Oral-History:Arno Penzias

From ETHW

About Arno Penzias

Arno Allan Penzias

Arno Penzias was born in 1933 in Munich, and received a B.S. from City College of New York in 1954, a Ph.D. from Columbia in 1962. Although his primary work has always been in research and management at Bell Labs, he has held appointments at Princeton, Harvard College Observatory, and Stony Brook. He was hired by Bell in 1961, became a radiophysics department head in 1972, and was named director of the Radio Research Laboratory in 1976.  With coworker Robert Wilson, he won the 1978 Nobel Prize in Physics for the discovery of the background radiation of the universe. At the time of this interview in 1980, Penzias was Executive Director of Bell Labs.

Penzias focuses on social issues such as the reciprocal obligations of scientists, commerce, and society. He discusses government regulation, the social effects of technology, and issues of privacy related to new technologies. Another significant theme is affirmative action and Bell Labs' efforts to recruit diverse researchers. The interview touches on Penzias' flexible management of creative talent, the issue of women in science, and the question of racial discrimination in the workplace.

About the Interview

ARNO PENZIAS: An Interview Conducted by Carol Lof and an Unidentified Man, IEEE History Center, Spring 1980

Interview # 037 for the IEEE History Center, The Institute of Electrical and Electronics Engineers, Inc.


Copyright Statement

This manuscript is being made available for research purposes only. All literary rights in the manuscript, including the right to publish, are reserved to the IEEE History Center. No part of the manuscript may be quoted for publication without the written permission of the Director of IEEE History Center.

Request for permission to quote for publication should be addressed to the IEEE History Center Oral History Program, IEEE History Center, 445 Hoes Lane, Piscataway, NJ 08854 USA or ieee-history@ieee.org. It should include identification of the specific passages to be quoted, anticipated use of the passages, and identification of the user.

It is recommended that this oral history be cited as follows:

Arno Penzias, an oral history conducted in 1980 by Carol Lof, IEEE History Center, Piscataway, NJ, USA.

Interview

Interview: Arno Penzias

Interviewer: Carol Lof and an Unidentified Man

Place: Bell Telephone Laboratories

Date: Spring 1980

Personnel Issues at Bell Laboratories

Affirmative Action

Penzias:

I was talking about an affirmative action program I run. We're making a lot of progress, but there are some real problems.

Lof:

With affirmative action?

Penzias:

Not with affirmative action, with everything. People problems with our company. Basically you have a problem in organization, of typecasting people, somehow, when you have a staff organization. We can go to technical for awhile. Suppose you want somebody and they say, "Get to the affirmative action organization," for awhile. Now, after some period of time, that person is going to be promoted to supervisor in finance or public relations or something else, right? Or whatever. Because we have this business of this terrible number of levels. We cut jobs and define each job in terms of some level. Then as you move along, as you stay in a company, you have to keep getting promoted. We try to get out of that in the technical area here. That's one of the things that we try to do very hard. You don't have to get promoted in order to succeed. You can develop without this hangover from grade school.

Lof:

Promotion usually means you've got to supervise people.

Penzias:

Yes. We don't have at Bell Labs the opposite system, which is this business of giving somebody some kind of title. You know, dual ladders and all these things. All that ultimately is a label on their salary. They don't really do anything because it turns out somehow that these things are viewed as non-promotions because they don't cost anybody anything. We can go down the hall and call everybody a "telephone hero" if we wanted. Or we could call them a "senior scientist." Or "senior distinguished scientist." All that costs is typewriter ribbons. Basically, this kind of thing is like government currency with no backing.

Lof:

Well, it's like the magazine that I work for. It's run by the IEEE, and all the people that contribute to the magazine are volunteers. So they get titles, you know. And then they get promoted. They work twenty hours a week on the magazine, and they get a bigger title. Then we get "senior distinguished editor-in-chief."

Penzias:

Ahhh. I like that. It's like the lodges, you know, with the "imperial." But anyway, in our affirmative action program, we have a situation where it's one of many things. So people want to work in that. We get some really good people. We work on the program. If the person is good — and they usually are — well, that's only two years along they're punching their ticket. They work on that, and even if they're concerned with it and they really are, after a couple of years, well, they have to be promoted. They usually want to be a department head after that or whatever. People are constantly moving through all the jobs. Inside a company maybe that's okay. I run a fellowship program, but I run it on the technical side. We have a staff person also. We have about thirty minority students who are out getting Ph.D.'s at places like Stanford, MIT, and so forth, and doing very well. But we have the staff person who calls them up, and gets talking to them, and then finds out their problems. It's not just handing them money. What we want to do is provide the kind of support that other people sort of have automatically. If you start out from a poor family, and you're starting out —

Lof:

Are you really getting these from poor families, or are they coming from the middle class?

Penzias:

Some of them are really poor. On the other hand, it is not an income redistribution program. I mean we don't look at their income. But we are not out there just to get numbers, and we're really looking for the very best and the brightest. It turns out that the system really does work. I mean America really works. That is, when you look and you look for the best students, you find that a lot of them really do come from poor families. So we don't select on either basis. Even for people who are fairly affluent — occasionally we do have some students who are fairly affluent — the chances of them having an uncle who is a dean at Harvard is very small.

Lof:

Absolutely.

Opportunities for Women

Penzias:


Audio File
MP3 Audio
(037 - penzias - clip 1.mp3)


In fact this is why we separate that program from the women's program. We also have a women's fellowship program. The chances of the woman having an uncle as a dean at Harvard is just as good as for any white student. The family attitudes are different, but the kinds of support that the people need are different. Whereas in this case, students may do things. So they need a certain amount of support, not just money. We have a mentor concept. Many of our staff members start in the beginning after the students graduate and have a summer where they establish a relationship with them. Over the years they work with them. The trouble is that the professional on the staff side of the house changes about every two years because they're on their way to getting promoted. That's a problem I was just on the phone about for the last twenty minutes.

Lof:

That's one of the things that they say about women. That one of the reasons they've been held down is that they've never sought out mentors. That very few women have mentors and many men do.

Penzias:

Yes, yes. Sure. That's exactly right. This whole concept, the whole business of mentoring. In the New York City Police Department they refer to a mentor as a "rabbi," even though they're almost all Irish. In fact, that's one of the things about Jews in the New York City Police Department. I still remember when they announced promotions at a Communion breakfast. I was a Jewish New York teenager, and even then this was in the days before anybody worried that there should be such a thing as equal rights or whatever. It did strike me as a little funny. But I do remember the Daily News talking about the Communion breakfast, so on and so forth. The New York City Police Department was announcing their promotions. Commissioner So-and-so, and it was covered by a reporter, and it was a Communion breakfast.

Lof:

I like that.

Penzias:

Everybody had their own city department in the old days. It wasn't all that terrible. But it does mean that after the groups of people had sort of divided up the city, there wasn't anything left for anybody else. There certainly wasn't anything really left for the women except certain kinds of jobs.

Lof:

Women could work for the telephone company.

Penzias:

Yes, they could be telephone operators.

Man:

You needed a rabbi in the telephone company, too.

Penzias:

The mentoring, right. The business of having somebody who can provide the kind of support, the kind of access to good jobs.

Lof:

Somebody who's been there and knows the system and can clue you in.

Penzias:

Yes, somebody who knows the system and clue you in and fit you in and keep you with it. That's something women certainly don't have to the extent that men do.

Lof:

I don't think they have it today either.

Penzias:

They don't.

Lof:

They talk so much about networking and stuff like that. But it's sort of the halt leading the blind.

Penzias:

I get into terrible arguments on this. There was a young woman on a plane I was talking to about this. She was going to start an organization to career counsel, and somehow counsel women executives, women in business, and she'd never been in a company.

Lof:

Isn't that incredible!

Penzias:

She had gotten this academic job. She'd read all these books. Now she was getting her Ph.D., and she was going to open up a company. It depressed me that she might even succeed. If women are themselves so devoid of resources, they might be taken in by somebody like that.

Lof:

Well, I recently went to a little meeting. It was women in publishing. What they had were a bunch of editors from fashion magazines who were up there counseling this whole group. It was the silliest thing I've ever been to because all these little starry-eyed kids were raising their hands and saying, "Oh, I want to be in fashion magazines", you know, and "I'm interested in shoes and stuff." It was just ridiculous. This isn't where the jobs are. You don't walk into Glamour Magazine and start there. There was absolutely nothing realistic. You start at the IEEE sitting at a desk, copy editing Transactions. And you learn how to —

Penzias:

Author, should this be a Greek rho? [Laughter] Hopefully there's some daylight on the other side of a job like that.

Lof:

Well, they get the tools. Then they go someplace. But some women still look for glamour jobs. They don't pay; they're closed.

Penzias:

Some of them are very dead-end, yes. But we're trying to make jobs here somewhat more attractive.

Lof:

Are women going into engineering and physics in enough numbers?

Penzias:

No, not in enough numbers. We certainly have more than we used to have. It certainly is better than it was. I think we still don't find the numbers in either minorities or women that we ought to.

Lof:

Why?

Penzias:

I can see a lot of reasons. There is still the eighth grade syndrome. By the time you're in eighth grade, the boys had better be smarter than you if you expect to have a date for the prom. If you think that has stopped, it's not true. I would say it stops with some people, but it doesn't stop with others. I look at my own suburban high school. I have two daughters and they have not both gone the same way on that. But there is this kind of different tracking that somehow society still seems to put on large numbers of what are at that point still girls, who will ultimately be young women. That's one problem. That is an early problem and we are still typecasting. We are still typecasting in the society. Furthermore, the role model problem is still there. We also have certainly not come to terms with the concept of marriage. We certainly act in industry, in government, every place else, as if marriage didn't exist.

Lof:

Exactly. Yes.

Turnover at Bell

Penzias:

This is this business of promotion, right? The companies themselves are able to make allowances for individuals, and we make some allowances, but really not very much. Somehow we have a standard — we have this concept of motion. Bell Laboratories is a unique place. I mean I have had only one job in my entire life. In fact, I've only been in one organization for my entire life. I never even took a lateral transfer to some other division. I spent the first eleven years in a department as a scientist. Somebody wanted to promote me to department head, and I said, "No, I'm very happy here." Then my boss retired. Then his boss retired, and his boss retired. Each time I got the job. That staying is different from the standard

Lof:

But this is a rather unique place.

Penzias:

Well, some of it's not a unique place. If I talk to my opposite numbers who are executive directors, they laugh. They've lived in ten places over the same period.

Man:

I think yours is the exception.

Penzias:

Yes, mine was unique.

Lof:

Bell Labs is unique. The people who come here either leave immediately or stay a very long time.

Penzias:

A lot of them do. But even so, you look at it from the outside. You know, all Bell Labs isn't alike. Research has rather less motion to it than the others. Because the science is so complicated and the telephone business is so complicated, we have branch laboratories.

Funding of Bell Labs After Corporate Separation

Lof:

What's going to happen to Bell Labs with the new legislation that's going to separate the lines from the subsidiary things, which would be Bell Labs? How are Bell Labs going to be paid for if you can't get it from the revenue on the toll calls?

Penzias:

I don't know. Well, we get revenue from everybody, right? There was a concept which was set up years ago — a basic notion. It was set up maybe with less formal or structured decision-making than you might think. That is, now I actually understand that some people have done economic modeling, and it turns out to be exactly the right answer. The argument is as follows: The telephone customer is the ultimate beneficiary of advances in communications science. It pays for the telephone customer to just fund that research directly. When that research just supports all kinds of telecommunications improvements, the products the customer gets have a much wider variety, and are cheaper and better and more effective. That has been essentially the license contract method for funding. One might have to put more boundaries around it, but I could imagine that principle could continue. I'm not going to say it's going to. I'm not a politician. I don't know.

But in Bell Laboratories Communications research, our function is to put stuff into the open literature. Ultimately much of our work finds its way to Western Electric products, just as it does in the hand-held calculators that aren't made by Western Electric. Just as the results of our work go into Comsat-owned satellites. Ultimately these are things which get into communications products. We invented much of the fundamental work in digital modulation. We certainly invented microwave radio. A lot of the digital communications in the country is going on at Collins digital radio. Collins didn't invent microwave radio; we did. But the telephone customers had better look because they have digital radio. Okay, it's nice wherever they get it. So there are competitive entities out, and Collins Radio could be one of our competitive subsidiaries if the world were a little different. Well, it isn't. We don't own Collins; they're on their own. But they still use the results of our work, just as Comsat uses it.

Lof:

Do you see some sort of continuing funding?

Penzias:

I didn't say that. I can't predict the future. I can say that is an option society really ought to look at. That is, to continue the idea that Bell Labs research ought to be funded by the ultimate customer.

Man:

That's the entire nation.

Penzias:

The Bell Telephone customer. Not based on taxes. That would be running it by the government, and it would be terrible. But the communications user. If you use communications today, you are drawing from a fund of basic knowledge. Every time you make a toll call or you get just plain telephone service to your house, which is from one of the operating companies that puts something back in. It doesn't really cost you a lot. The people our in California cut our research support by nine million dollars. So they save the rate-payers of the state this huge amount of money, nine million dollars. On the other hand, a few months later they came in with a $300 million rate increase. Well, what would that rate increase have been like if there hadn't been fifty years of telephone research before it? How much would they have had to go in when the whole state was full of telephone operators, when they were still using open-wire or analog carrier or what have you? What kind of a rate increase would they have gotten if all these — and it would still have had to be women; I don't know who else would do it — all these women got a rate increase to match inflation? Do you think it would just have been $309 million?

Lof:

No, of course not.

Fluorocarbons, Ozone and Cleaning Electronics

Penzias:

Basically, that's the point, which is: if we can support research and have that go out to not just our competitive subsidiaries, but everybody, then everybody is better off. One of the other things people don't understand is that a lot of the research — or some of it, anyway — doesn't even have a product associated with it to pay for it. In other words, there are things we do which don't appear in a product. There's a problem even if everybody said, "Well, when you buy a switchboard, you should pay for all the research behind that switchboard." But when you have a switchboard, you've got to clean the contacts. No matter how electronic it gets, everything has contacts. There's only one way in the world that I know how to clean contacts, and that's using fluorocarbons. So the telephone customer as an individual has a tremendous interest in deciding how much freon ought to be put in the atmosphere. We have a radio astronomy facility at Bell Laboratories which is measuring atmospheric chlorine monoxide. Because the thing about freon is not the fluorine in it, but the chlorine that's bad. If you put chlorine in the atmosphere, it takes ozone, which has three oxygens in it. It takes a chlorine atom, and grabs one of those three oxygens, and makes CO1. Now you've got oxygen instead of ozone, and the ultraviolet is supposed to come right through it and do bad things. Nobody's even sure if it happens or not. It's one of these hysterical things: Somebody said the ozone is going to go away. The ozone didn't go away, but they still gave the guy a medal because he called attention to the problem.

Lof:

But is the research going on also to try and find some other way to clean these contacts?

Penzias:

Of course!

Lof:

And there's no breakthrough?

Penzias:

Surprise, right? What have you done for us since the transistor? [Laughter] There are people who are working on other cleaning agents, sure. Telephone customers can't, say, "We're paying $1.89 a spray can for this stuff. When somebody comes up with something instead of freon, we'll pay $5.89." Whoever's in the freon business won't say, "All right; we're in the freon business; we'll go off into some other business." There's no manufacturer making a switch contact cleaner that's perfect for the phone company. They'll make one that has carbon tet or something else in it which works ninety percent as well, get most of the market, and the heck with it. It costs only a few bucks now because we do it here in somebody's spare time. It's so little money it's nothing. But if we're talking about switch reliability in the Bell System, we're talking hundreds of millions of dollars a year. Instead of having these two guys monitoring chlorine monoxide, we could just make a law and say, no more freon. That, I think, would be expensive and probably stupid. Let's monitor it and see, if in fact there is a problem, how bad is the problem? How much can society use? Then put it on those uses which are the most reasonable ones. But you have to find out how bad a problem it is first.

Lof:

You're talking about the same things that have been banned from the public market, that were in hair sprays and...?

Penzias:

Oh, of course. Yes. You don't need them for hair sprays. But you also want to know how bad the problem is.

Lof:

I was under the impression that they were outlawed across the board.

Penzias:

It comes in your refrigerator, right? To me it's getting rid of it in spray cans. They were putting huge amounts of it in. The question is, if you're going to use this stuff for the next century, ought there to be a crash program to invent something to replace fluorocarbons? Well, you don't know unless you know how much chlorine monoxide is up there. Is the chlorine monoxide even a function of how much ozone there is? The ozone changes by factors of two with sunspot cycles. Somebody has to monitor that stuff. As part of our radio astronomy, we have a unique capability for doing that kind of atmospheric stuff. It's a small back-burner project, but it's one of the things, because we have an integrated research facility. We don't have to go try to find some fluorocarbon manufacturer and pay for that research — we'd spend much more time trying to find somebody who's willing to pay for it. The one fluorocarbon manufacturer can't afford to pay for it because somebody else can make fluorocarbons, too. Not unless you get all the fluorocarbon manufacturers together in some motel room somewhere, get the Department of Justice to agree, and then they all pay for it. Then the Japanese could make it, right? Or somebody else. You just can't do it.

I'm giving that as a kind of a very far-out example. I'm mostly in the electronics business. But basically there are these kinds of economies, which, unless you were setting up the Bell System totally for ideological reasons, that somebody ought to take into account. What we really want to do is consider the possibility that we ought to just continue to fund research, put the results in the open literature the way we do now, and patent it. The patent royalties then come back to the basic organization and help defray the costs of the telephone customer for some of the stuff we're doing. Because we license to other people, and we let the operating companies who are paying us —

Telephones and Regulation

Lof:

Don't you worry that this is not what's going to happen now?

Penzias:

Sure, I worry. I worry about lots of things. I worry that somebody can just decide that the telephone company is too big and say, "After all, it's too big, and the telephone has worked so well all these years, so that must mean it's an easy job, right?" It looks so easy. Harry Houdini became famous largely, I think, because he had this guy with the sledgehammer standing near the tank. The concept was that he was risking his life, and somebody was counting out that in one minute and forty-three seconds he would drown. If it gets to one minute and forty-two seconds, the guy's supposed to smash the tank, let the water out, and save his life. The fact that it might go wrong, was part of the show. The difference between the performer the audience appreciates and the person who bores the audience is the thought that it's somehow difficult. The lion-tamer. If all the lions did their things like pussycats instead of roaring once in a while, it would make the difference. That's why you get excited, and watch the lion- tamer. If this guy could make the cats do it perfectly, you would be bored. One of the troubles with the telephone is it's boring.

Lof:

Yes!

Penzias:

You pick it up, and it always works. Right? When a blackout goes out — we have blackouts in New York, right?

Lof:

I was part of it.

Penzias:

People congratulate the radio stations for still being on the air because they called up their place in New Jersey and stayed on the air. I remember those things. People said, "We were still on the air during the whole blackout."

Lof:

It didn't do me any good. Mine won't plug into the wall.

Penzias:

Okay, you have a little radio. You take your little transistor, and people had their transistors in their car radios, and WOR was still on the air.

Lof:

Right.

Penzias:

But the telephone doesn't use electricity. There's no electricity in it because it works in blackouts. The telephone didn't do anything. The blackout goes away, and it still works. It's boring. It always works. Therefore you can do anything you want to it. It's just there. You don't have to pay any attention to it. So this is one of the problems we have. We make it work too well. One of the few times people appreciated us was when we had this terrible fire in New York.

Lof:

I was part of that, too.

Penzias:

One of the buildings burned down, right? People sued us. Then we found out what telephone service was worth. It was several hundred million dollars. Florists were suing us for a million dollars because they didn't have telephones for a few months. Three or four hundred thousand dollars for one guy! More than his annual income. Then we found what telephone service was worth and we started getting a really good press because there we were, in this burned building, it was black and just full of junk. We had to put it up and we put it back. Then people suddenly realized, "Hey, there's a telephone service. You have to do something to it." Most of the time you can't kill it. Since you can't kill it, that means you're free to legislate, you're free to do whatever you want.

Lof:

Well, they are legislating.

Penzias:

Sure, and there are really good reasons to legislate. That is that communications has become far more pervasive than it has in the past. Since now the telephone has become so handy that it's connected to everything else, these boundaries have to be somehow defined by legislation. People demand more from the telephone. Before that they had themselves, and they had secretaries. They had human beings, and it was just a question of everybody using this really neat thing which fits between your ear and your mouth and you talk over. There was an easy boundary; it was between the human beings and the network. So it was an easy place to define it.

Well, now people want the pet machines to talk to each other. Now that it's possible, if you have a computer of your own, to fill out even more complicated forms, the government wants that more complicated form. You have to keep all these extra records. Everybody wants more things. You have the machines, so now people have found work for these machines, some useful and some not-so-useful. You do all that, and now these machines have to talk together. Well, you can't now decide to grow the telephone company into all these things, because some of these things are really part of the telephone and some aren't. At some point you have to decide where to put those boundaries. You have to legislate, but you also have to regulate. The government has a legitimate role in deciding when there's any utility and how that utility ought to best serve the public. It isn't up to us, after all. We've always been a regulated business.

Long Line Competition

Lof:

Yes. You're going to cease to be so regulated and go into competition.

Penzias:

Some portion of the business.

Lof:

The Long Lines.

Penzias:

Yes. Some portions of the business will now go into a competitive posture. I don't know which ones will, but some of them very well may.

Lof:

There are a lot of different theories as to how exactly it's going to be done. I spent some time in Washington talking to different people, and each one had a different theory. Each one believed it. A completely different idea, and they were absolutely positive it was correct.

Penzias:

After all, the telephone is here, and you can't kill it, you can't hurt it, and you can't build another one. Nobody has the idea of building a new telephone company or anything. But at least you can rebuild it.

Lof:

That's not true!

Penzias:

Well, a new telephone network.

Lof:

Somebody has the idea of completely separating all of the Long Lines from the —

Penzias:

But that's not building a new one. That's rearranging. Building means starting from scratch and providing communications service to all the people in the country. It's already been done. Now that it's there, you can't put your name on that. What you do can't change it around. It's like the ancient Egyptians of the pharaohs. When some of the pharaohs came around, they started building new pyramids. What finally brought the Egyptians down was that one of these pharaohs got the idea of saying, "It's a lot easier to put my name on some of the old temples than to build myself some new ones. In fact, all the best spots have been taken." Since it's a lot harder to provide some new service for all the people in the country, let's leave the service and give it another name. I'm not saying everybody, but that's a tempting thing to do. That's what you run into a lot of people, each one with a different idea. That may be what they're doing. They're taking an existing thing, and by putting their own stamp on it, starting over.

The point is, though, I think people ought to realize two things: One is that telephone service as it exists is a very valuable service which people ought to want to keep. Second, that for the quality of life to improve or hold steady, we have to do a lot with the limited resources of our society, and one of those resources is our own brains. One of the things we've got that most of the other countries in the world don't have is a damned good communications system. Maybe we can afford to have a little more expensive labor, and maybe more hours off. Maybe we can even afford to live in the suburbs instead of having to be packed in units. We can afford some of these things which maybe we might find luxuries. We can afford some of these things because we have other advantages: economies of scale, a very efficient agriculture. One of the other things we've got is a damned good communications system, and a growing one, and a very highly sophisticated one with a lot of support — like the people that work here — behind it. If we perceive that as one of our strengths and advantages, it may be that people will continue to understand that that's a way they can keep treading water and keep from ending up with their expectations diminishing.

Social Issues and Technology

Employment

Lof:

What new developments do you see in the communications field? What's the scenario for the year 2000, as you would see it?

Penzias:

I think it really depends a lot on what people want. That is, I don't think we are totally technologically driven. There are people who say we can do almost anything, and I'm not sure that's true. But what people might want to say is that we can do anything at a price, or we can do anything given enough effort, and so forth. This is barring something which would impede progress, but assuming that society continues to take advantage of the possibilities. There are ideological things. People get carried away, but we're not suicidal as a country. We have a lot of basic strength. When the shoe pinches and things start getting bad, for instance, we stop using electricity so much. We may find that after we get enamored about clever little Japanese telephone sets that look prettier, basic telephone service is really a neat thing that we don't want to lose.

Given that kind of background, I think one of the kinds of things we want to see is machines. We're probably going to have to find a hard time not making much more use of machines than we have already. I don't mean things with rotating arms and cams. I mean some sort of computers and computer processing in many more facets of our life, hopefully to replace those parts of human effort people find demeaning, uncomfortable, unnecessary. I would like to look forward to the day when the job of secretary doesn't mean that you automatically have some woman working for the executive. Right now there's a labor pool which fills the job, which I would imagine in a longer period of time, for instance, ought to be taken up in a kind of a natural progression. I see that from machines.

Lof:

We've been doing this for the past fifty years. We've removed the un-nice jobs.

Penzias:

We've done very little of it. We've done some.

Lof:

The electric dishwashers replace the black boy that used to go and wash the dishes. Now what is he doing? He's unemployed.

Penzias:

If you want to live in a society where you can say that we can't do anything with women, women are incapable of doing anything but being secretaries.

Lof:

We're not saying that.

Penzias:

But we're almost saying that by saying that that person is unemployed. The hard-core unemployed in the country, since the War on Poverty started, has gone way up, I understand. I'm not sure it's true. But I don't think that's a response to additional technology, as much as to social things. I don't think you can blame General Motors for the South Bronx. We have done certain things —

Urban Renewal

Lof:

I read an article that blamed the South Bronx on the telephone.

Penzias:

Really?

Lof:

They said that in the 1940s and 1950s, not everybody had a telephone, and everybody got their messages down at the corner drugstore. The people were outside. When air-conditioning, television and the telephone came, people moved off the curbstones.

Penzias:

What actually happened is that some people defined slums in certain ways, and they found a lot of people sitting down at the curb. So the federal bulldozer went in, knocked down all those buildings, and put in high-rise projects. When I was a kid in the Bronx, the safest place to stand waiting for a bus was in front of a bar.

Lof:

Sure, where there are people.

Penzias:

There are people, and there are a lot of men. If a little girl were to scream, you know, people would investigate. But in a high-rise apartment building — and these are the ones that were built — you get the poor people out of sight by stacking them vertically. Now, of course, any brownstones that the city hasn't knocked down are terribly valuable.

Lof:

Incredibly valuable, yes.

Technological Determinism vs. Social Choices

Penzias:

But basically they were knocking those things down as fast as they could, putting up high-rise apartments for the same people, and then they found it was socially destructive. We have things like waiting for some number of years to get an apartment in a housing project unless you have a fire. If you have a fire in your apartment, then you get to the head of the line plus you get up to $3,000 for new furniture and clothes. So again, you can't blame the telephone company for that.

Lof:

No. But what you're saying is —

Penzias:

The technology. I'm saying that the social change, the things that we have done or don't do, are not really to be blamed on technology. I think society has made a lot of choices. We have, for instance, the idea of eligibility and sufficiency in welfare. The idea that you're only eligible for welfare if there's no father in the house. Secondly, that anytime you earn a dollar, you lose one whole dollar of your welfare, and you can't get even. I think the damage we have done, the fact that people are still unemployed, the fact that we haven't made use of the talent which has been liberated from menial and meaningless work by technology, is not the fault of technology. That's a clear distinction that has to be made. The fact that technology provided napalm and the helicopter gunship does not mean that technology was what brought us into Vietnam.

Lof:

Then you're saying it's the people that are using or abusing it.

Penzias:

Technology has given us opportunities. We no longer have somebody yelling, "Boy!" Whatever that means. There were copy boys, and in those days they weren't even black. They were white, but they were young. They'd be at the newspaper where someone would just stick up his hand with a piece of paper and say, "Boy!" I remember there was a friend of a friend who actually one day got so sick of it, that he took this thing, threw it on the floor, said, "Me Tarzan," and walked out. He was a college graduate who'd gotten the job after World War II. He was just sick of the whole thing, of being called "boy," until he finally could get to a typewriter. He just got sick and walked out and did something else. Surely you can't blame technology on inventing the cotton gin and then saying, "Look at all those happy people in the fields."

Lof:

You can't blame technology on anything. Technology is inert. It's not human.

Penzias:

But technology has a momentum of its own. Technology does have dynamics, and I don't want to get around that. The fact that we put a ringer on the telephone does force people to answer it. They stop conversations. If the telephone rang right now, it would take an awful lot for me not to answer it. Technology, the existence of a telephone ringer, does impinge on our conversation. So I don't want to go that far.

Lof:

But so would it if somebody were to knock on the door.

Penzias:

Yes, but the technology is different. The technology is different because it provides different kinds of constraints. That is, the guy has to face me in a very different way from that. Or if I have an automatic alarm clock. We can do all kinds of things, technologically, which do impinge on the way we do other things. You have cars for one year so you can have better cars the next year. The existence of automobiles means you have highways and you plow under potato fields. You have driveways, and then you have crossing guards. The technology does build. I'm not arguing that it doesn't. But I am not prepared to say that the evils of society can be traced back to technology. This is probably the generation which has to some extent begun to pay the price for the racial inequities of the past. We are beginning to look at the sexism that goes back into society's history as far as —

Lof:

Forever.

Penzias:

We're doing some of it in this generation. As I said, we are making some progress. We shouldn't say that we ought to stop progress because look at what we did for the blacks; we got rid of all the menial jobs. You could make that argument to say, "Well, I should stop working on word processing because you're going to throw those poor women out of work, too." Just because the blacks are unhappy now, you're going to make the women unhappy because you ought to keep them as secretaries? I don't buy that. I think that the whole human race has a better future than that. Or at least has the capacity. And I think the technology can help provide it. Furthermore, once you get out of utopian things, there are lots of people out there who need more of the things that make life worth living. There may be very few people who starve to death, but there's plenty of malnourished people. The only way we can get to a just society is with a bigger pie. We are not going to take the pie of present size and cut it up more equally. We've never done that. We need a bigger pie. The only way to get a bigger pie is to use our brains. Now that the Arabs have found out how much we need their oil, we can't expect them to go back and be reasonable.

Lof:

True.

Penzias:

Therefore the only thing left is our own intelligence. You can't program for that. You can't just say, "Go and sit in a corner and invent." There has to be more focus. One of the things we have found that seems to make a lot of sense is to be able to bring people together in more effective ways to substitute for some of the things we do routinely. For things like travel, for things like record-keeping. So those sorts of things have to be done in a more economical way. Economical doesn't mean saving money. But it also means that the filing clerk, the porter, we allow the chambermaid at the Holiday Inn, to go do something else.

Record Keeping, Privacy, and “Big Brother”

Lof:

Let's talk for a minute about record keeping.

Penzias:

Yes.

Lof:

Our records are all computerized, right?

Penzias:

Almost none are totally computerized.

Lof:

They're getting to be more and more computerized. How are we going to protect ourselves against the possible evils of having our whole lives on a disk someplace?

Penzias:


Audio File
MP3 Audio
(037 - penzias - clip 2.mp3)


Because of our tremendous reputation in hardware, people may know less about the fact that we are also a very big user of computer science. We've got a lot of computer people. Besides all the computer science people in Division 12, here in communications sciences we've got a lot of software people as well. These aisles are full of people who are working on the problems of record-keeping. Well, look at this office. There are more pieces of paper in this room than any human being can read in a lifetime. The room is left unlocked, and someone can sneak in here at night and look at it. The only thing that protects the office person, presumably, is the size of the mess. The only thing that protects him is the size of the mess. Somewhere underneath all of these things is a scrap of paper which reveals the real reason why he went to dinner on April 17, 1977, and it wasn't what he put down on his income tax. If all the records were put in one place in computer-readable form, then you would, in fact, have somebody else who could go in and say, "Check that. Check the problem."

Now, I call that the liquor store problem. Let's say that all of a sudden I became a guy working on one of these missile-launcher things. I'm one of the three people with keys to blow up the world. I have a Master Charge card, and I go in, and I'm buying a lot of liquor. People could then say, because it's national security, "anytime one of the people with these three keys buys a lot of liquor, we want to find out about it." Maybe I won't get promoted to major because I'm a secret drinker. What I really was doing was bringing some stuff home for my daughter's wedding. This takes us back to this problem we ought to talk about anyway, which is, "What is the year 2000 going to be like?" The question we have to ask ourselves is, also, "Why are we doing all this?" The only reason for doing all this is to improve the quality of life. First, you want to improve somehow, the quality of life.

There are two things, if we just extrapolate. I know you want to interrupt me, but let me finish this thought. There are two things I want you to understand. One of the reasons you feel a little bit threatened about this whole thing is because the extrapolation is that somehow we are going to get to very big machines. There are two things wrong with that. The first is that that will not improve the quality of life. The second thing is that it's probably impossible. The thing I see in the year 2000 is far more individual control over machines than I see now. Let's say the government decides now Bell Labs is too valuable to lose, so they give us all GS ratings. If you're in the government, you can decide that the economic thing is to build one big computer for New Jersey, and one database. Then we can have your driver's license on the same thing as your bill. We can, in fact, take your driver's license away as soon as you leave the liquor store, before you even take that drink. After all, this is saving human life, so we ought to do it. What if we make a mistake once in a while? Well, we're not going to go that route, I hope.

If we go the other way, which is people getting what they want, then we will have a far more dispersed set of machines. Machines which talk to each other the way we talk to each other now. If we have a way for these things to talk to each other, the way we have people talking to each other, then we can more than compensate for the economies of scale with the conveniences of privacy. Plus we can do the other things. We can compensate for the convenience of privacy, with the advantages of what I might call customer programming. I want to use a technical word to say something which I mean in a much more human way. Everybody decides what kind of machine they want, what kind of services they want, and they put them on their own computer. That means privacy is far more maintained. Then the liquor store problem can be tractable. If I'm willing to put money in my wallet, I have to worry about losing it, and I have to worry about counting it. I'm doing those things for myself. Well, the electronic equivalent of a wallet is the electronic equivalent of cash. I'm willing, under certain circumstances, to go without receipts just for some kind of convenience. But also maybe for privacy. Maybe I don't want my wife to know how much her birthday present cost. If everything goes through a Master Charge Card, and the jeweler's bill comes in, and she signs them, she sees how much it is, You begin to have a problem if there's only a Master Charge card.

That need for the small dispersed computers is a real one; in a marketplace, I think people who can provide dispersed processing can win. But you can also win on economic grounds if you do it right, provided only that there is a network to make them talk to each other. They have to be able to talk to each other because otherwise it won't be there. They have to be able to talk to each other, and they have to be so tractable that in fact they can do the secretarial kinds of things that we do today. The secretary still does an enormous amount of anticipation and translation, handles a lot of details like register add-up, and so forth. It's like my first thing on the IBM 650: Take this number out of memory. Put it in this register. Take another number. Put it over here. And take these both into the add register. Operate the add register. Take the sum. Put that over here. Now, of course you've got four translations just for A+B, and it does that. We have some of these things already in research. Look at the telephone to the left. On that one, if you push a few buttons and you work it out, and you want to use one of our automatic programming jobs, then you can almost do it in English now.

Lof:

You can?

Penzias:

Yes. When I have somebody on the phone and the phone rings again, and I push this button, I want him to go on hold. I want to talk to the second party, and then I want to push the button again. When I push it the next time, I want the other one to come on and not lose the first one. In other words, this is programmed conferencing on your phone in the way you want it. We have much more product differentiation. We don't have to anticipate. In this building, there are no empty offices. We're full. All of these buildings are full of programmers. You know about Bell Labs, you read about these additions — we're building another building and so forth — it's all computer science. We're growing in these software areas. But we can't do everything for you. Where would American food be today if you still had these hooks, you know? Remember the long pole with the thing? Because you don't even know what you want. About half the time that clerk stands there while the customer's saying, "Maybe I want to get some oatmeal...." You're going down the aisle, and you say, "Oh, hair spray, of course!" I remember last night we were out of it. Or my daughter was complaining about whatever. I got this stuff when I got to the store. The customer doesn't even know what to ask for. The customer has to do it themselves.

In that same way, what we want to provide are machines that the customer operates and teaches. Then we get more individuality. Once you have the right kind of machine, you take over some of these record-keeping functions. If you have a record-keeping capability, then the need for someone else to maintain those records isn't as pressing. Well, now there's an alternative, and it allows the possibility of individual freedom and privacy which was not there before.

Lof:

Okay, but the American government is getting every day more and more information on every person in this country. Is that going to reverse?

Penzias:

It could reverse if people could keep their own records. If they want to. Right now there's no way of doing it, you see.

Man:

If they want to, I think, is a big key, too.

Penzias:

The point is when we get into the privacy situation. I don't think I can speak for people twenty years from now. Right now people in California want divorces. A few years ago they wanted safe contraception. Then they wanted meaningful lives. But the next generation may decide, "Damn it! We want privacy!" They may. We can't extrapolate from today. But it may be that some point in the future people will want to say, "I don't want to live in a country where the government knows absolutely everything, and therefore we're going to stop that." But then in order to do that, you'd better have a technological alternative.

Lof:

Okay. But you're counting on men of goodwill.

Penzias:

No!

Lof:

And women of goodwill.

Penzias:

No. Not at all.

Lof:

Government of goodwill.

Penzias:

No. There are many societies in the world which are far more rudimentary than ours, as far as technology is concerned, where they're enormously more oppressive than ours.

Lof:

Oh, our oppression is very benign, I grant you.

Penzias:

No. The Nazis managed with far less technology than we did. They were able to check the ancestry of tens of thousands of converted Jews back three generations. Even if they had to do it with pencil and paper, they found them and sent them to the gas chambers. So I don't think it's necessary. I say there is a capability. I can't give you a different government. Presently the digital computer has been invented, but it doesn't matter. Once paper and pencil and the ledger have been invented — and even more than that is available now — the government has the capability of keeping records on everybody. They do.

Lof:

They do.

Penzias:

One of the things I would like to see, as an individual, not as an executive director of Bell Laboratories is a technological alternative. I think the technological alternative is to distribute record-keeping to the point when if you go to the liquor store, you would be able to go back to paying cash with your Master Charge card, and I don't think that's impossible. Right now, if the government wanted to, they could have the guy record the serial number of every bill, and they still can trace you. If everybody had to write down every serial number and send in all the serial numbers, you couldn't even pay cash because your wife would still know how much the present cost. Or the government could still know, even if you paid cash for the champagne for your daughter's wedding. But that's to an absurd limit. The point is there are really limits to how much a computer can do. In terms of selection, we can turn that around. We have Master Charge, we have Bank Americard or whatever they call VISA, we have Diner's Club and American Express. If one of these outfits could also provide privacy, I bet they would get a lot of customers, right? That's talking about the marketplace again.

Telephone Encryption

Lof:

Well, let's talk about tapping your telephone. A computer can access a line and pick up a ring going to another ring and tap into thousands and thousands of calls.

Penzias:

Sure.

Lof:

It can be very selective. It can only pick up exactly what it wants.

Penzias:

Absolutely. For a few dollars, I can provide you with an unbreakable code on your telephone if you have a digital telephone. I may decide that you're going to live in a society where I don't want the phone tapped, period. So I'm going to give the guy on the other end a trapdoor code, and he's going to give me one.

Lof:

That means I can only talk to the person who has that code?

Man:

What do you mean by a trapdoor code?

Penzias:

A code where only I can break it; he can't. I can give him a code. I can tell him to encode, but he can't decode. If I've told him how to encrypt voice, now I can decode it, but he can't. There are simple ones. You probably even know about trapdoor codes. There was something in Scientific American about them some years ago. You're worried about people tapping phones. I want to provide you the following service, not through the telephone. You can have a little processor on your phone. It's a separate box. You take two prime numbers, and you make an array out of them, and you need to know these two prime numbers in order to factor things out so you can decode the message. You don't tell him the two prime numbers; you just tell him the product. If these prime numbers both have thirty digits in them it will take him an hour to break it. You can change those every five minutes. You tell him to encode, you tell him how to code, and you don't tell him how to break it. He just puts in the code. You've got the other code because you've just done it yourself, and there's nobody in the world who can break it. Now he's talking to you, and he tells you one, and you put your code in, and only he can break it. You can't break your own message again. Now the two of you talk to each other, and someone can tap into that line from here to Doomsday, and it won't do them any good. It's a simple thing that could be added for a few dollars once we have digital telephones.

Lof:

Sounds kind of cumbersome to do all these big digits back and forth before you can talk.

Penzias:

It might take five seconds.

Lof:

Okay.

Penzias:

In fact it wouldn't even take a second because presumably this telephone is going at ten or twenty kilobytes. So in other words, you can send certainly send a thousand numbers in a second, and you don't need a thousand numbers to do this. You need twenty or thirty. So the first second, before you even start, you call the guy up, right? You've even got your little library over here. If you ever stopped to think what goes on in a digital calculator, or if you ever stopped to think what goes on in that telephone, you would say, "It's so complicated, I wouldn't even want one." Suppose there had never been a telephone, and all of a sudden I'm coming here today, and I'm telling you about that machine. And I'm saying, "All right, look. When you disconnect that relay, when you lift that up, what that does is it touches a pair of contacts. That closes a circuit which goes back to a line concentrator which is up on the telephone pole somewhere. The line concentrator is a sensor. Finally it gives billing information, and then ultimately it connects, scanning to find an unoccupied ring generator which goes on to the other end." By the time I tell you what it takes, you'd say, "I don't want one of those things. Not me." But it's all hidden from you.

In the same way this little box is simple. You go to your generator of unbreakable codes, and you pick one. You push a button, and your little processor goes to the other end. As soon as it goes off-hook, it tells his processor to encode. At the same time, he has a handshake, comes back and gives you a number, and you code with that one. Anybody in the middle can't monkey with it because if it came through in some other code, your code wouldn't decode it. Just after you go off-hook, there's a slight click, and the two of you are now talking, totally encrypted in code. If you need it once a day, we could build it for you for a few thousand dollars. But I'm giving you that as one example of an alternative.

Lof:

Yes, I understand.

Distributive Processing and Empowerment

Penzias:

If we have a free marketplace, there is an alternative to one big computer running everybody if everybody has access to their own machines, and machines can make you free. It looks like a complicated one. The little calculator in your pocket is yet another machine. It can make your life easier and less complicated, because you can add up all the buttons. Right?

Lof:

Oh, sure.

Penzias:

If you have a little pocket calculator, then Proctor & Gamble or whoever can't fool you by giving you 27-1/2 ounces for thirty cents off. In fact, people are putting a little label on to do the calculation, a little sticker. In other words, somewhere in the store there's a little calculator that divides those two numbers for you, and they put it on there automatically.

Lof:

Exactly.

Penzias:

So with the machine you are freer to make an intelligent choice than you were before. You can have machines not only on their side, but also on your side, provided one of these machines is small enough and you can own it. Without having to feel like Buck Rogers, if I had a digital telephone, I would want a privacy feature. Truthfully, I wouldn't want the telephone company to do it for me. They might provide me with the blank box, but I would want to, with five minutes of prompting, put in some random numbers of my own.

Lof:

Yes.

Penzias:

Then that box is unique from all the boxes in the world because it came out of my brain. I'm going to put in a few code words, and then that box now is generating unbreakable codes. I don't want the telephone company back in some central office having a big thick book and say, "Oh, yes, I want an unbreakable code. Please give me one out of the book." Right? First off, the nice thing about that is we wouldn't make any extra money opening the book for you. We've already made the money because we've sold you the service. What we want to do is connect you from here to there. That's our business, and we want to continue to have that business. Now there's a nice little box on the end which keeps you from being eavesdropped on. The telephone company makes a lot more money if there's no wiretapping, right? If there's wiretapping, people go to motels to talk, and we don't make anything. We want the secure network. The wiretappers are a very small fraction of our customers. They probably don't even pay us because they just go out to the pole.

Lof:

That's right.

Penzias:

So we don't want wiretapping, and we would like to help our customers. That's one of the nice things about being in business just to make money. We're not trying to save anybody's soul, so we'll give you a scrambler on each end if we can give you a digital telephone. A scrambler's just one example. Now you've got privacy again because you've got your own machine under your own control. That takes a certain amount of work. We are doing some of that, and I hope that when all the smoke clears we continue to do that. I would like to see a lot of little machines. Because I want to own my own machine, and then I want to be able to provide a political alternative. If the government has one big machine, they can say, "Look, if we don't have our big machine to take care of everybody, the world will come to an end. We can't keep records anymore. There's no way of planting crops, mailing Social Security checks, and whatever." If the only way of keeping records is a big machine, then you guarantee that all the power will go to the government. One of the things we're doing here, under words like "distributive processing" and "customer programming," is allowing people to be individuals. I've got a sociologist who isn't here this morning because on Wednesday mornings he baby-sits at home. From nine to twelve he's home with his children, working on his terminal. The connection allows him to be more of an individual. If we didn't have telephones, we'd go back to the days with this huge room, where everybody sits at a desk. There was a little sort of glassed-in office where Mr. Smithers (or whoever he was), the guy wearing a costume like mine, would sit there and make sure that everybody's head was bent over until lunchtime.

Lof:

Yes.

Penzias:

Telephone has allowed us to disperse, to have our own lifestyles, to talk to each other from our own little piece of the world where at least we can put a picture on the wall or we can do things differently. In that same way that allows for individual diversity, if the telephone, as it should talks to machines, we should also be able to provide for machine diversity. It's one of the things I really would like to see. It's not just a hardware problem, because we can make microprocessors. You've probably seen some of the ads for our MAC-4 and all the various little chips that we're making now. Lots of computers are getting small.

Lof:

Yes, I know.

Penzias:

The problem is to get them to talk to each other, and that's where we're talking about in our advanced ACS; that's what we're talking about in all our research. We're doing a lot of trying to get the smarts there, which is not all that hard, but getting them to talk to each other is very hard. This is such an exciting place to work in the software area. In fact, I hope it's an impossible job.

Lof:

Why?

Penzias:

I'm thinking about the next step. The thing I hope is an impossible job is that one big machine with everybody working on it. That's the easy job, and I hope that job is impossible. At some point the economies of scale on computers have got to stop because, after all, what happens is that programmers only work a certain amount of time. If the program gets too big, and the people start bumping into each other, by the time that whole program gets together, some of those people have left and done something else. Then nobody knows how that part works, and they bump into each other, and you get software crashes. I hope that problem is not going to be solved. I would rather have smaller things. But in order to have the smaller things, which I think in the end will be easier, we have to be able to talk to each other. Then we need a lot of flexibility because the smaller things require the customer to do a lot more. One of the prices of freedom is a little bit of work. But it only can be a little bit. We can't expect people to learn FORTRAN or C or one of these things. What we have to do is teach these computers to understand people. If we can do that, and then allow for these various idiosyncrasies of the computers to be taken care of by the intervening telephone network, I think we can have a different kind of society in the year 2000. That's the kind of society I'd like to see us get into. That's what I think we're really doing in the telephone company. I mean if everything is inside of one big computer, the telephone company isn't anyplace because nobody's talking to each other, right?

Lof:

Exactly.

Penzias:

We need lines in the middle of this thing, or we're out of business. People will still talk to each other. Presumably the IRS guy will still call you up. Or Big Brother will call you from Washington. Or who knows? There will always be a telephone business, but it wouldn't be a very nice country to live in.

Lof:

No, it wouldn't be very nice at all.

Penzias:

I don't want you to think of technology as going only in that direction. If I can leave you with one idea, it would be this idea of distributive processing.

Lof:

How are you going to educate the country?

Penzias:

You don't have to educate the country. Everybody hates big computers. Everybody likes little things. I don't think it's education. People are far more sophisticated than you give them credit for. I don't think you need a Big Brother out there. I think market forces teach us that.

Lof:

But we're allowing Big Brother to take over all the time. We're giving up more and more to centralization.

Penzias:

Because the alternatives don't exist.

Lof:

I could save money for my retirement, could I not, instead of letting the government take care of me?

Penzias:

Not easily. Nobody has in the past. People would decide, “Well, I'll save it next year.” It's not just retirement; it's disability, right?

Lof:

Yes.

Penzias:

Somebody's husband runs off, or somebody get paralyzed, or any one of a number of things happen, and then the society would say, “Well, I was going to start saving next year, but this year I really wanted that boat.” So the whole society's supposed to take care of it. We institutionalize some of these in Social Security, whether it's good or not. People who are healthy and independent don't like Social Security until they need it, because nobody expects to need it. That's why people smoke, right? People don't always take care of themselves. In a discussion on telecommunications, I shouldn't have to defend Social Security.

Lof:

I agree.

Penzias:

But I will say that if there is no alternative to Big Brother, it's inevitable. In communications we’re trying to provide an alternative. I really shouldn't have spent a lot of time today talking about this. If you wanted to discuss Bell Laboratories and the good things we're doing, we ended up in a different conversation.

Lof:

Yes, we did.

Penzias:

I would like to see communications with machines for a purpose. There could be two kinds of communication. There could be little sensors sending stuff into the central place, and then you have the one big machine in control. Next door I've got a computer made by Tandem. One of the reasons we bought one, was because there's no one in control. Tandem does not have a central processor. Tandem is a machine where each of the processors is equal, and they all talk to each other. We are investigating for various other applications that sort of environment. You could invent something where there is a controller. In fact, some of our new machines work in both directions. Usually, until now of course, we do have this hierarchy; somebody is in control. Down here in Bedminster we have one thing that has the control. Obviously we want to control the entire network, one product. In order to control the entire network, you need someone to decide, “You'd better start moving some things through Atlanta because Chicago is messed up with a snowstorm.” Whatever. So you do need somebody in control for some kinds of things. That doesn't mean we have to exert the same kind of control in the information flow. It doesn't have to be a network where everything feeds into Washington. Washington's got too much already.

But if we can provide the individual with the capability of doing for himself and herself things which are only now done in a centralized way, we at least give the society the option in the year 2000 of voting in the political process. I assume twenty years from now they'll still have free elections. We've had them for 200 years; I see no reason to stop. Maybe the person who can get elected will get elected, just as I think the credit card company could get the business, by providing freedom for the individual. You can sell freedom to people. I think they're smart enough to understand that they don't want to be controlled. In the political marketplace which is, “Vote for me; I'm going to do so-and-so”, you can do exactly the same thing as we hope to do in the economic marketplace. We are not a government agency. You have to want to buy our product. One of the products I think we can sell you is the capability of having your own personal computer or record-keeping or whatever. This would be for individuals, but also for your business. Through our ACS, for example, we can provide you with the ability to talk to other people so that your computer can talk to other people's computers. Your little computer can talk to other people's computers, so you don't have to give up the whole thing at one centralized computer somewhere in a vault in Des Moines. Then when you want to take a vote, it isn't a vote about going back to an agrarian society, because we might do that. Then the society may also not be one worth living in because then you get back to sleeping-car porters or the equivalent — somebody's going to do the menial jobs — and secretaries and so forth. You'd have somebody in control and some other people helping keeping the rest of the people down. A backwards society — a low-technology society — is inherently an unjust society because they're not going to move the pie equally. I just don't know of any society in the past where the shrinking pie has ever been divided up equally.

Lof:

No.

Penzias:

If we say that the only alternative to centralization is chaos, we're probably going to stick with centralization. But if the alternative to centralized control is individual capability in a complicated society, then I think I would like to vote for the individual control.

Lof:

I would, too.

Privacy and Electronic Cash

Penzias:

So you're a potential customer, and you're a potential voter. But in order to do that, we need that capability; we don't have it right now. It's very hard to get computers to understand people. Now we say you have to have the one big computer back there is because we need programmers, and you're not a programmer. So therefore, just fill out this tax form — Or never mind. Don't bother filling out anything. We'll watch you, and it'll be coded for your benefit, and it'll all be on there. Every time you get a paycheck, we'll know it. Every time you spend some money, we'll know it, we'll put it on there, because we've got programmers that can do that. You can't program; therefore you can't keep your own records.

When given the choice between being a programmer or letting the government keep records, most people will just decide they can't be programmers. They want to do something else; they'll let the government do it. But if you could keep those records without being a programmer, if you could get a computer which you could talk to, then in fact you could say, “I don't want the government to have that kind of control over me. I want to go to Bank Americard or to go to this other record company because they're not going to send statements someplace else.” In fact they don't even know how I spend my money. Because what we've done is when I go to a store, we have both agreed that they're keeping my wallet, and they're going to use something like what kidnappers do. One of the problems with kidnapping is, how do you get paid? The way kidnappers usually do it is go take some money, put it in a box, and then walk away. So in the same way, what you do is you go to the liquor store, and you say, “Okay, how much money do you want?” Then somehow in there you call up Chase Manhattan Bank, and you say, “Put some money into a box, and walk away from it. And give the trapdoor code.” They get paid because you've told them that's the box. Now it's your problem. The guy is now in the liquor store and says, “I've opened the box. I found it, and now I pay you.” It can take a few milliseconds.

Lof:

The whole transaction is done, and there's no record of it?

Penzias:

I'm not saying it absolutely can be done. I can't promise you the future twenty years from now. I'm not sure what we're going to do. We have progress reports coming in here every month. But I'm giving you a conceivable scenario, right?

Lof:

Yes.

Penzias:

The credit card company that solves the liquor store problem, has a marketing advantage over the others.

Lof:

It certainly does.

Penzias:

You're beginning to become intrigued with the idea of finding a company where, if you want to buy a present for somebody, you can somehow keep the price of the jewelry store bill out of the monthly Bank Americard.

Lof:

Without carrying the cash in your pocket.

Penzias:

Without having to carry the cash in my pocket. I'm not saying I'm even solving that problem; I'm giving you a class of problems. It doesn't have to be done all in one day. But there are these kinds of things which are possible, given distributive processing. Given the possibility of computers talking to each other. Given computers that you can program yourself without knowing that you're programming.

Lof:

With your voice....

Penzias:

Nobody likes to type. One of the other things we are working on is speech recognition. In this system we should have these machines look as much as possible like the human beings they're replacing. Obviously voice recognition is one of the things that will also have a big advantage.

Alternatives to Centralization

Penzias:

This is what I see as one of the opportunities. It isn't enough just to have an idea. The ecologists do this a lot, some of them, the anti-technologists. They're not really anti-technologists. They're people. Do you remember seeing Dustin Hoffman's movie The Graduate?

Lof:

Yes.

Penzias:

This one guy says, "Come on. Over here." He grabs him early on in the party. He grabs the kid over, and he says, "I've got to say one word to you." He says, "Plastics," and walks away. In the same way as somebody in our society says, "solar energy," and then walks away. I don't want to say, "distributive processing," and walk away. We need Bell Laboratories for the next generation. I'm merely trying to say that there are technological alternatives to centralization which make life more worth living, if we decide as a society, to go that way. We certainly are not going to go that way if the same people who already run centralized record-keeping decide what technological alternatives are going to be there. Obviously, if you run centralized record-keeping, there's no reason to put yourself out of business.

Lof:

Right.

Penzias:

People can decide whether to use the telephone or go to a motel or have a conference. People can decide whether or not to buy a picture phone in the days when it was supposed to be an office set rather than a conferencing set, which seems to be working very well.

Lof:

For conferencing?

Penzias:

Picture phone conferencing seems to be much more acceptable than having one of these things on your desk. When we are pushed in the right direction by market forces, we have this feedback, we have the correction. We have to provide something the customer will buy.

Lof:

Sure. Of course.

Penzias:

We could legislate. We could, for instance, take out all the plain black phones and put in blue ones if we wanted to pay for it ourselves. But, given the regulatory situation, if we somehow decide blue telephones are better, we might be able to put over some silly thing like that on our customers. Any change of that nature is very hard for us to do because we're not the government.

Lof:

Exactly.

Penzias:

The telephone company looks very big and is very big, and has a slow turning radius when it comes to anything to do with change. Sometimes people get mad at us, I know. I'm a telephone customer, too. There are things you'd like to change. But we have to be responsive. Because as we move, we can only move in those directions that our customers are willing to have us.

Lof:

Because you're selling your service.

Penzias:

We're selling a service. If it's no good, you may still have one in your house. But if your teenagers don't want to talk over it, they're not going to buy the second one. So we've always been in a competitive business. We’ve realized [not] as much as we might have. I see us increasing in our computer science areas, partly for our own purposes.

Lof:

And programming for people and not for machines?

Penzias:

We are programming for machines, but we have to program for machines that people will accept. We also have to do the machines that work. The big machines have advantages of scale, but they have lots wrong with them. Even if we decide we like the idea of having government do everything for us, when one of those big machines has a disk crash, we are in a lot of trouble. At some point the computer gets too big. The people who are trying to program it bump into each other. So there are probably limits, just given to human life expectancy: From how long somebody wants to stay in one job, to how big a machine you can get and how many lines of code you can document. We're trying to do things in an economical way because we don't want to have to go out there and ask for rate increases. We get our head beaten in, you know. We try to provide an economic service. It's the only way to put change in. You have to try to show the new thing's cheaper than the old. The big machines aren't all that easy to prove economically. For economic reasons, since we're not the government, we can't tax people. We have to provide a cheap service they're willing to pay for. Technology doesn't inevitably have to lead to large-scale things. The first steam engines had these huge locomotives. What you saw then was the muscle power being replaced on a very large scale.

Lof:

The first computer.

Penzias:

Yes. But the mine owner or the railroad owner was the only one who owned the machine. You would think of bigger and bigger machines, and it never occurred to you that someday you'd have an electric toothbrush. In the same way, when we now see the machines getting bigger and bigger, we also see that there may be alternatives. It may be the little one, right? It was the truck that beat the railroad.

Lof:

Right.

Penzias:

Just as the bus and the truck beat the railroad, maybe the small processor will beat the large. We have calculators and mini-computers, but they're limited. Doing something fairly complicated right now takes a big machine. Writing a line of code is so expensive, that you put it in a big machine so everybody shares the same program. You can't afford to program each program individually because the cost of programming is so high. Even if you did it yourself, in order to get this stuff to the consumer, we have to make it easier. That means we have to be able to understand language and people's desires. We're working on that. If you came back and talked about the details of Bell Laboratories, people would find that sort of thing in great numbers. But it's a very tough problem, and this is the thing we started with originally when I came in here. We spend a lot of time on people problems here. They're not problems in the sense that we'd be better off without people. What we're really trying to do is get the best possible people to work on these things, and to make our own climate as compatible as possible. We don't want the standard solution. We don’t just do everything we've done in the past, which is a lot easier. We don’t hire the same people who used to work in the past, and don’t do things and manage in the same way. It would be much duller. But it also would be inevitable that we wouldn't do these things —

Recruiting and Managing Creative People

Lof:

How do you find the people you're recruiting these days? When they come straight out of college, how is their education compared to what it was twenty years ago? Are the schools giving them the proper tools?

Penzias:

It's not a question of the schools giving them the tools. It's a question of how much they take away.

Lof:

How much are they taking away?

Penzias:

I don't know. On the one end, you put in curious five-year-olds, and then at the other end you get somebody out at the age of twenty-five or thirty with a certain number of skills and a little bit of curiosity left, a certain amount of inventiveness. People are durable, fortunately, so schools don't hurt them all that badly. That's the best I can say for school. We're really trying to make use of the natural resources nature gives people. Whoever those people are, we try to help them survive in school, with the programs for minorities, for instance. Inherent intelligence and creativity are within us and reinforced maybe, and we can teach them skills along with it. We really need the brightest people we can get to work on these problems, and hopefully people who can think in novel ways. We want somebody who isn't just blindly loyal to something. One problem I have is people who think they're working for a different company.

Lof:

What do you mean by that?

Penzias:

That somehow there is a boss who wants some kind of line loyalty and very uptight. And who wants, when they send him a message, for them to talk about personnel instead of people. They're people who have a mistaken idea there's some kind of very structured corporate image.

Lof:

And there's not here?

Penzias:

There is some, but there's a lot less, and that isn't what we're looking for. That isn't going to solve our problem. We're looking for the most creative people.

Lof:

You're looking for them not only to solve a problem but to define a problem.

Penzias:

Absolutely! Because I don't know what the problem is. I personally could barely write a FORTRAN program to do my income tax. That's the total level of my understanding of computers. But I do understand when somebody says, "I've got a box over here. I want a big controller and a whole bunch of little processors." And I say, "Well, that isn't the way the world is. What I really want are a whole bunch of processors. I'd like a democratic computer." So that kind of thing I can do. But beyond that, these people have to find their own problems and then tell me: “This is what I want to work on because...." So we need people to define. We need really creative people. We need independent people. If anything, I wish we could get more independent people than we do. Wearing jeans is not enough.

Lof:

No, of course it's not.

Penzias:

But a lot of people think that if they come in here in blue jeans, and they say "you know," and maybe smoke marijuana at home, therefore they're really loose. But they're not.

Lof:

Well, it was like letting the hair grow long.

Penzias:

Yes.

Lof:

Even construction workers were letting their hair grow long in liberation.

Penzias:

We're looking for variety and creativity. One of the nice things about variety is that it also helps creativity. If everybody looks alike, and acts alike, then people begin to think we really want people who think alike. If we keep hiring people who look alike and act alike, we're probably giving them more of that message than we realize.

Lof:

But you still have to supervise about 300 of them?

Penzias:

I don't really supervise, no. "Supervise" is the wrong word. They don't ask me what to do. What we do is provide an atmosphere of criticism and direction.

Lof:

But in some way you have to control that atmosphere.

Penzias:

Influence, not control. I can't control it.

Lof:

Influence. Thank you.

Penzias:

It's a much better word. I influence it. They take a magazine interview and put it on the bulletin board. Like the one from Omni. There was one in Omni where I said, "With all this computer processing, I could do better with a case of dog food and a pooch from the pound." We’ve got some people down here, and we're going to do a better job.” People were saying, “Well, damn it! That opinionated son of a bitch! I'll show him.” As a result we've got people now doing speech understanding in rather more ambitious ways than they might have.

Lof:

Well, that's good then.

Man:

Yes, indeed.

Penzias:

Sure. But I can't go pick the person and say, “I want you to stop what you're doing and work on speech understanding.” I can do it occasionally, and I might. It's a gradation. That is, there are some people whom we leave totally alone. Others we guide a little bit, and others we guide a lot, depending on what our perception is of how well they're able to pick problems. Some people are luckier than others. What we really want is more lucky people. Some scientists are lucky. They always seem to be working on the right problem.

Lof:

I think people make their own luck.

Penzias:

Okay! But that's what I'm saying by lucky. I didn't say where it comes from. You're saying the same thing I'm saying. I don't think being lucky is an accident.

Lof:

No, I don't doubt it.

Penzias:

That's what I mean by lucky. Branch Rickey said, "Luck is the residual of planning."

Lof:

Okay. I'll buy that.

Penzias:

We're looking for people who are going to be lucky in some way. People who are going to take chances.

Lof:

You're looking for the personality that's going to go out and make a contribution.

Invention of UNIX

Penzias:

This is really what we look for at Bell Labs. When you start thinking that research is only a tenth of this company, and there's something like a thousand Ph.D.'s in research, you think, “How can I make a difference?” Of course you can. You start by going down to one problem. You break all these different things down. The individual makes the whole difference. We don't put six people on a project. In Development, if they need a switching machine, a digital switch, in two years, they go out and put hundreds of people on it. But in Research, usually it's one to two. UNIX was invented by two guys who got so sick of operating systems that they decided to go into an attic. They made this terrible operating system. The nice thing about it was they were in Research. It was inefficient and awful. But then they were left alone, and they finally made UNIX, and now UNIX is a household word among users of certain kinds of computers.

Lof:

What is UNIX, please?

Penzias:

It's a computer operating system, which works on some mini-computers. A couple of people invented it for themselves here, and we actually sell it. We license it free to universities, but we made more than a million dollars licensing it.

Man:

It's widely accepted outside of the Bell System.

Penzias:

It's a computer operating system, but two guys did it. Two guys who were sick of —

Lof:

With a idea.

Penzias:

Somebody didn't say, “Well, we need a new operating system.” They just got on it. Individuals make a difference, and what we try to do is identify and encourage those individuals to look at those problems. I'm not smart enough to know what I want, what can be done in the year 2000. All I can say is that I've got some questions where I really care about the answers. Sometimes I'm right; sometimes I'm wrong. Sometimes the department head argues with me and says, "We ought to be in this business." I say, "Well, I don't understand it." "Well, that's too bad, Arno. But we're going to be in it. I'm sorry you don't understand it." They're going to fight because they think it's important. That kind of thing is really good, because if they really care enough to argue with me, they're probably right. It sounds pollyanna almost. But the people here will do their best. I remember how people worked on the outside. You think of the telephone; that means being infinitely rich. I don't have enough space, people, materials, anything. Yet I've got my bosses' confidence. I've got a lot of support. But there are so many problems, so many good things to work on. We can only work on the most important ones, and we try to get the few people we have to work as effectively as possible.

Lof:

You're always going to be beset by some sort of limitations.

Penzias:

Yes. Some things pull their own weight. But right now, in a healthy organization, you'd always like about twenty percent more. So what we try to do is get as much as we can out of the people we have. And the way to do that is to get the most creative possible people. Because we can't afford to throw twenty people at a problem in research. It's usually ones and twos. The amazing thing is how few people go to a meeting. There are those few who are really working on the problems.

Lof:

I read a quote by a Professor Panofsky, a fellow physicist out in California. He talked about the dilemma of a prominent scientist who if he uses his name to endorse something, he's called arrogant. If he keeps quiet, he's called cowardly. How do you deal with that, being prominent, having achieved a great deal of recognition?

Personal Opinions vs. Obligations to Bell

Penzias:


Audio File
MP3 Audio
(037 - penzias - clip 3.mp3)


With difficulty. You don't want to be cowardly. On the other hand, you don't want to become an endorser. You can't say, “Well, now that I'm a prominent scientist, I'm an instant expert on everything, and so I will go around giving people advice.” That doesn't work. On the other hand, if you do have something to say, it's usually better to say it. It had better be something important, and it had better be in an area where you have a right to an opinion. You do have some freedom. I now can talk about how much I hated some aspects of school, which I really did. In some ways it was very helpful. Education turned me from a poor refugee kid into a quite prosperous and well-supported scientist and a member of the upper-middle-class. So I don't knock education. On the other hand, I can see things that happened to me in school, where I was different from the other kids, and where that sort of thing was used against people. It's used in the same way to de-motivate people who are different by reasons of race or sex or economic background or whatever. In an area like that, I have an obligation to talk about it. I think what you have to do is not talk about everything, but try to talk about things that you feel are important. That's how I think I deal with it.

As far as dealing with it at Bell Laboratories, you have the question of how to deal with the requests for information, the requests from reporters. What do you do? How do you deal with the fact? Bell Laboratories, fortunately, has a number of Nobel Prize winners, so they don't have to bring out their tame Nobel Prize winner every time somebody comes through. So there isn't that insurmountable pressure in having to talk to people, or having to do the kind of thing we're doing now. When it makes a difference, when maybe I have a bigger audience, and I feel strongly about something, then I think of it as an opportunity and, to some extent, an obligation. For most individuals the environment in which the prize was won was also contributed to by the people at Bell Laboratories, and Bell Laboratories is an institution. So I have an obligation — and this is just earning my salary — I have an obligation to the institution. But it’s not so much of an obligation that I have to lie or somehow distort how I feel. In those areas where I think it can help, I give a certain amount of my time. I'm not selling a product exactly, but there is a certain amount of salesmanship involved, I suppose. I'm trying to get across the idea of what Bell Labs is like.

Lof:

Well, you think it's very important.

Bell Labs as Caring Individuals

Penzias:

I think it's very important. I don't want people to think of Bell Labs as this place which is still working on the transistor, and just endless corridors of identical people. And that all we're marching for is in lockstep with oppressive technology which is ultimately going to destroy privacy and make us all some kind of monolithic state or whatever. Quite the contrary, it's a group of individuals who are working to make life better. I hope every day that something we have done is going to make life a little bit better for everybody. People say, “Bell Labs wants,” or “Bell Labs does,” or “AT&T's opinion is.” What do you mean, AT&T? There is no AT&T, there is no Bell Labs. They're people. I mean if you want to say Ian Ross, I'd pay attention. He's my boss. That's not to say that everybody's equal. But it's down to individuals. What we are is the sum of individuals. It not only has their best qualities; it also has their greed and their suspicion and their ambition and all the bad and good things. It's not all perfect, but it is a bunch of individuals. You're not willing to forgive a mistake of somebody who pretends something they're not. If you pretend to be perfect, and some kind of angel, and then you make a mistake, you can get into a lot of trouble. But, usually people understand that by talking freely you're taking a certain number of risks, and people expect that once in a while you'll say something wrong. I talk about my minority things, for instance. Occasionally I get misunderstood. Once in a while something comes out that even sounds racist because I haven't explained myself. I jumped ahead or something else. But basically the alternative is to pretend that there is no race problem in America, and I'm not going to do that. I'm even talking about the fact the telephone company has problems. The alternative is to not talk about it at all, and then give the impression of a monolithic company which is blindly marching ahead to lockstep, independent of its customers. So, by talking about the faults we have, and our warts and things like that, I hope I'm giving a more accurate, but also a more acceptable, picture of the place. If we could get them, what we really need now is a few more creative people. Right this minute.

Lof:

Scientists?

Penzias:

They don't have to all be scientists. They can be engineers. We need technicians. We need all kinds of people — mostly scientists. But right now we need more creative people. We also need a public understanding that you can kill the telephone company.

Lof:

That you can?

Penzias:

Sure. You can kill the creative part.

Lof:

Well this is one of my original questions to you. From the research that I've done, I have —

Penzias:

We'll drive you to New York. We can talk in the car a little bit. I was originally going to drive you in myself because I want to be in New York. But I talk with my hands, so I've gotten a driver for us, and that's why everybody is on me.... [Pause] The interaction between people and machines make it really the biggest. I wanted to talk a little bit about microprocessors. [Speaking while walking] That area really is growing. That's not to say there aren't people here in other things. You know, an x-particle physicist knows computer programs. And whatever. They're all coming in at various times. So it's not to say any one of these isn't being pushed.

Lof:

But doesn't everybody have to be part computer scientist today?

Penzias:

I'm not. Up until a couple of years ago, when I wanted to measure the area of a curve, I did it on an analytical balance.

Lof:

Are you doing any astronomy now?

Penzias:

Oh, yes. I do some research at Princeton, part-time.

Lof:

Do you find yourself doing more business now than science?

Penzias:

I've always spent about half my time on telephone company things. Years ago that used to be just doing thing like satellite work, satellite communications. But now the phone company part of my job is more managerial. I have an administrator who worries about the budgets and things like that, but the managerial part takes over most of my phone company work.

[Drive to New York City]

Penzias:

Are you going to be taking us to New York?

Chauffeur:

Yes. Follow me. [Garage sounds]

Penzias:

This is an unusually nice car.

Chauffeur:

We're going to go for a diesel ride today.

Penzias:

We're trying our fuel economy thing, using different cars. This is about as nice a car as Bell Labs has.

Lof:

Yes?

Penzias:

Yes. They have these for executives for business purposes.

Lof:

Well, this is certainly service that I'm getting. I got picked up by my door this morning.

Penzias:

Yes. Well, as I say, I was going to drive in to New York. I have a couple of meetings in New York today. I was going to drive myself, and I would have driven you. But just as I got here, I had the feeling that it's almost a little dangerous. I get excited when we start talking. I'm going to go wave over there, and I won't be watching the traffic, and that's not such a good idea. So I decided this would work out better. Well, what else would you like to discuss?

Lof:

Well, you've been off onto your own track, and I just let it go.

Penzias:

Yes. But we still have an hour or so. Maybe you want to talk about the things that you have on your mind. I came in this morning, and I got hit by this affirmative action concern.

Fellowship Program for Women

Lof:

How are you going to get more women to study electrical engineering and advanced physics?

Penzias:

Now we have fellowship programs, for instance. We have fellowship programs for women where we pay all the expenses. We take people at the bachelor's level. We have other programs for undergraduates. But there are a lot of bachelor's graduates. We have summer research programs. We bring people in in the summer, usually at the end of their junior year and we let them work in an industrial environment. Then they find it's not as threatening as they might have thought; it's more interesting. Gets them turned on. Then when they're ready to go to graduate school, we have a small number of — In dollars it's a lot of money for us, but in total numbers of people it's not all that big. If you count minorities and women, it's in the dozens. These people are off in graduate schools studying with some support, because at Bell Labs, they have an individual who acts as a helper, a mentor, through this graduate program. Then afterwards, when we start getting such women on board, as with minorities, we get a kind of word-of-mouth, or you get a build-up.

Lof:

The academic community, I know, is very upset because the starting salary for somebody with a bachelor's — I'm more familiar with electrical engineering — is getting to be astronomical.

Penzias:

For a starting salary sure.

Lof:

The kids are opting out of graduate school. They're feeling that there's going to be a big crunch. I was reading someplace where the most qualified people applying for professorial jobs are not American nationals; they're not citizens.

Supply of Ph.D.’s

Penzias:

There was a big hiring problem, I know, in physics and astronomy — certainly in astronomy, and astronomy was easier. I think you're mis-connecting a couple of things. In most Ph.D.-level things, there is still a big oversupply of Ph.D.'s. But it's coming down as people don't go to graduate school and find undergraduate employment more attractive. My own son, who's an electrical engineer at the University of Pennsylvania and is finishing his junior year, for instance, is thinking that he'd rather go off and work at the end of his bachelor's degree than take any more schooling, because there are a lot of bachelor's-level job opportunities. I don't think you can legislate such things. I think those things tend to even out, and ultimately move off into line. There's always been a shortage of really bright people, even in the days of the fullest surplus and the big budget crunches in the hiring and everything. We still never got as many truly top-notch people as we would like. There's always a need for top-notch people. Universities have had some problems. They're probably right now not perceived as very good employers. The faculties are heavily tenured. I think Time Magazine said that college professors' salaries relative to the population in general have shrunk more than anybody except people on welfare.

Lof:

Really?

Penzias:

They were the second lowest category.

Lof:

Do you pay well at Bell Labs?

Penzias:

I hope so. That is, we try. What I hope we're doing is staying competitive. We're paying enough to hire the people we want. That's how salaries are set.

The FCC and Regulation

Lof:

How do you feel about a bunch of politicians legislating on scientific matters — the FCC?

Penzias:

You're saying it in a pejorative way. I don't think it's necessary for a judge to have actually gone out and committed murder before he can sit in on a criminal trial. You don't have to. Society, through its elected representatives, has a right to decide what it wants. People in Congress can pass draft laws even though they're all over the draft age. I'm sometimes frustrated about the lack of understanding that exists on both sides. It would be nicer if people understood the issues a little bit better. Resolving some of these things may look easier to people than it really is sometimes. The government has a legitimate right and a need to have parts of society responsive to the entire society. So I think the government has every right to pass laws.

Lof:

Pass laws on —

Penzias:

On anything.

Lof:

Things they don't understand?

Penzias:

Well, they have to pass the laws. I mean you don't have to understand how a telephone works. It would be helpful. In fact they have staff and people who try to understand the details, the technological issues involved. You hope that the legislators are enlightened enough to understand things like the nature of basic research. That's a problem. You can't always assign the cost of doing basic research to a product that's bought.

Lof:

No, you can't. I understand that.

Penzias:

I hope that legislators, for instance, understand the need for continuity in research, for enough independence and freedom for research, and for people working on the important problems. The things we talked about today wouldn't be part of this.

Lof:

No, they wouldn't be.

Penzias:

I'm not against passing laws. I just want to make sure that good laws are passed; that's a different thing.

Lof:

How do you deal with a statement like: “We're in a deregulatory climate, so we will deregulate the phone company?”

Penzias:

I don't think that's been said.

Lof:

I have this on tape from a congressman.

Penzias:

In some cases they feel that regulation is not appropriate, and feel that the marketplace will do better. At the same time, they ought to understand that you can't deregulate everything about a utility. You certainly can't deregulate basic telephone service. You might be able to deregulate terminals or something, other things. There are going to be problems with that, but I think the problems can be and are being addressed. Whether they're being addressed successfully, I'm not well enough informed to know. You have to keep some of the benefits of vertical integration. On the other hand, you also want to give the consumer a maximum opportunity to make use of a diverse number of suppliers. That means you have to have very good accountants. I'm not an accountant, so I don't know how hard that is. I hope it's possible. I would like to continue to do basic research; I would like to see Bell Laboratories continue to provide the best and most reliable information design criteria for communications equipment in the future. How you make sure that money which comes from a regulated service doesn't subsidize something which is competitive, is an accounting problem.

Lof:

But aside from the basic accountancy, if they're going to be competing with companies and lowering rates, there's going to be less money.

Penzias:

Oh, no. no, no. There's nothing about lowering rates. I don't think people seriously argue that when they're all done, this is going to be cheaper. Telephone service is not that expensive. We are saying we should give people access to other technologies. If somebody else wants to buy a Japanese switchboard, you have to let them. Let's say we were making extra money on what we call vertical services; say, the second telephone. We used to make money on the second telephone to subsidize the first one. We gave everybody the first telephone at a rock-bottom cost, and made profits on the second telephone. Well, now people go buy the second telephone at Radio Shack, and that means you have to charge more for the first one. The total amount is the same, except that there probably is some advantage for people who have six telephones in a suburban house. Presumably, when we get out of certain businesses, the revenue from those businesses won't be there anymore. But that doesn't mean, necessarily, that the total revenue of the telephone company will go down. All these other gee-whiz companies want to get in the phone business these days because it's a growing business.

Lof:

It's getting back to your expanding pie.

Penzias:

So now you've got an expanding business, and it may be that there'll be many more opportunities generated. The revenue upon which development and research is based is quite large and may still be growing. It's not obvious that if the terminals become deregulated, there won't be as much money for research. It's really more the question of the basic decision: What is the benefit to the telephone communications customer, to having somebody working in a general way to make communications better, rather than waiting for the Japanese or Siemens or somebody in America making the product and then trying to sell it at a higher price later on? There's a balance between those two. I think the first one has proven to be so much better, that we'll continue to do that. We have to have a national discussion about the benefits to communications of hanging onto this tremendously successful enterprise called Bell Laboratories. The problem is that you may do things for ideological reasons, and then find afterwards that you've really done something you didn't intend to do in the first place. That's why we are doing a lot of talking, and I think there are people doing a lot of listening. We are learning at the same time. So there's a dialogue going on right now.

Wish-Fulfillment and Research Mistakes

Lof:

You're in science. You make findings and predictions on your findings. Somebody else can be doing research on something and make a finding that's diametrically opposed to yours. The Sugar Council found that saccharine caused cancer. The saccharine people, the Calorie Council, found the exact opposite. How can you have two sets of facts saying exactly opposite things?

Penzias:

I think doing science is extremely hard and scientists, like everybody else, can make mistakes. One of the easiest ways of making mistakes has to do with wish-fulfillment.

Lof:

I understand.

Penzias:

My wife is a teacher, so let me go for that example. If you see the child coming in the first day, and you put a mental label on the child saying, "brat" or "spoiled" or whatever, you'll be right. Your own expectations tend to influence him. It’s easy to make mistakes when you let your prejudices to get in the way. And scientists, like everybody else, have to recognize their prejudices, allow for them, and correct for them. It's not easy to do. People who look at the telephone company as a big monopoly, as Big Brother, or whatever kind of words you want to use — also have this problem. You don't have to be a scientist to make mistakes. On the other hand, we have this perception of scientists as somehow having to be perfect. Some scientists would like to believe that. One of the problems we have about images of scientists is self-image as well as the one society puts on us. But we are not perfect, and we make mistakes.

Lof:

How do you deal with your mistakes?

Penzias:

We correct them. Everybody makes mistakes. The people who gave us the South Bronx also made some mistakes; they were poor scientists.

Lof:

Nobody's acknowledging guilt for it, though.

Penzias:

That's the difference. Nobody's acknowledging the guilt. Look at the people who somehow decided the business of being paid for having a fire in your apartment, making some kinds of drugs illegal when all they're doing is jacking up the price, and deciding eligibility and sufficiency in breaking up families on welfare. All these can result in really disastrous things, far worse disasters than technological things. I don't know of any technological disaster approaching the scale of the things we have with our social experimentation.

Lof:

Well, look at the people.

Penzias:

We're talking about numbers of dead bodies. All the people in the country who have ultimately died because of technological things is peanuts compared to the kinds of things we've had from social things. Racism.

Three-Mile Island

Penzias:

The best example I can give is the business of Three-Mile Island. There was an article in the New York Times a few weeks ago. They put a circle around Three-Mile Island and counted the number of deaths, and so forth — and it was larger than the state average. They got hysterical. It was statistically valid. Then somebody came back and said, "Of course. We have Harrisburg in there. We know what's wrong with Harrisburg. It's full of black people." So we take Harrisburg out, and now it's back to state average, and it's fine. There are minority people in Harrisburg who die in much greater numbers. That changed the statistics. But the point is, they're just ignoring the real problem, Harrisburg. We know what happens when poverty is there. So we're comparing social with technological, and everybody is beating the scientist over the head.

Lof:

Yes, they are. Particularly with that Three-Mile Island.

Penzias:

On the other hand, the scientists did deserve a few things. We probably were a lot smugger than we ought to have been. We tend to believe some of our own reputation for infallibility, which is also bad.

Lof:

I've done some reading about Three-Mile Island.

Penzias:

Yes?

Lof:

One of the suggestions, of course, is that you get some very highly trained people with tremendous skills to sit and watch these machines which probably won't go bad maybe more than once in their entire life. How could you get anybody to take that job?

Penzias:

I've got a better story for you. I don't think it's anything that hard. I don't think you have to invent something new. Let me just say the following: When Three-Mile Island went sour, the Department of Energy to my understanding had 40,000 employees, twenty-six of which were field inspectors.

Afterwards, I talked to some people, and I said, "Look, if the government wants to convince the local population that place is safe, you'll live there." Now Congress has allocated money for resident inspectors. You need somebody watching. Well, no inspector went to that plant for two weeks. The thing that's using a million gallons of oil a day can certainly afford to have a full-time inspector. He doesn't have to be a genius, but he has to be there. We have not already used the limits of our ingenuity. I don't think we've gotten anywhere near it. It’s a human problem. People imagine that somehow you need some enormous technological fix, that you have to do something dramatically different. If we had somebody who said, “We've got seventy plants, and twenty-six inspectors. One-quarter of them are going to be on sick-leave or vacation or something. Who's watching the plants?” It doesn't take a lot. If you want to convince people the plant is good, have the inspector live there. Buy him a nice house. Somebody with a family. Because his kids schools are in the local school system and he bowls with whomever he is now plugged into the community, and he's going to get lots of input, whether he wants it or not.

Lof:

But isn't that a sign of being very bright finding a simple solution?

Education and Diversity

Penzias:

No! I think lots of people could find simple solutions. Maybe you have to be very bright because you have to get around all this intimidation. We all have this creativity within us, but it's been beaten down by education. We are afraid to come up with simple solutions because we think that the solution has to be complicated.

Lof:

But doesn't our education teach us to look for something complicated?

Penzias:

Of course! I'd like to change education. But on the other hand, I could then be kind of an inverted William Shockley, right? I could come up with my own hobbyhorse and then go around the country preaching it, whatever it is. But then that gets back to being arrogant. I don't claim to have the solution to solve education in America. All I can do is try to do my own little thing, which is that I want more bright people, I want more creative people, and I'd like to give people permission to think in simple ways on a small scale. I'm not trying to be a folk hero. All I'm trying to do is run the Communication Sciences Division at Bell labs, plus a few other things.

Lof:

But important.

Penzias:

Important to me, and maybe to the world. Maybe if I do it, other people will do it. I don't have a monopoly on this. Even if you're talking about the fifty or so executive directors, there's a lot of texture, a lot of difference. Bell Labs has a lot of diversity, and is likely to try and increase it.

Lof:

There has to be diversity.

Penzias:

There doesn't have to be. There ought to be. You can go out, as people did in the past, get all your executives from some Big Ten college, and have your organization chart and executive suite read just like the Mayflower passenger list.

Lof:

The CIA.

Penzias:

I have no idea. But any organization like that, it seems to me, has an inherent weakness. You can't do one thing. Then everything disappears. It's like, you know, somebody decided it would be nice to plant Dutch elms everywhere. No other tree. When those disappear, you don't have any trees anymore. If you have a diversity, when, one goes, things just work better. You don't even know what you need. If everybody thinks everybody else is doing it, nothing happens. When somebody gets saved on a subway stop, or somebody goes after a mugger, it's usually a person who doesn't belong there. It's often a tourist. It's often somebody who doesn't feel part of the crowd.

Lof:

My, that sky is ominous.

Penzias:

I'm looking forward to summer. I get a day like today, and bring a raincoat; it's not so terrible.

Lof:

That was terrible last week when we had to walk to work.

Social Control over Technology

Penzias:

Well, there are lots of reasons for that. The biggest reason it seems to me, is what we were talking about, the shrinking pie.

Lof:

Yes. The city is a shrinking pie.

Penzias:

The city is a shrinking pie. These people asked for twenty percent over two years. They were asking for one year's worth of inflation in two years. They don't want to go out on strike.

Lof:

No, they don't.

Penzias:

When somebody tells me that these technologists are trying to make the world more complicated, I think of the subway workers. When the pie is shrinking, everybody has their elbows out, and then it's a contest of wills, of who gets thrown out of the lifeboat last. That's not to say we have to waste our resources on meaningless things. You have to do intelligent resource allocation. We can say, “Let's stop making them, and we'll just follow our ideology blindly, and say we'll do without whatever.” We'll do without the benefits of brain power, because brain power invented napalm and the helicopter gunship. Or recombinant DNA. Would you rather do without that? All of a sudden you find that maybe recombinant DNA is going to do something nice for you, like keep you from getting certain kinds of viral diseases. You really have to decide what it is you want. It means, on the other hand, that the whole society has to make the choices. That's why you have politicians. The scientists can't do it for society.

Lof:

The scientists can only make the breakthrough.

Penzias:

They can provide the information. They can be guided. But the application and the use to which it's put is really up to the society as a whole.

Lof:

Take Mr. Nobel who invented dynamite, right?

Penzias:

Right.

Lof:

And developed these prizes because he was so guilty over the uses of dynamite.

Penzias:

Dynamite's probably killed a few people in industrial accidents. But it isn't much of a war weapon, really.

Lof:

No.

Penzias:

But basically it's the use. Certainly Genghis Khan was able to slaughter people just with swords.

Lof:

Very efficiently.

Penzias:

the time they went back to school.

Lof:

You couldn't go to the swimming pool when polio was running around the neighborhood.

Penzias:

Sure. This kind of thing has to be addressed and dealt with. Technology is providing a lot of very useful things. Those things aren't just given to you. Technology does not have an infinite capability. I think the people who are knocking technology also think of it as being infinitely powerful. Because, after all, all they have to do is ask for something, and it'll be provided to you. Which of course it won't be. Technology can't do anything just because somebody asks. There are limits. Also, in order to get the things out, you've got to put something in. You've got to support your universities, and other places. If you decide you want telecommunications, you'd better support either Bell Labs or a place like it.

Lof:

Yes, I agree 100 percent.

Penzias:

You can go for some period of time making Japanese copies of whatever. Well, they also do some basic research. I'm not knocking them, but do we really want to see everything in the world made abroad?

Lof:

I don't.

Penzias:

Well, I don't think people do either. They want very much to continue their lifestyle.

Lof:

And we're allowing basic industries to leave this country.

Penzias:

Well, so far we will have our telephone industry here. Western Electric makes its stuff here.

Lof:

Yes.

American Litigiousness

Penzias:

For our picture phone meeting service, we buy Japanese Codex. At least the basic technology is still being done here. Maybe we can afford to export our steel mills, and we may even be able to export other things. But we can't afford to export our knowledge factory. If you also export basic research, then you've given away an awful lot. Not only do you give away jobs, but also the money with which to train the next generation of scientists and the ideas. The day will come when you're just giving away everything. You'll have nothing left but a bunch of lawyers.

Lof:

Are we heading there?

Penzias:

Well, the Japanese have one-twentieth as many lawyers per capita as we do.

Lof:

I heard one-fifth.

Penzias:

It's more like one-twentieth. The English have one fifth.

Lof:

The legal profession is very self-perpetuating.

Penzias:

Yes. Somebody once said this town is too small to support one lawyer, so we have to have two. We are a very litigatious society. In part this is because we have this idea that everybody should be comfortable. If they're uncomfortable, they ought to be compensated.

Lof:

Are we demanding too much from our society?

Penzias:

I don't know. I think that's a question that has to be seriously addressed. We ought to at least take a look at the question: What is our society capable of providing? What resources do we have, and what can we do with them? How many things of what kind can we make? Then we can make intelligent choices, select among the possible.

Lof:

Are we making intelligent choices as a nation?

Elimination of Leaded Gasoline

Penzias:

Are we making intelligent choices? Yes and no. I think we made a very intelligent choice getting rid of lead in gasoline.

Children in the cities were getting lead poisoning from the dirt. Not from chewing paint, by the way. All the buildings in New York had to be repainted because the landlords had put paint on the walls? Did you ever see a child chew a wall.

Lof:

No. I've never seen a child chew a wall. I’ve never even seen one lick it.

Penzias:

Okay. But somehow people got the idea that black children are different, and that they chew walls. Isn't it absurd?

Lof:

Yes, it's absurd.

Penzias:

But the basic idea was obviously if black kids have a higher lead level, it must be from the paint. Well, of course it's the dirt from the automobile emissions that are covering the city. They're picking it up on their hands when they touch anything. They were getting it from the cars. But we didn't even make the simple connection. I'd be very surprised if there were many people who'd ever been or had any connection at all with a slum apartment or a slum neighborhood, had ever seen slum children except at a distance through the window of their air-conditioned car.

Lof:

That's true.

Penzias:

Then they made this faulty connection, and went off in a totally different direction.

Lof:

They had to repaint all the schools and everything, didn't they?

Penzias:

But it didn't make any difference! The point was it's the dirt from the automobile emissions. You get sort of a handful of lead or more out of every gallon of gasoline. With every car in New York City once a week, everybody throws the equivalent of a little paper bag full of lead right out on the street. We're just choking in our own lead.

Coal vs. Nuclear Power

Lof:

We're still choking in other things. I mean the acid rain.

Penzias:

There are people who call themselves ecologists who are arguing that power plants, coal-firing plants, are really cheaper than nuclear plants by eleven percent or something. People are arguing for coal based on something they call "ecology", which, as far as I can see, is totally nuts. That's putting it a little too strongly, but only a little bit. I don't think you ever should argue for a coal plant based on — you may want to do it for other reasons — but on a health reason? To call yourself an ecologist and argue for coal plants is really stretching things pretty far. There are certainly problems. I mean I wouldn't want to live near a coal-burning power plant. The small particles, the ones that really hurt you, can't be taken out by most processes. Those are the ones your lungs can't deal with.

There are lots of problems. We have some capability for solutions. One of them is, you do have to collect some sort of centralized data on health, because then you begin to recognize what the problems are. You can measure a radiation level of a billionth of what can hurt anybody, and even less. It's easy to measure. If you don't have good statistics, if you don't have health records, if you don't have experience, then you just get rid of whatever you can measure, rather than whatever hurts you. That's a hysterical and a stupid way to operate. It doesn't do the proper job. We cannot base our socialization’s on measurement. We have to base them on damage, on risk. That means we have to make certain assessments. We have to deal with it. That's a much tougher job. It's very easy for some guy in the government to buy a little better analytic machine and then find traces of whatever in anything. There are traces of everything in everything else.

Lof:

You're saying that we've got to go and get the basic statistics as to what these things are doing.

Penzias:

We have to understand. That means we have to get notions of probability. That also means, of course, that as a scientist you have to take the public into your confidence. What are the odds? What is the risk of this versus that? How many cigarettes would you have to smoke before you took the same risk as with the radiation from a ski trip in Colorado? You want to balance those two. If any place in New Jersey had the radiation that you get on a typical street in downtown Boulder, they would rope it off and evacuate the people.

Lof:

Really?

Penzias:

Oh, sure. A week in Boulder is much more radiation than Three-Mile Island put out.

Lof:

Yet Boulder, Colorado, is where you write to get all of your government pamphlets and everything on —

Penzias:

Oh, sure. Lots of people live in Boulder. But it's radioactive for lots of reasons. One of them is that they're sitting on a lot of uranium.

Lof:

Probably Boulder always has been.

Penzias:

Oh, sure. Madame Curie got her pitchblende right out of Boulder. The place is really radioactive. One of the reasons we know that nuclear plants don't hurt us so much is because we use statistics.

Lof:

The people in Boulder don't have a higher incidence of cancer or anything else?

Penzias:

No. If they did, nobody would live there. But they're getting about a 100 milliards extra a year.

Lof:

Why didn't I know about that?

Penzias:

The New York Times had an article. A guy went around New York with a Geiger counter and pointed out, for instance, that you get thirty or forty milliards a year extra if you just work in Grand Central Station, because granite is radioactive. Cosmic rays hit granite. You get quite a bit of radiation if you work in a granite building.

Lof:

By the same token you get quite a bit of radiation if you live in a granite mountain.

Penzias:

Somebody went into the Rayburn Office Building with a Geiger counter. He turned this thing on, and he said, "Look! If this place were a nuclear plant, it would be closed." And somebody said, "Turn that thing off before somebody makes us build a new building!" So he turned it off. It's in the Congressional Record. But we don't quantify it. It's difficult.

Lof:

Of course something like Three-Mile Island makes newspaper copy for months and months.

Penzias:

Months and months, and for one reason deservedly so. I would never have believed that a danger sign could be left on through forty-two shifts.

Lof:

That's incredible.

Penzias:

There was a story there. There was something that whoever was running the place — and by the place I mean the whole country — did wrong. There is the point that technology is not totally benign. In using technology you run risks. I think you run fewer risks than taking your chances with polio every summer. But you have to understand what those risks are. Once you understand them better, you can also deal with them.

Lof:

Well, of course polio was such a spectacular disease.

Penzias:

Sure.

Lof:

It crippled infants, children. Are you running late?

Penzias:

No.

Lof:

I have a digital clock on my clock radio, and I realized that you tell time differently.

Penzias:

Sure. You tell time very differently.

Lof:

I'd say right now it's two minutes to twelve. And if I looked at the digital clock, I would say eleven fifty-eight. The next generation won't have to learn to tell time. I got upset about that for about two minutes...until I realized that I never had to learn how to hitch a buggy.

Penzias:

Sure. That's right.

Lof:

Well, here we are in downtown scenic New York City.

Penzias:

It's a great place. I love it.

Lof:

I think I got my apartment because it's right on the same bus line as the Metropolitan Opera House.

Penzias:

Good for you! Well, we're almost done. So if you have one last question....

Lof:

No, I really don't.

Penzias:

You probably got much more than you wanted.

Lof:

Well, I found it very, very interesting. I thank you for your time.

Penzias:

I think it's important to do. If I get a certain amount of sympathy, I'm willing to take the chance that comes along with being misunderstood.

Lof:

Well, I'd really like to minimize that, and if you don't mind, I'm going to send you a draft of what I write.

Penzias:

Fine. I'd very much appreciate that.