A black and white issue?
Colorization, preservation and the digital future by Richard Price
Colorization. Colourisation. It is easy to say that whichever way you spell it there is nothing good about the process. The term is a generic one for the technique that involves transferring black and white film to an electronic form and then computer colouring the picture by selectively filling in the parts of the image with colour pixels. The finished result is then sold for home viewing either on television, or on video or DVD. Colorized black and white films are not transferred back to 35mm for theatrical projection. Many visitors to this site are likely to have strong views against colorization, but this article will attempt as balanced a view as is reasonably possible and will touch on related, but much larger, conservation versus restoration issues. The whole colorization debate has a slightly 1980s feel to it, though there has been a recent resurgence of interest in the process, by its promoters if not the public.
Colouring monochrome films is nothing new: it is nearly as old as Cinema, with the earliest hand coloured films dating from at least 1896. Before then, photographic lantern slides were often coloured in by hand, achieving a certain realism for Victorian audiences. A development of hand colouring was the Pathecolor stencil process, a semi-mechanised form of the same thing, used from 1905 to the early sound period. If, for instance a film was to coloured with three colours, the technique would have involved taking three identical films and cutting out minute portions of the image on each frame. For example, if a brown cow on a green field with a blue sky was the image to be coloured the operator would, on a device like an editing machine, view a black and white frame of a positive print back projected onto a ground glass screen. He (or normally she: cheap female labour seems to have been the rule) traced the image on screen with a pointer.
The movements of the pointer were then reproduced in much smaller scale by a pantograph (left) connected to the pointer. At the other end of the pantograph was a vibrating needle that cut out a very small section of the film. This was done for each frame of film as the operator advanced the film frame by frame. Using our example above, the operator would trace and cut out every image of the cow on a given sequence of film. This film would then be the stencil for brown. This process would be repeated for green and blue, resulting in three stencils in all. The first stencil and the monochrome positive print would then together be run passed a pad impregnated with brown dye. The process would then be repeated on the part coloured film with the green and blue stencils until there was a copy of the film coloured brown, green and blue: the projection print.
As would be expected, this would be a long and costly process where five or six colours were being used in each frame (the release prints of the feature Cyrano de Bergerac (1923) (right) took a couple of years to colour) though a considerable number of prints could be made from one set of stencils. We might see colorization as an update of this process: black and white films being made more commercial by having them in colour rather than genuinely achieving a greater realism, even if the latter is the stated aim.
Certainly the techniques are similar in that areas of individual frames are coloured artificially, though colorization does not suffer from the colours misaligning with the image: stencils often could not be cut with accuracy, the stencil and the film being coloured sometimes did not line up in the printing process, and the dyes might run.
Colorization, despite improvements over the years, does not in the view of most achieve success in the way of tonal range or subtly, despite claims to do otherwise, the colour tending to be solid "blocks" of colour, the backgrounds being particularly unsuccessful and the choice of colours often arbitrary. The finished result has been likened to a moving lobby card, though this is one of the kinder comments.
Colorization is not a recent process. Universal toyed with a Japanese system in the early sixties and experimentally coloured sections of Psycho. It is perhaps a paradigm, both as to artistry and authenticity, for the whole colorization debate. This was a film deliberately designed in black and white allegedly because it was believed that red blood going down the plughole in the shower sequence would not get past the censor. One assumes that in the colorization test the blood was coloured red, not, as one might facetiously argue would have been more accurate, brown because apocryphally chocolate sauce was used since it photographed better in black and white. [Left; that's another fine mess you've gotten me into...]
Neither Hitchcock nor Universal was impressed with the results and the idea was shelved for a few years. Some stock footage and Betty Boop cartoons were colorized in the 1970s, and Universal again became interested in the early 1980s, by which time there were two rival processes in existence, Colorization Inc and American Film Technologies Inc's system Colorimaged. Both these systems had their experimental roots going back several years. The writer remembers the interest shown in the early 1980s when on British television the BBC aired sequences of a worn Chaplin print that had been colorized. Interestingly, and ominously, no attempt seemed to have been made to source a decent copy to colorize.
Colorization became popular when a certain level of competence had been achieved and when Hal Roach Studios Inc (by then having no connection with the great man himself) acquired a substantial holding the company that soon was to become Colorization Inc. Roach's film Topper was the first colorized film to be released in the United States on home video in 1985, followed by Way Out West, The Outlaw and It's a Wonderful Life. The novelty value led some who had been associated with these films to write favourably, even warmly, about the results, with Cary Grant, Stan Laurel's daughter Lois, and, initially, Frank Capra giving their endorsement.
A greater commercial impetus came with the decision in June 1987 of the Copyright Office of the Library of Congress to allow colorization to extend the copyright life of a film, though only in the colour, not the black and white, form. And of course television stations could now run the same old films colorized if not at prime time then at least no longer relegated to the late late show. [Right: Colour film from stencil]
We have to ask whether colorization, as it was in the 1980s or is now, does any harm. Is the criticism justified or is it just armchair snobbery from the highbrows? What is wrong with watching and enjoying the It's a Wonderful Life colorized? It has long been a Christmas favourite, especially on American television, and many who watch it no doubt do so without considering that artificial colour detracts from the black and white film, let alone considering it to be a "moral" issue, as do others who consider themselves purists.
More importantly, no damage is ever done to original material. It has even been argued that the process has gone some way to assist with the preservation of film by raising the public awareness of old films. The more you treat colorization as a moral issue the more arguments and counter-arguments present themselves. Does, for instance, the fact that Stan Laurel in later life regretted that the Laurel and Hardy feature Babes in Toyland had been made in black and white justify its colorization?
From the mid eighties when colorized films began to be shown on American television, opinion of audiences was generally against them, favouring original black and white. Then the veteran filmmakers started to come out against the process: Frank Capra changed his mind about It's A Wonderful Life and Billy Wilder asked whether colorization was intended to make classics better - and if so how - "or do they", he asked, "hope to palm off some of the old stinkers by dipping them in 31 flavors?" But for a while they found an outlet on home video.
In 1986 Turner Entertainment announced plans colorize a number of black and white features in its newly acquired MGM library. Ted Turner famously relished the controversy and was a great promoter of the process. Several MGM, Warner and RKO films were colorized in the late eighties, including The Maltese Falcon (left) and Casablanca (below). Despite starting work on a colour version of Citizen Kane a clause from Orson Welles's contract with RKO was invoked to prevent its completion and release. One can only guess what this would have done to Gregg Toland's chiaroscuro photography and one hopes they would have drawn the line at colorizing the News on the March sequences.
As the process improved, attempts were made to make the colours more authentic - an interesting concept in this context - by trawling through studio records to find out what were the original specifications of the decor and costumes, forgetting that while in life the sets and props no doubt had colour to them, the whole film was designed, lit and photographed with a black and white end result in mind.
If the James Whale Frankenstein films were colorized would Boris Karloff's face be green as was his make up in order to photograph as white as possible? In any event films made in black and white in the thirties and forties do not look like Technicolor films of the same period printed in black and white: the subject, style, the average length of shot and therefore pacing are different, the result of contemporary commercial and technical considerations and not least because of the heavily prescriptive influence that Technicolor itself had on the use of the process.
In 1988 Congress established the National Film Registry, which was to a list of important American films, 25 each year, to be labelled with a disclaimer as to "authenticity" if sold in an altered form. The almost universal opinion was that this was toothless bureaucracy at work that had nothing to do with film preservation and an expense to the US taxpayer that would have been better directed to film preservation. The first list of 25 submitted in 1990 contained Dr Strangelove - a partly British made film.
Similar calls came from a forum of directors in the United Kingdom who suggested that there should be a category of inviolable classics that should never be colorized. Not much thought seems to be given to how this would work, perhaps too little was given to this exercise too: along with Hamlet and The Third Man, Rebecca as cited as one such classic. Rebecca (left), despite its British cast and director, is of course an American film. Colorization has always been less of an issue in the UK. Very little material has, to the writer's knowledge, ever been shown on British television. Only a few American television programmes have been shown, such as a Judy Garland special and episodes of the Steve McQueen western series Wanted: Dead or Alive, the latter looking like Cinecolor seen in a nightmare. Whether one objects as strongly, more strongly, or not at all to colorized black and white television programmes is a matter of opinion, but the arguments either way are surely the same as for film.
Colorized films, viewed on television, videotape or DVD with the colour turned down do not look the same as the original monochrome picture. The shading and contrast are not the same. Since the image is degraded, albeit only in the colorized form, it is not possible to argue that colorization is archivally sound. It may happen that in the process of colorization, genuine conservation or restoration of original sound and picture elements has occurred before the process of colorization is started; at least claims have been made to that effect.
[Right: the Legendary Lon Chaney in The Phantom of The Opera (Handschiegl colour)]
Turner Entertainment has otherwise shown an enlightened attitude, naturally a commercial one, to both preservation of films in their original state and, just as important, access to those films, and it was certainly for commercial reasons that it quietly dropped the colorization programme as the demand for colorized films started to dry up in the early nineties.
The debate is not over. The series The First World War in Colour was recently shown on British television - on Channel 5 - exploiting the idea of showing actuality footage in colour of subjects otherwise only ever seen in black and white, as in the series The Second World War in Colour, Britain at War in Colour and The British Empire in Colour. Each of these showed astonishing original colour footage. The First World War in Colour used cheaply colorized, battered prints and the result looked as bad as any colorization done in the 1980s. In some ways colorizing documentary footage seems even less justified than fictional subjects.
The weakness of the series was underlined by the producers resorting to the to-camera reminiscences of game old boys aged 105 who (for example) remembered sharing tins of bully beef in the trenches in 1917 and by hiring Kenneth Branagh to do the narration. (Interestingly a small number of military subjects were shot at about the same time in the three colour Gaumont Chronochrome process, though none of combat. Some of these were shown in a rival Channel 4 documentary series made at about the same time.)
Claims have made for recent improvements in the process with comparisons being made to the colour saturation and tonal range of three strip Technicolor or later colour processes. In the writer's view the achievements fall short of these standards. Colombia TriStar recently released on DVD some colorized Three Stooges shorts (left) and all the old issues resurfaced. The discs have both black and white and colour versions of the films to satisfy both preferences and the same claims for authenticity as to original colours of props have been made as in the 1980s.
Columbia TriStar made the not very interesting point that colorization makes old films more attractive to modern audiences, but also the much more interesting (though debatable) point that the process is no more vandalism than the conversion of an analogue soundtrack to 5.1 digital audio. Even George Lucas weighed in, asserting that the Three Stooges in colour took them out as their natural 1940s black and white context, though of course with their final Shemp replacement they made a number of colour films in the sixties, and George Lucas himself has not been uninvolved in the re-releasing of digitally enhanced films. The discussions can go on and on, though they always seem to focus on the rights of film makers rather than on film preservation.
As long as the technology and demand for colorized films exist it is not realistic to prevent their availability for those who want them. If there is to be a place for colorized films, it must be on video or DVD, available for sale or for rent, and clearly labelled as such. A colorized film should not be allowed to become the received version of a black and white film, not even if they are "the old stinkers dipped in 31 flavors". So like cropped pictures and panning and scanning, colorization has no place on broadcast television.
The technology itself of course has its uses. No criticism can reasonably be made of using similar techniques to colour grade films before their release, notably in the Lord of the Rings trilogy, or in much current television post-production. One day there may be admirers of colorization in the same way that people today (the writer among them) find much to like about artificially coloured films and old so-called natural colour processes like two strip Technicolor which were so often denigrated at the time, or at least when the novelty had worn off. There may even be a website devoted to colorized films. But if that time ever comes the same people will no doubt be exercising themselves about some other threat to film, or by then digital, heritage.
A well-known archival use of colorization, perhaps the only valid use, was in the Photoplay restoration of the bal masque sequence in the Lon Chaney version of Phantom of the Opera, where only part of the footage survived in Technicolor with the remainder in black and white. Using the surviving material as a template colorization restored colour to the monochrome footage. In another sequence the deep red of Lon Chaney's flowing cape was restored where the original (right) had been coloured frame by frame on black and white stock using the Handschiegl process, a process not unlike stencil colouring.
As to preservation of whole films, there have already been major digital, as opposed to optical, picture and sound restorations of films, the results being transferred back to film. Studios such as Disney and Universal have produced exemplary digital restorations in the last ten years, the costs being recouped by theatrical re-release or home sale. A very recent digital restoration is of Gone With the Wind (below left) by Warner Brothers, digital techniques being used to get back to the original imbibition-printed nitrate look. In 2004 the film was shown digitally at the National Film Theatre. National film archives, perhaps out of conservatism, have been slower to move away from traditional techniques, though this may change, especially as digital techniques can clean images and repair damaged sections of film in a way that conventional techniques cannot.
Such areas are fertile grounds for debate where currently there is no consensus, and in some ways they make all the old colorization arguments seem trivial. Digital restoration and even projection may become the norm, even though they could be said to be antithetical to the very idea of film, particularly projection, where the medium on which sound and image are carried and the process of exhibition has nothing to do with film in its usual meaning. Nevertheless, against this, digital projection promises greater distribution, and therefore greater appreciation, of classic films.
Perhaps the debate about digital enhancement should be widened to include undetectably manipulated images of actors of different generations being paired in the same film or more sinister, and only just round the corner, the Orwellian possibility of creating convincing fake documentary images. Imagine, for example, footage which appears to record a secret wartime rendezvous: Hitler and Churchill greeting each other warmly at Hitler's Berchtesgaden hideaway, perhaps with Goering patting Churchill on the back, Himmler looking on and Churchill glancing furtively to camera; the colour and everything else about the images being seamlessly manufactured to look like one of Eva Braun's Agfacolor home movies.
All arguments return to the nature and degree of any change made to original sound and image. Perhaps we should say that all restoration must bring us nearer to the original film (the look, the sound, and the whole experience of watching it) but not beyond it. But there could be special cases where we should add intention to the above list: there would, for instance, seem to be justification for using digital techniques to cure the bedevilled soundtrack of Chimes at Midnight, where, but for economic and technical hardships, the dialogue would all have been properly recorded and synchronised.
But from Orson Welles and Orwell let us return to our subject and leave the last word to Hal Roach, every inch a producer but one who gave considerable artistic freedom to his contractees. In 1992 at the age of 99 he travelled to Britain to publicise the UK release of the colorized Way Out West (right). Roach's studio of course had produced the film and the re-release was then being marketed by Hal Roach Studios in the Colorization Inc process.
Holding court at the Park Lane Hilton (unforgettably telling people "I always stay here…. my girlfriend back home is Louise Hilton. Her old man did these places out you know. I feel a kinda loyalty" and handing out pocketfuls of propelling pencils embossed "Compliments of Hal Roach - 100th Birthday") he was asked what he thought about the repackaged classic. His diplomatic response was as characteristic and it was wise. His words were those who of a businessman who may have eschewed pretensions to art but certainly understood his craft: "The picture looks good. It looks professional. Apparently people like color these days. I know one thing: the movie cannot be any funnier in color."
No funnier. No better. We ought to conclude that all audiences, casual viewers as well as the many hues of film enthusiast should strive for and learn to appreciate films in their original form.
©Richard Price 2004