated with it, those are different ways of saying the same thing.

Sometimes you get an answer. You're sure it's the answer and you work a little further you find this no good at all. All right go back to the drawing board. Except you now know what you don't have, can't do. You know that that path doesn't work. You know more of things not to do so you have a more focused view. Another thing you can do, which I often do when I'm stuck, I ask myself what would an answer look like if I had it. Are there conservation of energy laws? is there conservation momentum? What would an answer look like if I had it? This also helps you greatly to solve the problem.

Another thing you can do is ask, have I got all the information? what does it really depend upon? must I really know the position of the earth when I'm trying to calculate the satellites of Jupiter, or do I not need to know that? and what effect is position the earth got on the timing of satellite Jupiter's?  Robert Dicke found the velocity of light that way. You have to ask all kinds of questions.

A false starts and false solutions are common. Find you have an answer. You've found an answer in your own quaint way that is peculiar to you. You find you have trouble saying it. You also find the explanation doesn’t make much sense to other people. So you have to rework the idea for the other people. For example, Maxwell's equations. Maxwell originally derived them using all kinds of gear trains and everything else. A very wild mechanical model. That's how he found them. But he stripped out all the nonsense and gave you the equations directly without the mechanical model behind it. And that's often necessary to clean up your idea. Remove the nonsense. Remove the unimportant parts. In the process you will often understand the problem much better yourself by removing the irrelevant material.

For example, you assume the function is continuous and you look at the proof and you say, “where and the proof did I use continuity?” well I didn't. well then why do you assume continuity? I don't have to. that kind of a thing happens continually as you strip down the idea once you got a solution. and you gradually come up with a clean solution and you publish it finally.

Probably the most useful thing in creativity is analogy. This is like that. Maxwell in his head was thinking about a bunch of gear trains when he thought about how things might work. He finally decided gear trains were irrelevant but that's how I got there. you may think the analogy has to be close but frequently the analogy is very loose and weak. it's just a suggestion. “oh that reminds me of how it works over there.” “yes like that but of course it's different this way or like that.” “that's like that but you know that corresponds to that.” I see what to do next.”

It happens that way frequently, a very loose analogy like the snake biting his tail. loose analogy like that will often get you there. very good ones, not too often, but it happens. If you had an exact analogy you wouldn’t have something new. you’d be saying the same old thing again. There's always bound to be some difference but analogy is one of the things we work on. so how can we do this?

I worked with John Tukey for many years. You heard his name many times now. He was clearly a genius. He used to infuriate me frequently. We'd be doing something and he say, “well you know about the polarized light?” “well yeah I know about polarized light.” but I didn't occur to me to think of the situation where our own radar, that that would be relevant. When he says it, I see it. Why didn't I think of it?

I've spent a lot of years on it. Why was he so able to do it and I wasn't? Partly because he was smarter I was. Partly he had done something I had not done. Up until then I had learned things in the framework in which I had learned them in. I saw that when he learned something in the act of learning he asked himself, “what other things does it look like? In other words, he put little hooks on the idea so it had many ways of being recalled. I can only recall the framework in which I learned it. You see the difference? If in the act of learning, you simply learn in the framework which you're told, you can only recover in that framework. But if you think it over yourself and turnover and associate with it many other things in your mind, then when the time comes to recall it there are many more hooks. You can pull up the fish the other guy misses. This is what John did. Now I started doing that.

When I learned something new I said, let me stop and think and that's why I told you about digital filters. I told you how I went at it rather differently I said what is basic going on? what are fundamentals? how is this like that? I told you the story how walking out of my office and saying oh I know how to kill Gibbs phenomena down such a couple stories - that's nothing else than the fracture. Go a little further, that's the fraction electron microscope because that's where I'm headed for.

I have gradually learned to do these kinds of things. That's why I say several times that what you learn from others, you learn to follow. That which you learn for yourself, you learn to lead. You have to decide when you learn things to think them over. In the act of learning I look at in many different ways and then you'll have this lovely ability recall a relevant thing to the situation. You have a sense as I was with early days with John Tukey when he told me I knew but until he told me I didn't. And as I say that infuriated me. I didn't like that. Once or twice I know why, but when this happens ten times a year you do get rather infuriated being so stupid.

Many books were written on a topic of creativity. There are whole talks, seminars, everything else. There was maybe 15 20 years ago the idea of “business and brainstorming.” You assemble a bunch of people and they all brainstorm ideas without regard to details, just think wonderful thoughts and it was supposed to produce great ideas. We all have had the experience of talking with a friend or two about something, batting and idea around, and out of that comes some insight. But it appears these scheduled creativity sessions of Barnes brainstorming did not work very well because they've been pretty well abandoned. They just don't work, not scheduled. On the other hand talking, your ideas over to friends is help. I find what i have an idea i often go down the hall and explain to my friend. In the act of explaining I remove some of my confusion and get it in a cleaner neater form. Sometimes if you pick your friends well, they say, oh yeah that's just like so-and-so and then you've got another clue what to do.

I've always searched wherever I was to find those people who would stimulate me to think better, who are the people to whom i told something they would say something back to instead of “yeah that's interesting isn't it?” doesn't help you a bit. what you want is a person who's goes “yes that reminds me of so-and-so or that sort of like so” or “I remember that happened back in 08.” you seek those people out. you'll find those are people you go to lunch with. you avoid the other people. you pick your friends so they will stimulate you to think. you don't pick them necessary for a polite nice people. you pick them for ability assimilate. several people I found most stimulating were as people awkward difficult people to get along with. but if they will stimulate two good ideas they're beautiful. if they don't they're not.

Back to this topic mainly: can we teach creativity? I think the answer is both yes and no. I would not have devoted a lecture I would have not devoted the whole course to it if I did not think I can do something about it. but it's the analogy I've used many times, if you want to run the for minute mile you have to do the work. I have to get you to change yourself. I have to get you to change yourself. Physical habits you're familiar with. you put your left shoe on before the right shoe. But you have mental habits which are just about as rigid. You habitually think this way or that way. I have somehow or other to get you to break down these things and think in new ways, to accept new ideas. To search for them. To prepare yourself to find them. In short my job is to change you and how you think. The only way I've known is the method I've used. I tell you stories that you more or less have to believe because I tell them about myself. If I told you about somebody else they would not have the effect. I told you this the first lecture I'm telling you again. The necessity of lecture on direct experience is overwhelming if you're going to try and teach art.

A person teaching painting can show how they’d paint and they’d remark, “well da Vinci's painting with so-and-so” but when the artists said well I'd painted this way why don't you put something in like this, it's more convincing you see it done in front of your face. I've done this. I've tried very hard to get underneath your skin and I'm going to work on it further before we're through, trying to get you to change your ways.

In applying a change, you have to know yourself. It's very easy to promise reformation. I promise reformation to myself many times. Lots of times I failed. You know how difficult is. Anybody who has gone on a diet knows how difficult it is to stick to it, or quitting smoking, or something else. It's not easy to change your habits. Nevertheless, people do. People do succeed in going on diet and keeping our weight down. People do quit smoking. People do change things. You can change your mental habits the same way you can change your physical habits, provided you set down and start working on a project. But it isn't going to be easy.

It's better to practice on small steps than assume impossible ones and become discouraged. Frequently people who start dieting want to lose 10 pounds the first 5 minutes. Of course they don't. They get discouraged. Same way you have to pick small things to try and change small things and gradually build up a different pattern of thinking, a different way of living your metal life, and then you can be more creative than you have been.

In order to do this I have to really get down to something I've said it once or twice in this course. You have to take charge of yourself. You’re responsible for yourself. You're responsible for the way you behave. You're responsible for the way you think. You're responsible for yourself. You have to change yourself. You have to make yourself into the person you want to be. One of the things I've watched in my own life a lot:  do I want to be that kind of a person? no. then don't behave the way.

For example do I want to be a liar? no. then don't tell lies. if you don't want to be a liar then don’t start telling lies. When you find yourself telling lie, say hey I was not going to be a liar that's not right. I got to stop it. You have to stop a lot of things and start a bunch of other ones. You are responsible for yourself not the thing. I've got to get you to realize that more than you have. You are the one who must change yourself. I can't do anything about it.

I have not talked about a delicate topic. That is, dropping a problem. If you pick up problems and work on them and you get success, that is fine. You work one up and then another. But if you can't drop problems, the first time you meet a bad problem, that's the end. You'll never get it solved. The classic example is our boy Einstein. He had lots of good ideas. He produced a lot of things. But once he hits unified field theory, the second half of his life went down the drain and that problem he didn't get anything. It was the wrong problem, at the wrong time, in the wrong way. Not that it will not perhaps be unified field theory, I don't know but half his life went on a problem which he got nothing for it. Perhaps most creative fizzes we ever had in many respects

I have often argued that Oppenheimer at the Institute should have called Einstein into his office and said, “Al, old boy, do me a favor. Drop that unified field stuff for six months a year and work on anything else at all, but just drop that. Would Al have been able to do it? I'm inclined to think he could have, but who thinks they got enough nerve to tell Einstein how to run his own business. Oppenheimer might have but he didn’t, as far as I know. I sometimes thought Einstein could have been saved.

On several occasions at Bell Labs I've been asked to do that kind of a job, and one of them which is a failure, I can remember distinctly. This man thought well he was working on nonlinear filters, nonlinear filter is everything, with no restriction on this area. It's like looking at the moon and a falling Apple and saying “oh, gravity, common thing.” That's nonlinear filter ha you aren't going to build one.

He had the characteristic syndrome. He was going to get the results in 18 months. He'd made some progress for a while. But he thought he might get there fine at 18 months. A year passes he's made some more progress, he will better get there in another 18 months. That's a very characteristic syndrome.

You've got an idea, you're working on it, but you're the solution all always receding. We see it at about a fixed distance. It's not a guarantee but it's a suggestion you have the wrong problem. If you don't drop it there goes the rest your career. On the other hand if you drop problems too soon somebody else comes by and doesn't it. It all comes down to one of my favorite expressions: the difference between strong-willed and stubborn. George Washington was strong-willed. Had a lost the revolution, he would have been stubborn. It's a small, thin thing but it's essential that you understand when to drop a problem.

I think I've told you some cases the other day, I thought of how many times I have walked for a problem that looked too hard to do at the moment, and I should go off and do something else. Later on other people did it. Fine I'm glad they did. I couldn't because the working circumstance which I had were inadequate to the problem. The working circumstances is very common.

You may think you haven't got the right working conditions. Very common thing for creativity. Forget it. What the average person thinks is good for creativity is not. The evidence I will give you is the Institute for Advanced Study. There you're given a lifetime appointment, more than adequate salary, beautiful restaurants, and beautiful offices, everything else. Everything is laid out for you. If you compare what the people at the Institute did before they got there and what they did afterwards I think you'll be with me. The Institute has killed more good people than anyplace else has ever created. What you think of as ideal working conditions, are not. Frequently, very unpleasant work conditions are the ones that will stimulate you so don't give an excuse, “well I can't work until I have ideal working conditions.” when you get deal working conditions you probably will not find the answer. it's done under difficult circumstances.

One of the methods I've managed myself with and I think it works very well on me, I'm not saying it will work on you, is what I call the cornered rat theory. It's based on two principles which you know about me: I'm rather egotistical; I don't like being wrong. At the same time I got a fair amount of self-confidence.

Thursday comes and I’ll say, “well I'll have the answer for you Monday.” Sunday afternoon comes by and Monday morning we’re meeting soon, and I haven't got an answer. I started thinking about it well Monday night I still haven't got the answer. I thought really because I don't want to walk in Monday morning and say I haven't got the answer. Like a cornered rat more often than not, I've risen to heights of effort and produced an answer. I would not have done it if I had not promised “I'll have it to you by Monday.” I managed to convert what I consider a couple of character defects in many respects of me into an asset. I used myself. I deliberately would promise people answers why I didn't know how I was going to do it because experience taught me that was the way to be much more creative.  “I'll see if I can get you an answer.” oh I promise “I'll see you get you an answer” now maybe I get it by Wednesday, maybe I'll forget it by then. Something else well I promise I have Monday morning and Monday morning is a good time do it because you got the weekend to start working on it. With Monday morning you have very good time to promise an answer I'll be in your office in o'clock with the answer. it powerfully makes you think Saturday Sunday night on Saturday night if you've got those attributes I have that's one way of managing yourself. If you haven't then you have to find some other ones but you have to know yourself to manage yourself.

I'm coming back to this expression I said, “you have to take responsibility for yourself. You have to understand yourself and what you can and cannot do.”

I should also tell you the circumstance of changing things. When I went out to Nebraska to graduate school, I was going out where I knew no one. In those days you took a train because airplanes cost so much. On the way out I thought about myself. I thought I had had a quick smart answer for a lot of things. I was very clever quick-witted guy. When you do those things 9 times out of 10 it’s funny. But the 10th time it's only funny but it hurts the other guy's feelings a lot. I think you know it. These people have a quick answer. Sometimes the answer is really painful but it's funny so you have to laugh but it isn't good. I thought more and I realized I had lots of acquaintances but no really close friends. With a long trip out there by train through Chicago and Lincoln Nebraska I had plenty of time to think. I decided, you know it's because you had this snotty quick answer that you've gotten yourself in trouble, you've lost friends. Sooner or later you're going to say something which is going to hurt their feelings and they'll gradually drift off.

I decided I'll reform. Going to a place where no one knew me, no one turned to being expect me to make a quick remark. In among friends you expect to behave the same, when you move to a new situation you have a better chance of changing your behavior pattern then you have in a fixed static situation. Since you people get moved around a lot you have more opportunities than most of us have for deciding, “I am going to change some aspect of myself.” You have built-in opportunity to do it which I had far fewer than you did. After all 30 years of Bell Labs in 20 years here isn't exactly a lot changing around much it's pretty well fixed in a rut. Nevertheless you can do this and I say that when you change situations that is one of the best times you can change yourself because you no longer are automatic expect to behave the way you have.

You can behave differently and that includes not only physical material things but also ways of thinking. If you get a reputation for having answers in a new situation then people will turn to you. At Bell Labs I had a reputation for solving analytic integration of all kinds. The whole labs. When they got really stuck they would call me up and say I got an integrator, I hear you can do something. I would say yes and he tells me over the phone and I'll copy down the information and get the phone number and say I'll see what I can do. I got an enormous practice of that. Now you have a machine to it. But those days we didn't.

There were people around Bell Labs, people you knew. For example I knew two people who really knew Bessel functions very well. When I got stuck on an idea

they could tell me all about Bessel functions. I have other people who knew various things as I stayed in the organization. I knew to whom to go. They build reputation. How do you build a reputation for giving you answers a given kind? I think on the back end I had a reputation for the following: when nobody knows what to do at all, then you call Hamming. He is probably best when he nobody knows anything. We don't want to dwell where we're totally lost. Hamming  somehow or other can do something in those situations where other people are stuck. I think that's the reputation I had. Certainly that’s the way I probably worked best. You would have other traits. What are your best traits? Emphasize those and get a reputation for and bingo there you are. You're known as a person who does these kind of things. What kind of things do you wanna be known to be able to do, things that you can do.

There's no use to want to do some things you cannot do. I mean if you've one arm you can't become a great juggler. If you’ve got some other physical defects there are things you cannot do. Likewise mentally if you get some peculiar features. If you've got features I told you about mine I told you confessing it I got ingot ISM is pretty big and I got great self-confidence. I turn around and use those things for my advantage. You need to study yourself. Take charge yourself and find out what the heck is going on and go do it.

I have to say some things about careers. In mathematics, theoretical physics, and astrophysics, in the past, the best work was done by the person very young. Newton after all got much of stuff during the plague when he was sent home from Cambridge. He went back to where he father was and he spent time at a farm thinking. For 18 months as the plague went around. Most of the ideas are traceable directly from that period. Most great scientists their best work was done surprisingly young.

On the other hand, with music composition, politics, and novel writing, often the last is best. We value the last compositions of many of the great composers. That doesn't mean the composers did. The composers sometimes like some early composition that other people don't like. They like it because it was to them something of a breakthrough. People don't see it that way so in some fields maturity is the best thing but it's astrophysics mathematics and theoretical physics where raw creativity counts youth is a great advantage and experience is not. Other fields different. I don't know about your field.

There are a few exceptions. ?? Charles in mathematics taught high school for years and years years and he finally got to college became highly creative later on. There are one or two examples in mathematics where creativity was late in life, but by and large it's very young. If you want to go in a field like mathematics and you're 40, forget it. You're not going to do much.

What did I do in my career? Let’s look at a couple of stories.

After I came here Bell Labs I rolled sheet of paper like that with ten columns. Great research went across, and the years went in the down column. I unrolled it and looked at it. Boy was I good. I was connected with a lot of things. I been out of any transistors but there I was helping the bastards. I mean really good. Because it was rolled up tight, I scotch taped it on my office door so I can read it look at it. I come in and look at, admired it. I came in one day and looked. Everything the historians marked down that I did was important or connected with important things was in the first 15 years. There wasn't associated with a single important thing in the last 50 years . I couldn’t say I was reasonably closely associated with anything important. Needless to say, I tore it off the door and through the waste basket.

That's how historians judged my work. I knew this age business. What can you do?  You can do what I did and what ballet dancers do and aging athletes. When you no longer do it because of age, you become a coach. That's what I did. I came out here teaching which is really coaching. I can't do creative work anymore in my field. I'm too old. I can try and coach you into it. This is exactly what I'm doing. I'm trying to convince you you can do it. My error cracking codes occurred when I was 31 but my excuse is that I was working my way through college and the war tore up anything I might have done and got me involved in other things.

I got a little late start 31 is kind of late for a math teacher to do his best work. But probably error correcting codes are regarded my best work and they were don then. All the things the historians value were in the first fifteen or thirty years. I know more now. I got more skill technique than I ever had. That doesn't count. Creativity is something else. Originality is something else. I don't know how it comes out painting. I think in poetry often the best poetry is fairly young. TS Eliot made a great big splash when he was in his early 20s and 30s. Most posts get started very young. Others will get started late. Sometimes the age is a help sometimes is not. Your business depends on what aspect you're in in your business. You need to look the situation to say it over take charge of yourself and decide, “I am going to be great in these ways, I can do what lies within my ability, no I can't now suddenly at 40 become a great mathematician (almost surely).” yes I can do such and such.”

One other thing you can do which is worth doing because you've got one lousy life to live on earth as far as I know and if you do come back through reincarnation you won't remember what you to the first time anyhow so you've got one life to live here, what are you going to do with it? You might as well be creative.


Our lecture today is on the subject of experts. There's one definition of an expert: it is a person with a briefcase at least 50 miles away from home. The one I gave is an expert is one who knows everything about nothing, whereas the generalist knows nothing about everything. The expert tends to be so narrow they know everything about that a little small subject and nothing else, the generalist doesn't know anything.

I have been both an expert and a generalist. I can tell you the expert wins against the generalist almost always by the following devices. one you use a lot of jargon which the generalist doesn't know. secondly you invoke basic principles in your field which may be totally irrelevant but sound good. You snow the generalist and you lead them astray. Most times in an argument the generalist loses by those two methods. The specialist does not come down to the generalist level but rather stays as high-level. That leaves the generalist losing. It's a problem you will face. You people are by and large supposed to be generalists so we face a different problem.

A fellow named Kuhn wrote a famous book “scientific revolutions”  and he looked at the structure of science and the revolutions and he gave the name “paradigm” to name a pattern of what is going on. Typically when you are taught physics, you are taught not only the formulas but you taught a style of thinking about it. The style is not mentioned. It's just delivered. The kind of problems you can ask. The kind of answers you get, are all implied in this style. People by and large operate within the paradigm of the field. Suddenly there may be upsets. In physics there were two of them, relativity and quantum mechanics. Both can provide a different framework of thinking and different kinds of questions. Relativity opened up the whole field of cosmology, the origin of the universe. It really opened up the field now so there's lots of speculation. Dfficulty with cosmology is you have one sample only and you're supposed to account for how it happened. You haven't got a bunch of different samples and you haven't any power to experiment. So cosmology is a very interesting science if you think it's a science.

The contradictions will arise the field. In late eighteen hundreds there were numerous contradictions. I told you in discussing quantum mechanics how some of these led to something else. Most people in the field will ignore contradictions. They will dismiss them. They will do anything at all but face them. As they go on they don't do anything. It's only by noticing their contradictions and building up that you have a chance of making the big change. It's a very difficult thing to pay attention to.

What doesn't agree we'll accept the doctrine is because you are not popular if you bring up “yes but”. For example I brought up to you with regard to thinking. That although we talk all the time about the neural system neurons storing knowledge here there and yawn. One celled animals can apparently learn, and they don't have a nervous system. By and large that is totally ignored. It may or may not be relevant but I keep it in mind saying well you know maybe what I'm being told is not the complete story. Most people in the field ignore the fact and go on thinking within their framework, that the nervous system explains everything.

When a change occurs, it is resisted by almost everybody in the business. I can find no figures reliably or how much relativity and how much quantum mechanics was resisted. I can tell you that in late 1930s I saw in library in University of Nebraska quite a few books trying to claim that Euclid was a true geometry and all the other things were all wrong. Consequently, relativity was wrong. There was a lot of relativity books written against relativity. It indicates quite a few people did not accept relativity. If you go back to our boy (Max) Planck about adopting quantum mechanics there's a classic sentence of his, “we didn't convert them, we outlived them.”

I thought about that many times. The bitterness must have been in his voice, “we didn't convert them, we outlived them.” that's how we won. by and large entrenched people would not pay attention to new ideas. this is normal process. new ideas are greatly resisted.

It was supposed by (Thomas Samuel) Kuhn that new ideas oughta triumph. well yeah they oughta but I told you back in 1838 (Thomas) Dick wrote about what amounts to continental drift. In the early 1900s (Alfred) Wegener whole book about it but it got nowhere. It was adopted in the 40s well after the war or perhaps early 50s. Learned physicist wrote against why it couldn't possibly happen. of course, they assumed the wrong model. based a wrong model they proved continents couldn't drift.

The other one I mentioned to you is genetics, Mendel’s Peas. he might as well have not done it. But it was rediscovered in 1900 and then people found out he had done it earlier. it is not clear that a new idea will triumph. I cannot possibly tell you how many new ideas were lost and didn't triumph because there's no way of finding them. but my suspicion is that the idea that people have that all of the truth triumphs in science may be true if the element is long enough but it may be well past your lifetime. so the idea that we would like to have that we really went out early is simply not correct. it's a very great resistance when you have a new idea.

beyond just the continental drift because South America fitted Africa there was the fact that biologists had found the same kind of remnants of animals in the rocks in Australia in South America and Africa. IT’s natural to suppose they must have been connected together because the animals were in one place. to account for some of these things there was a belief that land bridges came up and sank down so the animals get across and they sank again, but there is no evidence. and there was the theory biologists had of one unified land Pangaea breaking up in the Gondwanaland and so on, other geologists wanted no part of it. until they finally had seen right with our eyes practically the split in the continents where the new land is being made right above the oceans. and then suddenly it was accepted. when you read down history, they say oh we always believed it, we just didn't have the final evidence. they were very very resistant to the idea. it's one of the best ones in my lifetime of total resistance to an idea which apparently now is fairly triumphant. most of us people believe something like plate tectonics is the way the planet is built. but it could be changed tomorrow I don't know.

I have one of my favorite one, a guy went to the Patent Office and applied for a patent it would lift water more than 33 feet. now you read in your physics book that vacuum will lift water 33 feet, no more. they wouldn't give a patent. because all the books so you couldn't. so he brought some equipment in and put it on top of the roof. a little valve here, a little valve up there, and a short-stroke pitched piston. the best is going so fast the standing waves are set up in the column. when is a rarefaction water comes in. well as a compression the valve shuts. and when is the compression the water goes out the top. he lifted two water 100 feet. they had to give a patent. but they didn't believe it because all the book said you can't lift water more than 33 feet because the air pressure outside is so much it's only going to push up so far. but you see he saw genius ways of producing perfect rarefaction and compressions and proceeded to lift water a hundred feet instead. I'm not saying it's a good method. I'm saying it's typical that the Patent Office knew that the books all said you can't lift water. but you see everything like that is based upon something. no possibility proof rests on one statement. it rests on a whole bunch. if any one of them is wrong…. and they never visualize. they thought if I simply tried to suck the water up yes I could only lift about 33 feet to perfect vacuum won’t lift it any further. what if I produce local vacuums. then I can. so you see one of the troubles with the expert, you're proposing to do something he knows can't be done. but you may be doing a different way and he cannot hear or she cannot hear. it's a very very great troublesome thing.

I said the geologists claimed everything is right. and well they had to revise themselves. as a well-known saying, “if an expert tells you something can be done it is probable it can be done. he tells it can't be done, it may pay to get another expert who may tell you you can.” certainly my experience of Bell Laboratories had many experts that have been marvelously wrong lots of times. they don't understand the problem. I had a young fellow working for me for a while, a bright energetic nice guy, but he didn't understand. he grabbed the wrong problem, solved it very elegantly and as a result his work had to be undone before he can get to the right problem.

The misidentification will problem is very great. the expert sees some parts of it. that’s that and this is this…they forced the situation into a situation they know or think they know, and then look at it that way from their trained eyes, and they don't see the problem has got some elements which are different. the expert sipping cannot see it. that's the same words I told you earlier this lecture that they can't remember there are small contradictions in any theory. they conveniently forget them.

well this guy was very nice but I found frequently he was a nuisance. what he said was correct. but he had the wrong problem. he had the right answer to the wrong problem. and trying to find a reasonably good answer to the right problem is something very difficult to do when the guy is given the exact answer to the wrong problem. it's very hard to undo that. but that's one of the things they do.

now Kuhn and historians of science have concentrated on the big changes of science. it's my impression that the smaller changes science work the same. there are many small changes of current science and they don't get adopted frequently. for example one that I didn't succeed with. working at Bell Telephone laboratories it was natural that I would meet the frequency approach. you remember in several mechanisms you analyzed by frequencies. you imagine a transmission and you're worried about the frequency bandwidth. I told you earlier I had tried to do what I observed a half by Max Planck. I tried to use the right formula so my calculation would fit with their beliefs in their field not just approximately polynomials which is traditional. so I gradually developed the technique of approximating not by polynomials but by frequencies. sines and cosines are complex exponential's. and you know how powerful Fourier series really is since you're more or less electrical engineers.

my friends in computing kidded me about it. they never listened. they never adopted it. they just never did. in fact I wrote a book which is reasonably well expounded but most American Alice's books now will mention the fast Fourier transform. that’s all others will say about the fast frequency approach. yet by calculating that way on several occasions because the calculation returns a frequency the person for whom I produced the numbers could understand the numbers better they could if it were polynomial. if I said well I passed 70 frequencies if I use a polymer degree 5 all over b5 wouldn’t mean anything to him. “these were the frequencies we passed through” that he could understand. frequently not always so I led some people to small things not as great as quantum mechanics but to small help, by trying to adjust the computation who fit the person's beliefs.

there's other ones I also occasionally used. real Exponential's for some things. In some other fields., sometimes, I talk in the person over lunch I found out the kind of functions they believed in. I try to use those kinds of functions to help them find insight. I didn't succeed. there was an episode where I think to this day that I'm right but that idea didn’t yet not percolate into computing. although it is widely used by physicists.

I'm not bringing up these troubles just to poke fun. but for four reasons why I bring up the role of expert. first as you go on you'll have to deal with experts you ought to know the faults and good parts of experts. they know a lot. secondly many of you will become an expert. I'm hoping somehow or other that you won't be as bad as the average expert. there appears to me the rate of progress is increasing and will continue increasing your period therefore will be more need for you to adjust to new things and more off the experts will be wrong because the situation isn't like it was yesterday. fourth if only I could say the right things to you I would make you stay ahead and not let you become obsolete. that is a very sensitive point for me. I have had several my friends good friends left behind because they didn't adopt new things. I told you roughly about a friend of mine who was a great analog person. I learned a great deal from him but he wouldn't really convert to digital and he was left behind. we retired the same time him by encouraged retirement little extra money and me to go out to a different job. later on when we met it was clear that our attitudes toward our life careers at Bell Labs were quite different. mine was pleasant. his was unpleasant because he was sort of pushed out. since coming here I met quite a few captains in Navy who retired. they didn't make Admiral. someone reasonably happy what happened to him but some show good deal being disgruntled and unhappy. I think you one who when you get him drunk enough he's always back commanded a flotilla off South America. he didn't like being passed over several times retired. my problem is how do I get you people to rise to as far as you wish to rather being pushed out.

Thought Series: Creativity is about combinations


Thought Series provides actionable ideas and anchors for reflection on your life or your work.


You don’t have a climb a mountain and wait for an idea no one has ever thought about. When you learn about something new, you combine them into something new. The newness is your uniqueness. It is the sum of your values, beliefs, ideals, motivations and traditions that guide your ability to learn and even your identity.

Angry birds. Genius.

What ideas are you playing with?