Groundbreaking Computer Scientist Daniel Hillis

Show: FRESH AIR
Date: OCTOBER 26, 1998
Time: 12:00
Tran: 102601np.217
Type: FEATURE
Head: Daniel Hillis
Sect: News; Domestic
Time: 12:06

BARBARA BOGAEV, HOST: This is FRESH AIR, I'm Barbara Bogaev in for Terry Gross.

Danny Hillis was part of the generation of computer scientists who worked at the MIT artificial intelligence lab in the '70s. While they were supposed be working on computer vision and robotics, they would go in at night and develop fun things like word processing, e-mail, and video games.

This after-hours activity ended up producing the computer applications that are now an integral part of our lives. As a kid, Danny Hillis built a computer out of Tinker Toys, and dreamed of being a Disney Imagineer, which is what he is today.

As vice president of research and development of the Walt Disney Company, he helps design theme parks and other high-tech entertainment products. He's now written a book called "The Pattern on the Stone: The Simple Ideas that Make Computers Work."

Danny Hillis' first revolutionary contribution to computing was the very radical idea of parallel processing. He went on to form his own company called Thinking Machines. I asked him to explain parallel processing.

DANIEL HILLIS, COMPUTER SCIENTIST; AUTHOR, "THE PATTERN ON THE STONE": The idea behind parallel computers is pretty simple -- instead of trying to make one computer very fast, you use lots and lots of slow computers together. So, you sort of divide up the tasks between minicomputers.

The reason that's a good idea is because slow computers are very cheap, they're manufactured in large quantities, so, if you can just use a big pile of them it's a very cost-effective way of getting a fast computer.

It sounds like almost a trivial idea, and in fact, these days it's the way that all the very fast computers are built.

In the mid '80s, when I first started working on them and there were some mathematical reasons why people thought that it wouldn't work very well, and I guess we demonstrated that that wasn't true by actually building them and getting them to work.

BOGAEV: Give us an example of just how powerful, how fast it is, what fast means?

HILLIS: Well, one of these big massively parallel computers could be, say, a thousand or even 10,000 times faster than the computer that you have on your desk. Now, why would anybody want such a thing? Obviously, you don't care if you're running word processing to have that happen 10,000 times faster.

But what people use them for is things like designing an airplane. It would be very nice to know if the airplane is going to fly before you a test pilot into it. It turns out you can figure that out on a computer, but it requires a long time and a lot of computation.

Using a big parallel computer doesn't require such a long time, so you can try out a lot of different variations of the airplane and design a more efficient airplane, which is one other reason why airplanes use a lot less fuel now, and are a lot quieter, and so on, because they're designed using these supercomputers.

BOGAEV: If everybody was telling you that their mathematical reasons for this -- that this wouldn't work, how did you know you could prove them wrong?

HILLIS: First of all, I have to say I didn't understand any flaw in their arguments. It wasn't that I saw their argument was wrong, but what I did know is that the human brain was able to do computations much faster than a computer. And if you think about that, that's a clue that we must be building computers in the wrong way.

So, for instance, I can write a face recognition program on a computer, but it would take many hours to recognize a face and it would do it inaccurately. And yet, every two-month-old can recognize the face of his mother. So, the computer is somehow organized in a way that makes it less efficient.

And as a matter of organization, because the actual components of a computer, the transistors, are much faster than the components that your brain is built out of. So, transistors are much faster than a neuron. And yet, the brain does the job much faster than a computer.

So, that gave me the basic clue that it was possible, that somehow it had to be true you could use these components together in parallel to do computations faster. And since I was interested in doing that kind of computation -- things like recognizing images or doing common sense reasoning -- I just sort of bombed ahead without really knowing what was wrong with the mathematical arguments.

And, fortunately, it worked. And afterwards they figured out what was wrong with those arguments that said it was impossible.

BOGAEV: There's an idea now to use computers linked through the Internet, through local networks to work together. I think it's called distributive computing. It sounds similar to parallel computing. Did it grow out of it?

HILLIS: Yeah, it is a type of parallel computing. It should work in principle -- but one of the problems with using the Internet is, while it's very fast as far as people are concerned, it's not particularly fast as far as computers are concerned.

So, they spend an awful lot of time talking with each other and trying to keep coordinated, but I think that's a lot of progress in that area. So, it will definitely happen for some sort of computations.

BOGAEV: Do you think it's an important element of the future of the Internet?

HILLIS: Well, I do think that it's interesting when we start using the Internet as a computer, as opposed to just a telephone system -- a glorified telephone system. Right now, almost everything that goes over the Internet is something that's basically looked at by humans, so it's words of text or pictures or sound. It's not really something understood by the computers.

When the computers start talking in ways that are meaningful to computers -- they're actually doing this kind of distributive computation -- then we have potential for much more -- kind of interesting behavior to emerge. So, you can almost imagine the idea of the Internet being a kind of giant global brain that's thinking thoughts.

Interesting enough, that's not a new idea. It goes back from way before the Internet. As soon as electricity came out people started thinking of this and sort of realizing the possibility that we were kind of building a nervous system across the whole globe. And I think that's actually really happening now.

BOGAEV: I think your slogan for your company that marketed the massive parallel computers, Thinking Machines Corporation, was -- and correct me if I get this wrong, "We're making computers that are proud of us."

HILLIS: Right.

BOGAEV: What were you trying to convey with that?

HILLIS: Well, I think it was a kind of stake in the ground to say that we're really interested in this other way of using computers. Right now, as you mentioned, computers really are mostly used to handle old kind of media. So, they're used to handle pictures and words, which essentially came off of paper. Sounds that were humans talking to each other in their voice.

And so, they really don't understand anything about what the content of those pictures are, what the words mean. And I think that the really interesting thing about computers is, in principle, they can understand not -- they can actually understand the meaning of the words, not just the letters of the word.

They can understand the process that produces the pictures. And to me, that's the more interesting side of computers. It's the unrealized side of computers, that part of computers that we don't quite understand how to do yet, and we're just experimenting with yet. But I think it's going to, in the long run, be the much more important part of computers.

BOGAEV: What are the implications of the Internet becoming a kind of global brain?

HILLIS: I think one interesting thing about the Internet is, because it's not understood, and has the potential of having forms of behavior that weren't really designed into it, recall this emerging behavior. That somehow the whole is more than the sum of the parts.

Now, we've actually seen that in the Internet a couple of times where the behavior has been bad. It's been, for instance, the time that the virus went through the computers and shut down the Internet, in effect. And it is quite possible that forms of behavior will evolve which are beneficial and will encourage them.

And I think it's a little bit hard to predict what that's going to be, but I guess in extreme, potentially a computation could evolve on the Internet eventually that would actually start talking with us.

BOGAEV: What do you mean by talking with us?

HILLIS: Well, it's an old idea from science-fiction that somehow if you hooked enough computers together and let them evolve, and let a computation evolve on it, eventually it would become intelligent. And by talking with us, I mean that an intelligent entity would appear that could have a conversation with us.

And I'm talking far out science-fiction stuff, but I don't think there's anything that makes it scientifically impossible. And I find it a kind of plausible scenario.

BOGAEV: I thought the underlying assumption in computer science is that you can't make machines more like the human brain.

HILLIS: Well, that's a myth about computer science. The idea that we can make machines more like the human brain is a very threatening idea to people. And so, philosophers like to explain why it's impossible. There's a whole history of philosophers explaining why computers couldn't to various things and then being proved wrong.

When I first started studying computers there were philosophers who said that computers could never play chess like a human being, and that was a sort of fundamentally human attribute. There are people who say things like, computers can't make mistakes or can't surprise you.

I think that those are just kind of ways of us reassuring ourselves that we're important and there's not any real scientific basis for them. It's true that computers right now are very simple and stupid, and they're probably more like the brain of a cockroach than the brain of a human. But there's nothing inherent in them that prevents them from growing and getting more interesting.

BOGAEV: Are there practical applications being used right now or developed that would allow the Internet to think?

HILLIS: One interesting thing about progress and artificial intelligence is, every time you actually get a computer to do something, it ceases to become artificial intelligence. So, there was a time when, for instance, voice recognition was considered an amazing thing that required thinking. And now, you often run into voice recognition circuits on directory services, for instance.

There was a time when spelling correction impressed people, and now that's just something you assume computers can do. There was a time when computer vision was considered something that required intelligence, but now there's boxes all over the place that, for instance, on manufacturing lines that look at objects to see if they meet the specifications.

So, I think what we'll see for a long time is this kind of gradual progress. You'll see applications recurring that you won't see as breakthroughs, just seem like the sort of normal progress of computers. And increasingly, when you see something happen it may be happening with several computers working together across the Internet.

I remember one time recently I was demonstrating a simple artificial intelligence program to someone, and they said, oh, my computer does that. And I said, well, I'm very surprised that your computer does that because I think that this is the only computer that can do it. And she said, no, no I'm sure my computer does that.

And I traced it down, it turns out that their computer, in fact, did do that but the way that it did it is their computer hooked through the Internet to my computer and asked my computer to do it. So, they didn't even know that they were using my computer.

But -- because it was sufficiently invisible. So, I think you'll increasingly see things that are happening like that, but you probably won't even recognize them as happening across the Internet.

BOGAEV: Danny Hillis' new book is "The Pattern on the Stone: The Simple Ideas that Make Computers Work." We'll talk more after the break.

This is FRESH AIR.

BREAK

BOGAEV: Back with Disney Imagineer and computer scientist Daniel Hillis. You've written a lot about the, I suppose, the convergence of biology and technology -- that we've reached a point where technology can take on almost biological processes like evolution. I'd like you to talk more about that.

HILLIS: I think it used to be that technological artifacts were simple, controllable things. And biological things were complicated and totally un-understandable, and you just had to live with them. Both of those things are changing.

Technology is becoming much more complicated and it kind of has a life of its own, and biology is actually becoming much more controllable -- we can actually gene splice, and control, and things like that. So, I think the distinction between biological things and technological things is actually starting to blur.

A good example of that is evolving computer programs. It used to be the only way of designing a computer program was a kind of engineering approach -- you wrote each line of code like you were writing a novel. Well, now it's possible to actually evolve a computer program by a method that's very analogous to biological evolution.

So, what you do is you create inside a computer an environment, and you fill that environment up with computer programs which can start out to be completely random computer programs, and you just let them live in that environment and you select them according to how well they do whatever you want to do.

So, if you want them to arrange things in alphabetical order, you just allow the things that tend to arrange things in alphabetical order to survive, and you kill the rest of them. Then you actually allow the ones that survive to breed with each other and produce children that inherit their traits. And in a computer, you can actually simulate generations and generations of these programs evolving in this environment.

And after a few thousand generations, sure enough, they're very good at arranging things in alphabetical order or doing whatever it is you're selecting them for. The interesting thing about that way of producing a computer program is when it's finished you can look at the program. You can't describe how works, but you can just see how it works.

So, it is much more like a biological organism that you can see what it does, you can kind of select it a bit -- you can do selective breeding to get it to do more what you want. But you really don't know how it works. So, in a very literal sense in which it is almost like a biological organism. I think you're going to start seeing a lot more of that happening.

BOGAEV: When you say that you no longer really know how it works, is that a scary specter for you or only for us laymen listening to you talk about it?

HILLIS: Well, it's interesting -- if I produced a program like that I would actually have more faith that it works than I would if some very smart programmer programmed it. That sounds crazy, but if you fly an airplane, would you rather fly with a human pilot or would you rather fly with an autopilot designed by a bunch of engineers?

I would personally rather have the human pilot. Well, I don't know how the human pilot works, I do know how the auto pilot works. And I do know how the engineers work which is what scares me. It turns out understanding things doesn't necessarily mean that that's the most robust way of doing it.

Now, I have faith in the human pilot because I know that that pilot comes from a long line of people that survived. So somehow there's generations of people who had enough survival instinct to do whatever it was necessary to stay alive. So, I trust somehow that that pilot will do whatever is necessary, otherwise they wouldn't be in the plane.

So, I trust the process, in some sense, that created the pilot -- the training program, the hiring process, and so on. In the same sense, when I evolve a program I trust the process. I know, to take my example, that at the end of 10,000 generations I have a program whose life depended on sorting things correctly, of putting things in alphabetical order correctly. It comes from a long line of programs that survived by doing that, that have been tested in lots of possible ways.

Just to take it another step, you can actually involve not just the programs but the test regimes that test the programs. So, normally when you test a computer program, somebody, usually the same person that writes a program, thinks of a few tests to see if it's working. Well, they're not likely to think of all the odd cases.

So, for instance, people who were writing payroll programs in 1950 didn't think of what happened when you got to the year 2000. If they had thought of those cases they probably would have written the program correctly. When you evolve a program and evolve the test cases along with it, what happens is tests evolve that find bugs in the program that you never would have thought of.

So, I'm convinced that if you it used evolutionary techniques to produce a lot of software that controls -- keeps track of what date it is, then we wouldn't have a lot of the millennium bug problems that we have now.

BOGAEV: Do you think the millennium bug hoopla is exaggerated?

HILLIS: I have to be very careful about how I answer that question because I do think that people who depend on computers should be spending some energy to see what will happen with those computers when the year 2000 comes. If you have anything with a microprocessor on it you should ask yourself -- does that microprocessor know what date it is or depend on what date it is.

So, I do think there's a certain amount of responsible action that ought to be taken. Having said that, yes, I do think the consequences of the millennium bug have probably been exaggerated.

It's to the experts' advantage to say how terrible it might be. If you're writing a Sunday supplement it's good to start with the lead about airplanes falling out of the sky or pacemakers stopping or something like that to grab people's attention, but I think those are actually pretty unlikely.

In fact, I think what will happen is that a lot of pretty trivial things. So, it may very well be that certain bills will arrive late or some insurance policies get canceled by accident. And some of that will cause real inconvenience - if welfare checks are late I'm sure that would cause a real problem. But I don't think it's a world-stopping kind of thing. Society is pretty resilient, and I think it will be a glitch and we'll get past it.

I think right now there's almost a danger that we've oversold it, and there's a danger that we're -- we will create a panic. So, right now, I think one of the worst scenarios that people will be so afraid of the millennium bug that there'll be a big stock market crash; everybody will pull their money out. So, I think there's a danger of irrational fear about it also.

Can I say one more thing about it?

BOGAEV: Yeah, sure.

HILLIS: I think -- to me, one of the things that's most interesting about the Millennium Bug is that people are so interested in it. You know, of all the possible bad things that can happen, why is it that this one has kind of captured people's imagination? And I think the reason for it is that it's a hint for a lot of people that things are really more complicated than they've understood them to be.

So, it's the first time they've realized that the experts don't actually know what's going on. There's not somebody back in the computer room that really understands how this computer program works. And that's a sort of shocking concept, that we have now made technologies so complicated that we don't understand it in detail. So, I think a lot of people's interest in the Millennium Bug is really their interest in that.

BOGAEV: Danny Hillis is the author of "The Pattern on the Stone." We'll continue our conversation in the second half of our show.

I'm Barbara Bogaev, and this is FRESH AIR.

BREAK

BOGAEV: This is FRESH AIR, I'm Barbara Bogaev.

Let's get back to our conversation with computer designer Danny Hillis. He pioneered the concept behind the first supercomputers. Hillis now works as an Imagineer for Disney. His new book is "The Pattern on the Stone: The Simple Ideas that Make Computers Work."

You've been working on something called the "Millennium Clock," or the "Long Nail" project. Why don't you tell us what it is and what's the thinking behind it.

HILLIS: Well, I got very interested in the problem that the future seemed to be shrinking. When I was a little kid growing up in the '60s, people thought of the year 2000 as being the future, and they really tried to imagine what it would be like. But now it's 1999 or 98 and the year 2000 is still two -- just two years away, and yet it's still what people think of as the future.

People aren't thinking of what happens in the year 2050. So, it's as if it's a kind of a mental block that people can't think beyond the millennium. And I think that that kind of short-term thinking causes a lot of problems to seem like they're impossible. So, people basically only think of a problem as solvable if it's solvable in two or three years.

The story that really got me thinking about this is a story I heard about Oxford University. They have a college -- they're called New College, which is called new because was made in the 15th century -- started in the 15th century.

The common room of New College has these big oak beams that needed replacing, and this was fairly recently that they needed replacing. And these days, you can't just go down to a lumberyard and buy a 40 foot oak beam.

So, they started looking around trying to find a source of giant oak trees that they could make these beams out of, and what they discovered was that, sitting right next to New College was a growth of oaks that had been planted for the purpose of replacing those beams in a few hundred years when they wore out.

And that really struck me that that was a very different way of thinking about the world than we have today. I mean, you can't imagine today, somebody building a building and planting some acorns so that there'll be oak trees to fix the building in a few hundred years.

And I started thinking about how can we regenerate that kind of long-term thinking. What the Millennium Clock project is about is actually trying to undertake a project that in some sense doesn't get finished for 10,000 years. So, what I'd like to do is build a very big, very slow clock that ticks off the millennia as they go by.

So, imagine that this clock will strike two in the year 2000, but then it will strike three in the year 3000, four in the year 4000, and so on. I find that thinking about this project stretches out your sense of time, you start thinking, well, what's the world really going to be like in the year 3000 when the clock strikes three?

For example, if I put the clock in San Francisco, is it going to be underwater? Or if I make the clock depend on people winding it, you know, what will those people be like? So, it starts a very different chain of thinking, and all of a sudden the year 3000 starts feeling like a real year that you actually have to think about and have some responsibility for.

BOGAEV: You've said that people either -- when you talk about it, they either get it right away or they don't get it at all. What do people who don't get it say?

HILLIS: I would say that -- I would say that probably eight out of ten people that I explained this to, sort of look at me like I'm crazy, and politely try to get away as fast as possible. But occasionally, I think you hit somebody who says -- yeah, I've been worried about that people haven't really been thinking about the future. And I see how you would want to stretch it out and really undertake a project that takes a long time.

You know, there are always people around that like stretching the collective mindset a little bit, and this certainly does that. But when you do that, you have to put up with the fact that most people aren't going to understand what you're doing. So, in fact, there's also a lot of people that really resonate with this idea.

And at first, I just started talking about it almost as a hypothetical and kind of a fantasy. But found that so many people really got interested in it that I started thinking of it as a serious possibility. In fact, people would start writing out checks to me and handing them, saying, well, here's my donation to help build a clock.

And so, I didn't quite know what to do these checks. My friend Stewart Brand (ph) actually started a foundation and said, OK, well, we're really going to do this project, let's put together this foundation, which is the Long Nail Foundation.

And let's make the purpose of the foundation is to get people to think over the long term. And the clock will be our first project. It's attracted a very interesting kind of eclectic bunch of people every place from businesspeople to science fiction writers, to artists who, for some reason, this idea struck them as worthwhile and interesting.

And there is now a very serious effort to build a clock. I mean, I'm actually designing it, and we're looking for a site for it, we're raising funds for it. And that's only one of several projects that the foundation is doing now.

BOGAEV: Can you say what a potential site would be?

HILLIS: Well, I guess my image of the clock is that it's someplace far off and remote and hard to get to. So, that you really have to make a commitment if you want to go visit it, so maybe it's up at the top of a mountain or in the middle of a desert.

So, you might hear about it, and most people will hear about it and never want to go there, but if there -- somebody actually makes the commitment to go there and look at it, maybe wind it or watch it tick, or whatever it is you do when you get there -- then I think that whole process, that whole kind of pilgrimage is more likely to be a transformative experience for them that if you just, say, put in the middle of a city someplace where people would see it every day as they walk to work.

BOGAEV: Daniel Hillis, thanks very much for talking with me today.

HILLIS: Thank you very much.

BOGAEV: Danny Hillis. His new book is "The Pattern on the Stone."

Coming up, the roots of anti-gay violence. This is FRESH AIR.

This is a rush transcript. This copy may not
be in its final form and may be updated.

TO PURCHASE AN AUDIOTAPE OF THIS PIECE, PLEASE CALL 888-NPR-NEWS

Dateline: Barbara Bogaev, Philadelphia
Guest: Daniel Hillis
High: Daniel Hillis is one of the world's leading computer scientists and is the designer of the world's fastest computer. He's also vice president and Disney Fellow at Walt Disney Imagineering. His new book is "The Pattern on the Stone: The Simple Ideas that Make Computers Work."
Spec: Computers; Media; Walt Disney Company; Daniel Hillis

Please note, this is not the final feed of record
Copy: Content and programming copyright 1998 WHYY, Inc. All rights reserved. Transcribed by FDCH, Inc. under license from WHYY, Inc. Formatting copyright 1998 FDCH, Inc. All rights reserved. No quotes from the materials contained herein may be used in any media without attribution to WHYY, Inc. This transcript may not be reproduced in whole or in part without prior written permission.
End-Story: Daniel Hillis

Show: FRESH AIR
Date: OCTOBER 26, 1998
Time: 12:00
Tran: 102602NP.217
Type: FEATURE
Head: Arthur Dong: In the Wake of Matthew Shepard
Sect: News; Domestic
Time: 12:32

BARBARA BOGAEV, GUEST HOST: Matthew Shepard was murdered earlier this month by two men in Laramie, Wyoming, who allegedly targeted him because he was a homosexual. Twenty years ago documentary filmmaker Arthur Dong was attacked by four teenagers in San Francisco because they thought he was gay.

In an effort to understand what motivates anti-gay violence, Arthur Dong interviewed convicted killers of gay men about what drove them to murder. His 1997 documentary, "Licensed to Kill," features these prison interviews along with videotaped confessions. Jeffrey Swinford (ph) is one of the men who appears in Dong's film. Let's listen to a scene in which Swinford describes his crime.

(BEGIN AUDIO CLIP OF FILM -- "LICENSED TO KILL")

JEFFREY SWINFORD, CONVICTED MURDERER: I'm not going to let anybody get out on any of my friends.

ARTHUR DONG, FILMMAKER: Did you know what was going on?

SWINFORD: No, not at first. Bobby, he kind of -- he kind of jump to his feet and he said, mother (expletive), I'm tired of you hitting up on me. Like that, right there, and I thought, what the (expletive). And he hit this one dude, and I said, man, what's up? And he said, man, these are mother (expletive) fags. Like that, right there.

And I -- you know, we got to fighting and stuff. Dude back up into this corner with a knife and, you know, he said, I'm HIV-positive. Like that, right there, you know, and we're like, what the (expletive deleted) you know, what are you doing, trying to, you know, spread AIDS or what, mother (expletive), you know.

It just pissed us off more. It happened so fast, you know, stuff happened so fast in a situation like that. You know, it probably really didn't -- I really probably shouldn't have done it.

But if a man ever hits his hands on me again that's just where I'm going to have to take it. This guy, you know, he was trying to make homosexual advances on my best friends, he's dead because of it. This stuff, you know, this just, you know, like one big opinion and what you make of the opinion. I mean, I don't have any opinion whatsoever for homosexuals except they all ought to be taken care of.

BOGAEV: Jeffrey Swinford is just one type of gay-basher portrayed in Arthur Dong's film. I asked Dong if he noticed anything familiar about the two men accused of murdering Matthew Shepard from reading about them in the news.

DONG: There are certain patterns that I saw in my research for the film -- for my film -- that I saw repeated in what was being reported. For example, the whole notion of wanting to rob Matthew as a motive, and then being angry because he flirted -- supposedly flirted with them. And the outcome being the attack -- the beating of Matthew. That was so common in terms of the motives given by perpetrators of these kind of crimes.

BOGAEV: What were your first thoughts when you heard about the murder of Matthew Shepard in Wyoming?

DONG: I was numbed. I've been researching this topic for 20 years, culminating with the making of the documentary, and during that whole process -- I think in order to do my job as a filmmaker, and you know, I put quotes around that word "job" -- I had to put up an emotional barrier because if I didn't do that I would have been hysterical making a film like this because it was so personal and painful at the same time.

And during the making of the film and during the promotion of the film last year, when it was out commercially and also on PBS, I had this barrier, and subconsciously, when Matthew's death hit the news -- when his beating hit the news, not even his death, it just stunned me.

I said, wait a minute, I've been dealing with this issue and I thought I did something to contribute to the dialogue and debate. And here it is, smacking me in the face, and it broke those barriers just completely.

BOGAEV: When you talk about hate crimes, the implication is that hatred really, it lies at the heart of what motivates violence against a group. But were all these men truly motivated by hatred of gays. I'm thinking Corey Burley who said, among other things, that he was a tough guy, he was robbing people, it was just a going thing, that was, let's go rob a homosexual -- you get it in your head that they're are weak, you just do it to do it.

DONG: Well, I think you have a point there. It may not be just pure hate or hatred of homosexuals. I think a lot of it has to do with themselves. With Corey Burley, there were two factors I think that are important to point out.

First, he, as a young child was beat up a lot and pushed around a lot, he was you know, kind of weak, and he was teased a lot. And he says, well, enough of this, I'm going to be a bad guy. So he found out that, well, he can be really bad by beating people up and everybody started being afraid of him.

So, this was his process, his evolution, in terms of his bringing up. That was one thing, to prove that he was a man.

And another thing was that he was very affected by the media. He didn't really know gay people, as far as he knew. But he had heard that they were weak and you can just do things to them and they wouldn't fight back. And that was the thing, they were easy prey. Here -- what that points out is that the media has a -- is a factor in the perception of what gay people are to those who are not familiar with that community, and Corey Burley bought into that.

BOGAEV: It really surprised me that watching your movie, I was struck by how pathetic the men are. That some had been, or claim to have been, raped as children, that they were very confused about masculinity, about their sexuality. What did they tell you about the psychology of abuse? How it turns men towards violence?

DONG: Well, you know, the whole issue of child abuse and molestation comes up a couple of times in "Licensed to Kill," and in one case, with Donald Aldrich, he claims that he was abused as a child -- sexually abused, but in the same breath he also says that his friend's daughter was raped by a gang of homosexuals. So, you kind of, you know, take that with a grain of salt -- that statement.

However, we have a person in the film, William Cross, who claims to have been abused as a child. And that in his crime, he was motivated by the sense of rage when he was touched by another man, particularly an older man. And this physical touching enraged him so much that he just totally lost control and stabbed the victim seven times.

And he believed that it wasn't so much a fear of danger -- of physical danger, of been attacked by this older man. He was a 55 year old man, and William Cross, I think, was in his mid 20s time of his crime, but it was more a fear of losing his masculinity -- of being raped once again. And he made that distinction very clear.

I think in the making of this film, after the gathering of interviews and the editing process, I knew that child molestation and the issue of child molestation within the gay debate, let's say, was very touchy. And I sought out advice on this. I said, how do I deal with this? What kind of message are we sending out?

And in speaking to many experts and psychologists on this topic, they say, well, you know, actually when you look at the statistics -- and of course we should understand statistics are what we can always hold them to what we want them to reflect -- but many convicted murderers in prison have a history of child abuse. And that's not unusual, so why would it be any more unusual for a killer of a gay man to have that history as well?

BOGAEV: These guys are really quite casual talking about their crimes and their ideas about homosexuals. Did they know that you are gay?

DONG: Well, it was sort of a don't ask, don't tell situation. They never asked and I never told, really. And, you know, I think one can say, well, gosh, is that a cop out? You should have, you know, said something.

First of all, I was very, I think from the very beginning of my communication with each one of these men, I was very forthright on what the project was about, in that it was a film about gay murders taking it from the point of view of the perpetrators.

I also believe that taking it from a journalist point of the new, I thought if I did tell them that I was gay, it would provoke them one of two ways: one, they may gave me answers to please me, let's say, and not speak negatively about sexuality. Or they would try to provoke me and just start spewing out all these hateful statements, and I didn't want either of those things to happen.

I wanted, as much as possible, for them to really speak what they felt -- to speak from the heart. So, I was kind of like a nonperson in many ways, and really more like a sponge for their stories.

BOGAEV: One of the dialogues that's grown out of the Matthew Shepard murder is about the effectiveness of hate-crimes legislation. There are no hate crimes laws on the books in Wyoming. What are your thoughts about whether hate crime laws make a difference? Would they have made a difference in any of the cases of the men that you interviewed?

DONG: I think hate-crime laws are useful in sending a strong message to the public that this is not a type of crime to be tolerated. I think anything that discourages these types of crimes from happening is useful at this time. These are desperate times and we need all the help we can get.

My feeling -- my point of view in my work is that I want to get to the roots of the problem.

I want to get to the point where even thinking about this type of crime doesn't even -- the thinking doesn't even exist, that's where I want our society to be at. And that is my purpose behind "Licensed to Kill," is to look at the root of this behavior and seeing if we can do something about it from the very start, as opposed to doing something about it after the fact.

In many ways, you know, I don't even want that thought -- that anti-gay thought to exist in people's minds, no less the thought of committing a violent crime against a person just because of the sexual orientation.

BOGAEV: After getting to know the men that you interviewed pretty well, are you more or less optimistic about the ability to prevent violence against homosexuals?

DONG: I'm less optimistic. I thought I'd be more optimistic actually, but having done this film and having met these men, and having researched other cases that aren't in the film, I can see the widespread contempt for homosexuality out there. I can see, I mean, I've come face to face with it.

I think many of us come face to face with it every day, but we may not acknowledge it or we may not see it because it's not upfront. But I saw it upfront and I see it and so many places -- big cities, small cities, rich communities, poor communities, black communities, white communities, Asian communities, it's everywhere. And it's so deep-rooted in our society that I'm very pessimistic.

At the same time, I'd have to add that I'm optimistic because I see many steps being taken, I see a lot of baby steps that I hope will add up to that big boom -- that big explosion of equality for gay people as part -- as an integral part of our society.

BOGAEV: Arthur Dong's documentary is "Licensed to Kill."

Coming up, a review of the 50th anniversary edition of Thomas Merton's autobiography. This is FRESH AIR.

This is a rush transcript. This copy may not
be in its final form and may be updated.

TO PURCHASE AN AUDIOTAPE OF THIS PIECE, PLEASE CALL 888-NPR-NEWS

Dateline: Barbara Bogaev, Philadelphia
Guest: Arthur Dong
High: In light of the recent beating death of gay student Matthew Shepard, a discussion of anti-gay violence with documentary film maker Arthur Dong. In his 1997 film "Licensed to Kill," Dong interviews convicted murderers of gay men and asks them why they committed such acts. Twenty years ago, Dong himself was the victim of a gay bashing. Dong is a Peabody Award winner and Oscar and Emmy-nominated independent filmmaker. His other films include "Coming Out Under Fire" and "Forbidden City, U.S.A."
Spec: Arthur Dong; Murders; Movie Industry; Homosexuality; Sexuality; Violence; Crime; Justice

Please note, this is not the final feed of record
Copy: Content and programming copyright 1998 WHYY, Inc. All rights reserved. Transcribed by FDCH, Inc. under license from WHYY, Inc. Formatting copyright 1998 FDCH, Inc. All rights reserved. No quotes from the materials contained herein may be used in any media without attribution to WHYY, Inc. This transcript may not be reproduced in whole or in part without prior written permission.
End-Story: Arthur Dong: In the Wake of Matthew Shepard

Show: FRESH AIR
Date: OCTOBER 26, 1998
Time: 12:00
Tran: 102603NP.217
Type: FEATURE
Head: 50th Anniversary: "The Seven Story Mountain"
Sect: News; Domestic
Time: 12:55

BARBARA BOGAEV, GUEST HOST: A 50th anniversary edition of Thomas Merton's renowned spiritual autobiography, "The Seven Story Mountain," has just been published. Book critic Maureen Corrigan has this reexamination.

MAUREEN CORRIGAN, BOOK CRITIC: In the same week that the anniversary edition of Thomas Merton's "The Seven Story Mountain" arrived in the mail, I also received a new book called, "Games for the Soul: Forty Playful Ways to Find Fun and Fulfillment in a Stressful World."

There you have it, two book titles that attest to the radical transformation in popular American spirituality that's taken place over the 50 years since Merton first published his autobiography. Merton's Dante-esque title suggests effort -- the fearsome fight of the soul to pull itself up and away from the Inferno.

In contrast, "Games for the Soul" assures us that the path to spiritual growth does not have to involve discipline, sacrifice, and long hours of meditation. In fact, it can be a fun and creative process.

Merton would not have played well on Oprah, where every weekday New Age swamis preach self-love. Merton is more Jonathan Edwards then Marianne Williamson. For me, re-reading "The Seven Story Mountain" after decades of exposure to cuddly Christianity was to re-enter a pre-Vatican to world that is now so alien it might as well be medieval.

The no-excuses Catholicism that Merton arduously converts to in "The Seven Story Mountain" dictates that, if you want to make it to heaven, you have to do it the old-fashioned way -- you have to earn it. When "The Seven Story Mountain" first appeared in October of 1948 it was hailed as a modern-day version of St. Augustine's confessions.

The twentysomething Merton, who had become a Trappist monk a few years before he wrote the book, tells the story of his birth in France to two progressive artists, and of his subsequent childhood in Europe and America where Hollywood movies, he sourly recalls, became the family religion.

As a desolate young man, Merton was expelled from Oxford and wound up in New York at Columbia University where he became part of an elite literary circle around Mark Van Doren. Throughout these years, Merton tells us, he felt the obscure need for faith.

While at Columbia, he converted to Catholicism and then decided to go whole hog and become a priest. There's a deadpan funny passage here where Merton describes walking with his girlfriend around Riverside Drive as he's obsessively reading seminary brochures.

Towards the end of his very long struggle, Merton attains a separate peace when he is accepted into the monastery of our Lady of Gethsemane in Kentucky. The thing that's most striking about "The Seven Story Mountain" is its author's tone. Merton is not a warm guy, he's intellectually arrogant and scornful of society, and of his readers, whom he addresses as, you who are now what I once was, unbelievers.

His conversion wasn't easy, and he doesn't make the process of reading about it a snap either. You have to plow through thick passages of theology if you're going to make it to the promised land with him. But Merton always takes more than he dishes out. He contemptuously calls his earlier self stupid, dry, full of dust and rubbish.

A new introduction for this edition written by Robert Giroux (ph), Merton's editor and friend, illuminates one other reason for Merton's self loathing -- as an undergraduate at Oxford he'd fathered an out of wedlock child. Unlike the reformed rake, St. Augustine, however, Merton had to contend with the Trappist sensors who ordered him to keep the sexy bits out of his manuscript.

Critics say that "The Seven Story Mountain" became an instant classic in 1948 because a world hardened by war welcomed its stern message of salvation. I don't know how Merton's tough-love autobiography will strike today's readership, anxiously tottering on the edge of the next millennium. Personally, when I free-fall into the cosmic unknown, I'd rather be clinging to the rough monk's robes of Merton the Militant than to the flimsy gossamer wings of those wimps on "Touched by an Angel."

BOGAEV: Maureen Corrigan teaches literature at Georgetown University. She reviewed the 50th anniversary edition of Thomas Merton's autobiography "The Seven Story Mountain."

This is a rush transcript. This copy may not
be in its final form and may be updated.

TO PURCHASE AN AUDIOTAPE OF THIS PIECE, PLEASE CALL 888-NPR-NEWS

Dateline: Barbara Bogaev, Philadelphia
Guest: Maureen Corrigan
High: Book critic Maureen Corrigan reviews the 50th Anniversary edition of Thomas Merton's spiritual autobiography, "The Seven Story Mountain."
Spec: Religion; Thomas Merton; "The Seven Story Mountain"; Media

Please note, this is not the final feed of record
Copy: Content and programming copyright 1998 WHYY, Inc. All rights reserved. Transcribed by FDCH, Inc. under license from WHYY, Inc. Formatting copyright 1998 FDCH, Inc. All rights reserved. No quotes from the materials contained herein may be used in any media without attribution to WHYY, Inc. This transcript may not be reproduced in whole or in part without prior written permission.
End-Story: 50th Anniversary: "The Seven Story Mountain"
Transcripts are created on a rush deadline, and accuracy and availability may vary. This text may not be in its final form and may be updated or revised in the future. Please be aware that the authoritative record of Fresh Air interviews and reviews are the audio recordings of each segment.