Restoring Lawrence: An Interview with Robert Harris
by
Gary Crowdus and Alan Farrand
Cineaste, Vol. XVII, No. 2 Copyright, 1989
The restoration of Lawrence of Arabia--"Produced and Reconstructed by Robert A. Harris and Jim Painten," as the official credit reads--involved not only a worldwide search for surviving materials, but also turned into a "a two year odyssey," Harris explained, "encompassing months of exhaustive research, detective work, and a touch of modern archeology." Harris (who also collaborated with Kevin Brownlow on the 1981 restorations of Abel Gance's 1927 silent epic, Napoleon) recently talked with Cineaste about some of the difficulties involved in the Lawrence restoration.
Cineaste: What condition would Lawrence of Arabia be in today without this restoration?
Robert Harris: Honestly? It would be junk. The negative was dried out, warped, scratched on both the base and emulsion side, and the sprockets were torn. I was told the negative had never been cut, and they had full black and white separations.* It turned out, however, they'd also cut the separations. The separations had been made dry gate, so any scratches on the negative showed, and there was differential shrinkage, so we couldn't align the color on all reels.
Q: Who would the original materials have been cut?
A: Who knows? The irony is that Lawrence was printed in a process called Auto-Select which negated the need to cut it, because it was a selective process and even fades and dissolves could be easily moved. The negative was originally single strand so you could make short versions, long versions, and foreign versions. There was no need to cut it, but they did , probably to save time scanning back and forth to select shots. The Auto-Select process caused a lot of wear and tear on the negative. It had been run some 250 times. They even ran it to make replacement footage sections!
When we tried to make our interpositive, which is a duplicating positive from which you make a dupe negative, the splices started tearing right through the shots. The chemistry of the film had changed, the cement on the splices was going, and the edges were loosening up. In the middles of the process we stopped and negative cutters at Metrocolor used thousands of pieces of mylar tape to repair every splice that looked like it was ready to go.
That negative broke so many times, and every time it broke we'd have to pull in more cans of negative, trying to go back to the separations, or go for alternate shots or trims if that didn't work. We had t go to dupe for some sections. We were supposed to have the interpositive off by the end of May, but we finally got it off in October.
I called Dawn Steel (President of Columbia Pictures) and said, "Look, understand that you are never going to be able to use this camera negative again. Right now I think it will run once more, it's taped to the hilt." She said, "You meant that if anything happens to the interpositive, we've got nothing?" I said, "Yes." She asked, "How much will it cost?" I said, "$52,000." So she said, "Make us another interpositive." Dawn has given me 100% support. She did it for history and to protect the studio's property. Although the project has more than paid for itself, that never became part of the original equation.
Q: What was the condition of the sound materials?
A: Two magnetic tracks survived but were mislabeled. The one marked 'print master' turned out to be a copy, and the one marked as a copy, a protection dub, actually turned out to be the original four-track magnetic master. It had splices in it where they had lifted scenes, and the oxide was flaking off, so it wouldn't run. The 600,000 feet of original mixing elements, including Maurice Jarre's original music, was all junked in 1975.
What we finally had to use as our original was fourth generation materials. The original mixing elements, you see, are first generations, and some thirty to thirty-five effects tracks are mixed down to one, which is second generation. Your master is third generation, and your print master--which is used to sound 35mm and 70mm prints--is fourth generation. Once we realized that would run, we made a fifth generation Dolby-A encoded full coat, as good as we could get. What your hear today is seventh generation, and, had we not used Dolby to suppress the noise, it would have sounded like a pot of mush.
Q: It sounds better than I remember it originally. A: You can thank Goldwyn Sound for that, and you can especially thank Richard Anderson and Greff Landaker, both of whom worked on Raiders of the Lost Ark. The gunshots in Lawrence, for example, have been basically modernized, they're re-equalized.
Q: Were there any problems in restoring the colors?
A: Not really, because Eastman camera negative is great. Over the years the color had shifted, it had faded just a little bit. To determine the proper timing, I was able to find one of the few surviving Lawrence 35mm Technicolor dye transfer prints, which don't fade at all. Bill Pine, the timer at Metrocolor, did not use a Hazeltine but timed every shot by eye. Everyone at Metrocolor did an extraordinary job. They did it for the love of the film and David Lean. David, because of their extraordinary quality, considers them to be his laboratory.
Q: Apart from a few minor imperfections, it looks fabulous.
A: There are imperfections, and they disturbed me originally, but not any more. David's feeling is that if you don't have some imperfections, people won't believe it's a restoration. It's a phoenix risen from the ashes, that's what it is.
Q: After you had spent more than a year restoring Lawrence to its original 222-minute length, how did you feel when Lean wanted to cut it?
A: Actually we restored it to 223 minutes because we added material that was never in it before.
Q: What was that?
A: It was all in the beginning of the second half, the scene between Arthur Kennedy and Alex Guines, things like Kennedy saying, "I understand you've been given no artillery."
After we'd completely reassembled the film, David screened it and said, "It's great, now let's cut it." And I said, "Huh?" He saw that I was not happy, so he puts his arm around me, walked me away, and said, "Look, do you want to have the perfect restoration of the December 9th, 1962 premiere, or do you want to make Lawrence of Arabia the best film we can make it? If you want to do that, let's sit down and take a scalpel to it, let's do the director's cut," and that's what we did.
Q: What material did he cut?
A: The only scene that he questioned was the first officer's club scene, with the pool table. He felt that it showed a side of Lawrence's personality you probably shouldn't see at that time. Other than that, he made only editorial trims. No entire scenes were cut. He tightened it as an editor, not a director, and most of the things he trimmed needed to be trimmed. For example in the second attack on the Turkish train, where Auda gets his horses, there's a shot where Anthony Quayle pops out of the top of an armored vehicle and fires a flare gun. In all the previous versions, he comes out of the vehicle, he tries to cock the pistol but it doesn't cock, so he cocks it again, and then finally fires it. But you don't need to see him not being able to cocks he pistol. That's gone now. There were also trims in the first scene with Lawrence and Dryden; there was a bit after they left Murray's office and before you see them walking down the corridor. It didn't belong in there.
Q: We were sorry to see that the balcony scene between Allenby and Lawrence wasn't completely restored.
A: There were two problems with it. First, the dialog didn't work, the dubbing of Jack Hawkins's voice was just too different. Charles Gray did an absolutely Herculean job, but, impassioned as he was, the voice just didn't work perfectly. Second, because of the way the film originally cuts, there's a question of where Dryden is all this time.
Q: Do you think the deterioration of the Lawrence negative was more a result of the particular printing process, the improper storage of materials, or the natural ravages of time?
A: If I had to weight one more heavily than the other, it would be the overprinting first, and quality of storage second. Ravages of time? Not really. If properly stored, those materials should not have gotten to that point.
Q: If this can happen to a classic film like Lawrence, what does that say about the preservation of the average, run-of-the-mill film?
A: It's not so much the quality of the materials, it's the way they're handled, putting them back together to make new negatives. One of the biggest problems in the studios is that they don't know what their own films are supposed to look like. And if you don't know what it's supposed to look like, you have no idea what the timing is.
Q: Has there been a decline in the level of technical expertise in the labs?
A: No, the labs are fine. It's a matter of budget and the studios are pushing their people too quickly. If the studios want great materials, t hey have to give their people the necessary time and the necessary budget and the access to original materials in order to see what the film looked like originally. They have to allow them to spend a few extra dollars, we're not talking about a lot of money, to make a proper internegative and proper interpositive, rather than just a interpositive.
Q: What's the present condition of Lean's Bridge on the River Kwai?
A: The have the camera negative. They have black and white separations which hopefully will register but you don't actually know until you try them out. People have supposedly made quite unquote 'preservation' materials but hen not tested them. You don't have preservation materials until you're made a dupe negative off your separations. I think Kwai could be gorgeous, it could be blown up to 70mm beautifully and the tracks could be reworked.
Q: We understand that you will be supervising the preparation of a laser disc of the restored Lawrence.
A: It'll be a letterbox version and will have lots of extra material on it. The entire script may possibly be put on the laser disc, plus the original trailers, and samples of daily continuity and mixing/dubbing sheets. We also have access to 65,000 feet of 16mm behind the scenes footage on the making of Lawrence. We also have footage of the New York and Los Angeles premieres, footage of Peter learning how to ride a camel, Albert Finney's screen test as Lawrence, and actual shots of Lawrence in Arabia in 1918. It's going to be six sides, at least. Hopefully it'll be out for Christmas and will probably retail for $125 or so. It's the ultimate gift for the cinephile. [The final boxed version contained the following: Production photos and a booklet detailing the twenty minutes of restored footage.--ed.]
The restoration of Lawrence of Arabia--"Produced and Reconstructed by Robert A. Harris and Jim Painten," as the official credit reads--involved not only a worldwide search for surviving materials, but also turned into a "a two year odyssey," Harris explained, "encompassing months of exhaustive research, detective work, and a touch of modern archeology." Harris (who also collaborated with Kevin Brownlow on the 1981 restorations of Abel Gance's 1927 silent epic, Napoleon) recently talked with Cineaste about some of the difficulties involved in the Lawrence restoration.
Cineaste: What condition would Lawrence of Arabia be in today without this restoration?
Robert Harris: Honestly? It would be junk. The negative was dried out, warped, scratched on both the base and emulsion side, and the sprockets were torn. I was told the negative had never been cut, and they had full black and white separations.* It turned out, however, they'd also cut the separations. The separations had been made dry gate, so any scratches on the negative showed, and there was differential shrinkage, so we couldn't align the color on all reels.
Q: Who would the original materials have been cut?
A: Who knows? The irony is that Lawrence was printed in a process called Auto-Select which negated the need to cut it, because it was a selective process and even fades and dissolves could be easily moved. The negative was originally single strand so you could make short versions, long versions, and foreign versions. There was no need to cut it, but they did , probably to save time scanning back and forth to select shots. The Auto-Select process caused a lot of wear and tear on the negative. It had been run some 250 times. They even ran it to make replacement footage sections!
When we tried to make our interpositive, which is a duplicating positive from which you make a dupe negative, the splices started tearing right through the shots. The chemistry of the film had changed, the cement on the splices was going, and the edges were loosening up. In the middles of the process we stopped and negative cutters at Metrocolor used thousands of pieces of mylar tape to repair every splice that looked like it was ready to go.
That negative broke so many times, and every time it broke we'd have to pull in more cans of negative, trying to go back to the separations, or go for alternate shots or trims if that didn't work. We had t go to dupe for some sections. We were supposed to have the interpositive off by the end of May, but we finally got it off in October.
I called Dawn Steel (President of Columbia Pictures) and said, "Look, understand that you are never going to be able to use this camera negative again. Right now I think it will run once more, it's taped to the hilt." She said, "You meant that if anything happens to the interpositive, we've got nothing?" I said, "Yes." She asked, "How much will it cost?" I said, "$52,000." So she said, "Make us another interpositive." Dawn has given me 100% support. She did it for history and to protect the studio's property. Although the project has more than paid for itself, that never became part of the original equation.
Q: What was the condition of the sound materials?
A: Two magnetic tracks survived but were mislabeled. The one marked 'print master' turned out to be a copy, and the one marked as a copy, a protection dub, actually turned out to be the original four-track magnetic master. It had splices in it where they had lifted scenes, and the oxide was flaking off, so it wouldn't run. The 600,000 feet of original mixing elements, including Maurice Jarre's original music, was all junked in 1975.
What we finally had to use as our original was fourth generation materials. The original mixing elements, you see, are first generations, and some thirty to thirty-five effects tracks are mixed down to one, which is second generation. Your master is third generation, and your print master--which is used to sound 35mm and 70mm prints--is fourth generation. Once we realized that would run, we made a fifth generation Dolby-A encoded full coat, as good as we could get. What your hear today is seventh generation, and, had we not used Dolby to suppress the noise, it would have sounded like a pot of mush.
Q: It sounds better than I remember it originally. A: You can thank Goldwyn Sound for that, and you can especially thank Richard Anderson and Greff Landaker, both of whom worked on Raiders of the Lost Ark. The gunshots in Lawrence, for example, have been basically modernized, they're re-equalized.
Q: Were there any problems in restoring the colors?
A: Not really, because Eastman camera negative is great. Over the years the color had shifted, it had faded just a little bit. To determine the proper timing, I was able to find one of the few surviving Lawrence 35mm Technicolor dye transfer prints, which don't fade at all. Bill Pine, the timer at Metrocolor, did not use a Hazeltine but timed every shot by eye. Everyone at Metrocolor did an extraordinary job. They did it for the love of the film and David Lean. David, because of their extraordinary quality, considers them to be his laboratory.
Q: Apart from a few minor imperfections, it looks fabulous.
A: There are imperfections, and they disturbed me originally, but not any more. David's feeling is that if you don't have some imperfections, people won't believe it's a restoration. It's a phoenix risen from the ashes, that's what it is.
Q: After you had spent more than a year restoring Lawrence to its original 222-minute length, how did you feel when Lean wanted to cut it?
A: Actually we restored it to 223 minutes because we added material that was never in it before.
Q: What was that?
A: It was all in the beginning of the second half, the scene between Arthur Kennedy and Alex Guines, things like Kennedy saying, "I understand you've been given no artillery."
After we'd completely reassembled the film, David screened it and said, "It's great, now let's cut it." And I said, "Huh?" He saw that I was not happy, so he puts his arm around me, walked me away, and said, "Look, do you want to have the perfect restoration of the December 9th, 1962 premiere, or do you want to make Lawrence of Arabia the best film we can make it? If you want to do that, let's sit down and take a scalpel to it, let's do the director's cut," and that's what we did.
Q: What material did he cut?
A: The only scene that he questioned was the first officer's club scene, with the pool table. He felt that it showed a side of Lawrence's personality you probably shouldn't see at that time. Other than that, he made only editorial trims. No entire scenes were cut. He tightened it as an editor, not a director, and most of the things he trimmed needed to be trimmed. For example in the second attack on the Turkish train, where Auda gets his horses, there's a shot where Anthony Quayle pops out of the top of an armored vehicle and fires a flare gun. In all the previous versions, he comes out of the vehicle, he tries to cock the pistol but it doesn't cock, so he cocks it again, and then finally fires it. But you don't need to see him not being able to cocks he pistol. That's gone now. There were also trims in the first scene with Lawrence and Dryden; there was a bit after they left Murray's office and before you see them walking down the corridor. It didn't belong in there.
Q: We were sorry to see that the balcony scene between Allenby and Lawrence wasn't completely restored.
A: There were two problems with it. First, the dialog didn't work, the dubbing of Jack Hawkins's voice was just too different. Charles Gray did an absolutely Herculean job, but, impassioned as he was, the voice just didn't work perfectly. Second, because of the way the film originally cuts, there's a question of where Dryden is all this time.
Q: Do you think the deterioration of the Lawrence negative was more a result of the particular printing process, the improper storage of materials, or the natural ravages of time?
A: If I had to weight one more heavily than the other, it would be the overprinting first, and quality of storage second. Ravages of time? Not really. If properly stored, those materials should not have gotten to that point.
Q: If this can happen to a classic film like Lawrence, what does that say about the preservation of the average, run-of-the-mill film?
A: It's not so much the quality of the materials, it's the way they're handled, putting them back together to make new negatives. One of the biggest problems in the studios is that they don't know what their own films are supposed to look like. And if you don't know what it's supposed to look like, you have no idea what the timing is.
Q: Has there been a decline in the level of technical expertise in the labs?
A: No, the labs are fine. It's a matter of budget and the studios are pushing their people too quickly. If the studios want great materials, t hey have to give their people the necessary time and the necessary budget and the access to original materials in order to see what the film looked like originally. They have to allow them to spend a few extra dollars, we're not talking about a lot of money, to make a proper internegative and proper interpositive, rather than just a interpositive.
Q: What's the present condition of Lean's Bridge on the River Kwai?
A: The have the camera negative. They have black and white separations which hopefully will register but you don't actually know until you try them out. People have supposedly made quite unquote 'preservation' materials but hen not tested them. You don't have preservation materials until you're made a dupe negative off your separations. I think Kwai could be gorgeous, it could be blown up to 70mm beautifully and the tracks could be reworked.
Q: We understand that you will be supervising the preparation of a laser disc of the restored Lawrence.
A: It'll be a letterbox version and will have lots of extra material on it. The entire script may possibly be put on the laser disc, plus the original trailers, and samples of daily continuity and mixing/dubbing sheets. We also have access to 65,000 feet of 16mm behind the scenes footage on the making of Lawrence. We also have footage of the New York and Los Angeles premieres, footage of Peter learning how to ride a camel, Albert Finney's screen test as Lawrence, and actual shots of Lawrence in Arabia in 1918. It's going to be six sides, at least. Hopefully it'll be out for Christmas and will probably retail for $125 or so. It's the ultimate gift for the cinephile. [The final boxed version contained the following: Production photos and a booklet detailing the twenty minutes of restored footage.--ed.]
*color film is protected archivally by the production of three black and white (silver) masters. Each master (cyan, magenta, and yellow) carries a portion of the spectrum. When the three are reprinted on color negative film, a duplicate negative of fine quality can be produced.