The movie is Straight Shooting, directed by Jack

Transcription

1 Colours, audiences, and (dis)continuity in the cinema of the second period Film History, Volume 21, pp , Copyright John Libbey Publishing ISSN: Printed in United States of America Colours, audiences, and (dis)continuity in the cinema of the second period Nicola Mazzanti The sun is high, blinding and hot: the image flooded in a bright yellow tint. Under a tree, an old, desperate man is lying on a grave, and, like a plain-clothed Lady of Sorrows, a young girl is leaning against its wooden cross. They are mourning their beloved son and brother who was ambushed by rancheros while he was trying to get water from a fenced-off creek. Half hidden behind the bushes, a cowboy is watching them, unnoticed. Slowly, he comes forward and makes a move for his gun he s there to kill the old man. After all, that s what they pay him for. But he can t. He just stands there, hesitant, watching. He sees them looking up, spotting him amid the shrubs, staring at him. [Point-of-view medium shot]. But he can t hold their gaze, looks away, takes his hat off a sign of respect? [Medium close up]. When he manages to look at them again, the colour bursts into a violent, painfully reddish sepia tone. It s an unexpected flash, striking both the cowboy and the audience. [Same POV MS as the previous one]. Back to a medium close up and back to the regular yellow tint. We see him touching his eyes with the back of his gloved hand a tear, or just a speck of sand? We can only guess, but we know that he has just had his epiphany, like a Saul of the Wild West. Now he will help the farmer s family against the ranchers. He will fight for them, kill for them. After all, that s what he s good at. Later on, the cowboy will thank the girl: Thank you for opening my eyes. Yes, indeed, that scene opened his eyes, and the sepia tone opened ours, too. The movie is Straight Shooting, directed by Jack (John) Ford in 1917, a western revisiting a familiar topic of the Wild West mythology the war between ranchers and farmers for the open range. The redeemed cowboy switching sides is Cheyenne Harry, played by an aging but absolutely unforgettable Harry Carey. The scene is described as it appears in the only surviving original (and coloured) print, and in the restored version from the Nederlands Filmmuseum. 1 Nicola Mazzanti has been active in the field of film archiving and restoration for over twenty years, as both a film archivist and founder of the festival Il Cinema Ritrovato. He founded and directed an internationally renowned film restorationlaboratory, and was involved in the restoration of hundreds of silent and sound films. He also teaches and writes about the theory and practice of film archiving and restoration. Currently he is an independent consultant in Europe and the United States, a member of the Technical Commission of the International Federation of Film Archives (Fiaf) and a member of the Board of Directors of the Association of Moving Image Archivists. Correspondence to nicola.mazzanti.95@virgilio.it

2 68 FILM HISTORY Vol. 21 Issue 1 (2009) Nicola Mazzanti The effect of that one sepia-toned shot arises from its unexpected appearance in a movie that otherwise uses colour in a traditional and less unsettling way: the vast outdoors are strictly tinted in a bright yellow as days normally are in the Santa Clarita Valley, north of Los Angeles, where the film was shot, in or around Harry Carey s ranch. Indoor scenes are either tinted amber or an unusually dark pink verging on violet. Amber is used for all interiors (the saloon, the rancher s house) except for the farmer s house, which is always tinted pink/violet; thus, this colour soon comes to identify the girl and her space, in opposition to the harsher, violent yellow of the scenes outdoors, which are Cheyenne Harry s territory. As an example of this codification of spaces, one could take the scene when Cheyenne Harry and Joan (Molly Malone) part for the first time after having found each other. He is outside her house, immersed in his yellow getting ready to ride off to kill his old friend turned foe, Placer Fremont (Vester Pegg). In the kitchen, softened by the pink/violet tint, a saddened Joan glances at her dead brother s soup plate on the table. She picks it up, holds it to her heart, remains still for a moment, then slightly turns her head as if she was looking at Cheyenne Harry (but he s outside, out of her sight). Cut. Harry Carey, also pensive, is standing by his horse this time flooded in the yellow light that comes through the trees; he turns his head as if he could see her one last time. Cut. She carefully puts the plate away, not into the chest with the others, but into a drawer to be kept forever as a keepsake of her dead brother. The crosscut ends: Cheyenne Harry hurriedly rides off to his cruel duty, brushing away any thought of peace, tenderness and love that the girl, in her mellow violet light, made bloom for a moment. The timing of the crosscutting and the restraint of the characters gestures, barely hinting at their devastating feelings, are perfectly designed. And the colour is yet another ideal element, working in unison with the others, so much more than a mere reference to the light or time of day. We do not know and probably never will whether the same colours were also used in the American release version of Straight Shooting, nor if the only surviving print differs from those produced at the time for distribution in other European countries. In other words, we do not know how significant an example this is, and to what extent we can draw conclusions from it. We know, of course, that cowboys don t cry well, almost never, as even Harry Carey tries to conceal his tears. Still, we like to think they might, and we can only wonder if sepia is the best-suited colour for such a dramatic incident. It may be true in the case of Straight Shooting, but not enough by far to make it into a rule. An aging cowboy crying in sepia might just be the juxtaposition of two exceptions to the rule. But as we know, exceptions are often very telling. So, for the time being, we shall retain the strong impression the colours of Straight Shooting produced on us as modern spectators, and put it aside carefully, as Joan does, because impressions such as these, like china, are fragile and easy to dismiss. As we keep this sepia moment in the background of our discourse, we wonder about the use of these colours and the reaction they might have caused in the audience back in Unfortunately, wonder is really all we can do, since, as William Uricchio notes, the early cinema agenda has somehow managed to marginalize such topics as colour, non-fiction, and performance. 2 Here we are concerned with colour, with applied colour, precisely, a term we use to define those colouring processes (tinting, toning, hand and later stencil colouring) that were applied to black-andwhite release prints, as opposed to photographic processes devised to record natural colours on film. 3 Uricchio s remark does in fact ring true. The discourse on the colours of the silents is limited to a few analyses of individual films, research on very specific issues (historical or technical), and episodic although often inspiring remarks, comments and hypotheses; it therefore lacks a comprehensive and wide-ranging approach that can only be based on a study of the texts (the films), of their intertextual relations, and of the broader social and cultural context in which they were created and experienced in short, on a study of all the factors that informed the intentional use of applied colours, as well as their reception. Still, these questions do not seem to be unimportant or without consequence, if we consider that what s striking is the breadth of colour systems, the breadth of colour effects, the range of uses to which colour is put over an extended period. 4 Or perhaps it is just me, and the fact that I spent more than a decade restoring silent films, forced into a daily intimacy with their colours: often flat and bland, sometimes fascinating and surprising. In any case, convinced as I am that applied colour played a crucial and complex role in silent cinema, and because of my personal magnificent obsession, I in-

3 Colours, audiences, and (dis)continuity in the cinema of the second period FILM HISTORY Vol. 21 Issue 1 (2009) 69 tend to discuss the use of applied colour in the cinema of the second period. I am adopting here the term ( le cinéma de la seconde époque ) introduced by Eric de Kuyper to define a cinema markedly different from both the cinema of attractions and that of the classic Hollywood style of the twenties; a period (from roughly to 1917) chronologically placed between the two without being simply an evolution of the former, and even less a teleological preparation for the latter. 5 Although generally defined as the teens, as in any periodization the boundaries of the second period are hard to nail down to precise dates, influenced as they are by complex dynamics (1917 might be the watershed for the US, but not necessarily for other countries Italy for example). Abandoning, in a completely non-linear and non-deterministic way, the modes of the cinema of attractions, but before fully adopting the codes of narrative continuity, the cinema of the second period exhibited a fascinating degree of creativity and experimentation that extended to the use of colour, thus assigning to applied colours a wide and complex range of narrative, figurative, metaphorical, indexical functions not experienced before, and not to be seen afterwards. Colours, versions, and audience(s) The relationship between applied colours and their audience is undoubtedly long-lasting (three decades, almost one-third of film history) and, as we understand, a strong one. In fact, breaking up with colour in favour of the more artistic and realistic black-and-white must have been difficult, at least judging from the interestingly mixed positions expressed in the debate on this topic. See, for example, the mixed reactions to films such as Von morgens bis Mitternacht, and the fact that even some of the most important all-b&w films, as late as Metropolis, were also distributed in coloured versions. 6 Distributors evidently counted on moviegoers undying love for colour to recoup the extra costs of tinting. Moreover, the rather common practice of modifying colour schemes for foreign distribution versions, which suggests that a need was felt to adapt colour schemes to the taste of a different audience, provides further evidence of a deep relationship between colour and its spectators. Of course, colours were also changed for other reasons, cost-saving being one. Eliminating the most expensive and time consuming processes (as stenciling and toning), or avoiding colouring altogether, were means to reduce costs in markets expected to be less lucrative, and to facilitate penetration into lower portions of the market by diversifying the price structure (cheaper prints for smaller exhibitors and cheaper tickets). Evidence of this practice can be found in numerous surviving prints, as well as in Pathé s well-known and widely advertised policy of offering the option of coloured or cheaper black-andwhite prints, and in the commercial practices of other distributors. The George Kleine-Cines contract of December 1911, for example, reads: Societa Italiana Cines may tone such subjects as appear to them to be improved by the process, and shall charge for toning 5 centimes per metre upon that part of the subject which has been toned. Provided however Mr. Kleine shall be privileged to instruct the Cines Co. to omit toning, there shall be no charge for tinting. 7 The contract seems to indicate that by 1911 Cines and Kleine considered tinted prints as the norm and toning a reasonable (although more expensive) option; interestingly, black-and-white was not even taken into consideration. Considerable evidence, including other papers in the Kleine collection (contracts, shipping lists, lists of prints and releases, relative earnings, etc.); the fact that extant original prints conserved in Italy or elsewhere (including prints of the same films distributed by Kleine) show that Cines films of this period were almost invariably coloured tinted, toned and even occasionally stencil-coloured; and the fact that these prints were produced by Cines in Rome, which also produced the prints for Kleine, all suggests that Cines films distributed by Kleine in the US and Canada must have been at least tinted (while toning, being a more expensive process, was considered negotiable). 8 Unfortunately, neither the contract nor other documents in the Kleine collection sheds any light on what was to happen whenever Kleine would opt against toning: who decided what colour to use in place of the tones? We can t but presume that it was Cines decision, since all the prints purchased under this and other contracts were manufactured in Rome and then shipped to the US. 9 (In the case of major feature films, like Cines Quo Vadis?, negatives were imported by Kleine and prints produced in the US.) Needless to say, we can t answer the next question: Who at Cines made the decision? It is reasonable

4 70 FILM HISTORY Vol. 21 Issue 1 (2009) Nicola Mazzanti to presume that generally somebody in the production/distribution structure made this decision, although not necessarily the production team responsible for the specific film. Clearly, this should not be considered a general rule; rather, it depended on the company s structure, the importance of the film, even the role of the director. Furthermore, these are factors that changed over time: the procedures applying to a non-fiction film from 1911 were probably rather different from those applied to a feature film distributed only a few years later. Altering colour schemes was only one of the many elements shaping an overall strategy of adaptation to foreign markets. In the Kleine papers we find references to subjects not suited for the American public, and to shots or scenes that needed to be cut or shortened for the same reason. 10 In many instances films underwent much deeper modifications, as was the case with the UK release version of GiovanniPastrone stigre Reale (Itala Film, 1916). An original sample print (without intertitles and meant as a reference for the production of other prints), and the Itala papers at the Museo del Cinema in Turin, document how a completely different last reel was cut and some extra shooting took place as well to turn the original tragic ending into a happy one. The beautiful, unlucky Countess had originally died alone in her bed, abandoned by her lover who was nursing his sick wife; in the UK version the two lovers are happily reunited, there is no wife, and the Countess s husband appropriately dies in a fire of which we find no trace in the Italian version. Needless to say, the colour scheme was modified accordingly. Considering the importance of the changes, and Pastrone s role at Itala and in the film, it is obvious to expect that he supervised the production of the new version although we do not know to what extent he was involved in the choice of the colours. We do not know who was responsible for any change in the colours used in these different versions, although we understand that this could vary depending on the context. As a matter of fact, we know very little about who was responsible for the choice of a film s colour scheme in the first place. We have very few primary references showing a direct involvement of the creative crew (director, cameraman, set decorator ) in the process. When such references exist, they are generally very significant, as Murnau s instructions to use black-and-white in the dream sequence of Schloss Vogelöd, Vidor s remarks about the use of colour in Sky Pilot, or Lucio d Ambra describing Carnevalesca s chromatic structure. Nevertheless, these references are both rare and rather late. Not surprisingly, they belong to a period when the role of directors was already established. But for most of the history of coloured films, and most significantly for the films of the second period that made such an extensive and complex use of colour, we have nearly no testimony about how specific colours came to the world, and by whose decision. 11 So it is not surprising that we also do not know under what principles certain colours were deemed more appropriate for a foreign audience than those used in the domestic version. It is a fact that when two or more prints deriving from distribution in different countries are found, their colour schemes differ, sometimes dramatically. If cost-cutting accounts for the reason why toning and stencils were eliminated, it doesn t tell us anything about why a certain colour rather than any other was selected to replace them. In most cases, the change is intuitive. A sepia tone is replaced by a yellow tint, a blue tone by a blue tint, tint and tone combinations are replaced by the tint used in the combination (the chromatic effect of which was evidently considered predominant: a blue-toned, pink-tinted night scene would be often just blue-tinted in a foreign version). But exceptions are countless [Plates 1 and 2]. Besides, this does not account for changes taking place independently of any compelling costcutting reason. Il Bacio della Gloria (Film d Arte Italiana, 1913), a love story set against the background of the Italian-Libyan War, provides a typical example. The colour scheme used for the Italian release version is rather basic: a sepia tone (just two outdoor shots), a rose tint that is used for all the interiors, and a light orange tint for all the outdoor scenes in Italy, and in the Libyan desert where the hero is wounded while bravely fighting hordes of savage Arabs. Colours used in a surviving UK distribution print (released as A Glorious Scar) are identical, with the exception of a light green replacing the light orange in the outdoor shots. In other words, the Libyan desert is light orange for the Italian audience, and light green for the British spectators: surprising, to say the least (also, no chance of this being an error, as the following discussion of the Film d Arte s colour codes will prove). Similar examples of intuitive, counter-intuitive, or apparently incomprehensible changes are found in many of the dozens of coloured silents that I have

5 Colours, audiences, and (dis)continuity in the cinema of the second period FILM HISTORY Vol. 21 Issue 1 (2009) 71 restored over the past fifteen years. The examples are numerous enough to prove that adapting colour schemes for different audiences was a fairly common practice, and while we cannot always fully understand the principles that informed the changes, they appear to be the result of two inter-related and mutually influencing factors: cost-saving strategies and an attempt to conform to a different taste or sensibility. A particularly fascinating example of how profound an effect these changes could have on a film is provided by Carmine Gallone s Malombra, a 1917 drama whose colours play a central role in defining the atmosphere of an eerie story and in painting the light around the diva s body and space. 12 Two original prints of the film survive (both incomplete), one of the Italian release version, the other found in Uruguay. Their colour schemes are completely different, with the South American version having clearly undergone a process of simplification (fewer tones and combinations), the logic of which is sometimes clear: a sepia-toned/blue tinted shot turning blue-tinted only, a sepia-toned/yellow tinted scene of the diva in a garden is green-tinted, and so forth. But often the logic is not so obvious, as when a scene tinted in a very pale pink in the Italian version turns into bright amber for South-American distribution, or a somberly lit party scene (blue tint in Italy) acquires a lively, gay light thanks to a salmon-pink tint [Plates 3 and 4]. While the cost-cutting factor explains the need to simplify the colour scheme, it does not account for the shift in tonality deriving from changing colours not only in the shots formerly toned instead of tinted, but in almost every scene. From a chromatic standpoint, Gallone s original drama is a diva-film imbued with a symbolist imagery dominated by Lyda Borelli s acting and physical presence, and supported by an extremely sophisticated use of colours that play a counterpoint of somber tints (blue, violet, pink), dramatic tones (sepia, green, and two blue tones one shining, one dark blue, almost black), and subtle tint and tone combinations, always either engraving or sculpting the diva s body. Engraving. The madness at lunchtime scene: harsh sepia tone and soft pink tint. Lyda Borelli s insane gaze, her twisted muscles, her face contorted by madness where every wrinkle is to be counted; as in a bas-relief every detail is engraved in an unforgiving sepia-toned sharpness and inundated by a completely unnatural pink light (lunch is served on the terrace) that makes the inanimate vibrate (the tablecloth, her dress, the walls). Sculpting. Marina/Lyda Borelli is possessed by a long-deceased woman s soul: a dimly lit room, a body slowly emerging from the blue shadows into a mauve light filtering through the window (blue tone and mauve tint). We see her body being slowly sculpted by the colours themselves. Together, coloured shadows and coloured light become the diva s body [Plate 5]. She is her body, her body is but colour. As Michele Canosa writes, the diva is her body: possessed by passion. 13 If so, colour is really the flesh and the soul of that passion. The complexity of the effects and the degree by which colour is intertwined with other elements of representation are not surprising to those who know the Italian cinema of the late teens. So Michele Canosa points out, by quoting Louis Delluc: Il fauno made a certain type of filmgoer smile. But real artists were enthusiastic about it. Febo Mari, Elena Makovska and this symbolic impressionism of the avant-garde Italians conveyed by the staging, the ideas, certain landscapes treated as canvasses creates a rare and shining composition. Il fuoco, Il fauno, Rapsodia Satanica, Carnvalesca,... and others, all these attempts at cine-poems should be reassessed: they are the real avant-garde of the Italian cinema, predating the French first avant-garde. Had we only listened to Delluc and his followers (Surrealists), we would not have wasted time in hallucinations of improbable predecessors of Neo-realism, sperduti nel buio. 14 Canosa and Delluc do not explicitly refer to colour except that we can t really imagine the canvasses Delluc hints at as being black-and-white (and besides, these films are all vividly coloured). But Delluc saw these films in France, and we wonder which version of Malombra he might have seen: the Italian, one similar to the South American, or yet another? Across the Atlantic (this we know for sure), a new colour scheme turns Malombra s South American version into a melodrama adorned by vivid, bright, violent, shouting colours (yellows, ambers, blues, pinks, violets): a symbolist reverie turned into what our modern eyes almost inevitably perceive as a Mexican melodrama, a completely different tonality for a different type of passion.

6 72 FILM HISTORY Vol. 21 Issue 1 (2009) Nicola Mazzanti Malombra, and many other films of which we possess multiple copies with differing colour schemes, provides further evidence that in silent cinema colour effects were neither chosen nor perceived nor should they be seen or studied today as independent of one another. Rather, they were inter-related and inter-dependent: they created (or aspired to) a coherent system functional to the film for which they were chosen. While they may have had several co-existing, perhaps different and even competing functions narrative, figurative, metaphorical, indexical still, within the film, colour effects operate as one system based on their inter-relation, balance, rhythm, and interplay with the images and the narrative. With this consideration in mind, it comes as a logical consequence that when cost-cutting measures led to a simplification of the colour structure, it was not just a matter of switching one colour for another (for example, a pink tint for a combination of sepia tone/pink tint). Rather, since the original balance was corrupted, a new one had to be found, and this implied that the whole colour scheme had to be revised and sometimes completely changed. So, when cost-cutting measures imposed a simplification of Malombra s palette by eliminating the expensive toning process, this made it impossible to maintain the original inter-play between colours and imagery in the film which largely relied on key effects like tones and combinations and this in turn led Cines to modify the whole chromatic orchestration of the film. At that point, the need to find a new balance intersected with the intent to create something that, at least according to Cines, could better meet the taste and expectations of the South American audience. To use a musical analogy, the result is a new arrangement, as if suddenly the string section was not available any longer. To re-create a coherent balance, not just some, but all of its components had to be changed. When studying early films, it might come as a surprise that the opposite is also true: when the decision was made to maintain the colours unchanged, the level of consistency was remarkable. This is even more surprising if we consider the limitations of the technology and the techniques that were used at the time. The earliest example of such consistency I can recall is La poule aux oeufs d or, a stencil-coloured Pathé trick film whose three existing original prints show a remarkably consistent colouring, both in terms of the precision of the stencils and the characteristics of the colours (density, saturation, hue) 15 [Plate 6]. The fact that consistent results could be achieved as early as 1905 (with techniques that appear so pre-industrial to us) shows that mastering the process was a goal both sought after and often attained throughout the silent era. Besides, had the process of producing multiple coloured prints been completely inconsistent, it would have been useless to go through the painful, costly and time-consuming process of seeking subtly different colour effects within the same film, or to modify its colour scheme. But if consistency was possible, and chromatic variations (also subtle, as we will see) were the norm, this implies that when we are confronted with a strange colour effect, as in a different colour suddenly appearing without logical narrative explanation, we can t simply assume that an error occurred. Rather, we have to question whether it was the result of a choice, and if so, why. So, the practice of producing different colour versions reminds us that when we approach the use of colour in silent cinema we are confronted with an intrinsically coherent colour system conceived to interact with, and function within, the narrative and figurative system of a film, to a degree that this colour system had to be re-negotiated and adapted to what cinema thought were the spectators expectations, desires, emotions and taste. In other words, we are constantly reminded that colours in early cinema were always chosen, they were never there by accident, as Jacques Aumont points out: Cinema has never had a phase when it was not in colour: colours were there, immediately. Immediately as a deliberate addition, a supplement, contingent by nature, thus a fortiori implying intent and purpose. 16 This deliberate act of applying colours always implies a choice, and this invariably entails a purpose. It does not really matter how conscious this purpose is nor how open the search is for a stronger relationship with the spectator, or for a more sophisticated use of cinema s many narrative and figurative devices; we still sense a tension between an eye which is still sufficiently virginal, and the dawning consciousness of the possibilities of cinema. 17 Nor does it matter whether the intent is purely commercial; it still entails a stronger and deeper (although still largely uncharted) relationship with the audience and its world, as well as a more conscious awareness of the medium. As Eric de Kuyper writes: [T]ogether with a blatantly opportunistic and exploitative use (colour makes richer), one should not underestimate

7 Colours, audiences, and (dis)continuity in the cinema of the second period FILM HISTORY Vol. 21 Issue 1 (2009) 73 the plastic and narrative stakes colour also offers to the cinema of this time. 18 Colour was one of many elements (and not the least important, as we increasingly tend to think) that the cinema of this period carefully considered with varying degrees of deliberateness and consciousness when it confronted the need of capturing one or more audiences. And for this reason the relationship between colour and its spectators (or how cinema conceived its spectators) was both strong (coloured Metropolis) and complex (Malombra), not only because colour was clearly a key component of elaborate commercial strategies, but also because audiences were perceived as being so diversified that the essential stability of the films themselves, of their chromatic characteristics, figurative and aesthetic systems, was questioned and often re-negotiated and modified. In other words, fairly soon the question ceased to be colour: yes or no, to become which colour where, when, and for whom. Unfortunately, investigating the relationship spectators had with the colours of the silents is extremely difficult, as is exploring the effect, function and role attributed to colour. References to how spectators (both professionals and simple moviegoers) reacted to colour are a rare find, particularly when compared with the richer traces left by other elements of the cinema experience, like the advent of sound or natural colour reproduction. One is tempted to wonder if the reason for this (or at least part of it) could be that applied colours have never had a real advent. Is it possible, having been there from the beginning, that colours were taken for granted, perceived as an obvious component of the cinematic experience? The effect of colour (even from the most basic standpoint: it was good, it was not ) was not commented upon, and was rarely listed among the elements that impressed the spectator positively or negatively. In a way, it appears as if colour worked as a sort of undercurrent: we know the effect on the audience must have been strong (commercial success, resistance to disappearance), but it still doesn t get to the surface. Perhaps, as Eric de Kuyper argues, colour produces something similar to the effect of music on the images, it plays with them, there s an interplay, giving an extra-textual dimension, where colours add another dimension to the filmic discourse. 19 Could it be that what is commented upon is just the effect of this interplay: that is, the experience of colouring is such an integral part of the film that it cannot be singled out? If so, colour appears to be something hard to define, a subtle quality that relates more to emotions than rationalization. If Maudite soit la guerre and many other films could genuinely be compared to a sentimental song, something I often like to do, writes Eric de Kuyper, the colour would be to the film exactly as the voice of the singer is to the refrain: a personal interpretation, the grain of the thing. 20 Something in Maria Callas s Casta Diva hits us so much deeper than the technicalities of her singing. When we hear a recording, it is our guts, not our brain, which tell us it s her. In this sense colour, beneath the rational perception of its narrative or figurative functions, is (or could be) an aesthetic experience. Film colours between sensuality and normalization If we broaden our perspective to include reactions to colour at the time it began to invade other media and the public sphere in general, starting from the second half of the nineteenth century, there is little doubt that it encountered a great deal of resistance. This was largely due to colour being seen as an uncontrollable, sensual experience exciting the minds and the hearts of those most easily influenced : children, women, and the uneducated masses (not surprisingly, the same categories that had to be protected from nickelodeons). Besides, colour is tasteless : One has to accept that [colour] has something to do with vulgarity, crudeness, childishness, popular taste. [C]olour, measured by our cultural codes, is usually on the side of bad taste. 21 Ultimately, colour was seen as a danger to civilization precisely because of that irrational effect we have just discussed. 22 As Tom Gunning reminds us, there are numerous parallels between negative reactions to colour and to early cinema. 23 Gradually, as was the case with cinema and its gentrification, a compromise was reached: colour was somehow tamed by making it less aggressive, more tasteful, and the middle classes progressively came to accept that some degree of sensuality as provided also by colour was to become a component of everyday life. This ultimately turned colour into an important component of consumer society. Still, like the cinema, colour may have become omnipresent, but still had the ability to disturb cultural hierarchies through its asso-

8 74 FILM HISTORY Vol. 21 Issue 1 (2009) Nicola Mazzanti ciation with the emotional and sensual rather than the rational and ideal. 24 Middle classes are extremely responsive and sensitive to anything that might challenge established hierarchies, cultural or otherwise. If they perceived colour as a danger, it was clearly because the others (again, children, women, lower classes) were experiencing it as appealing, fascinating, and alluring. And so cinema, in its untamed, vulgar, not-yet gentrified form, tried to use colours that were as bright, loud, and strong as possible. It is difficult for me not to see this strategy applied to both the use and the quality of colours (bright, harsh, primary) as they appear in early films. Whenever I see an early Pathé-stencil, as La poule aux oeufs d or, La ruche merveilleuse, orles Bohémiens, I can t but draw a parallel with such popular nineteenth century prints as those from the Imagerie d Epinal. 25 This might be due to the fact that my earliest and strongest memory of colour as a child is linked to two prints from the Imagerie that were hanging right above my bed, where I could watch them for hours on end, every night. Although I forgot their subject (definitely some battle scenes), I will never forget the bright, flat, unnatural colours only primaries: red, blue, yellow, green and the way they were used to highlight details in an otherwise blackand-white image: uniforms, explosions, palm trees, a blue cloud in an otherwise white sky (!). Such strategy in the use of colour recalls Pathé s earliest stencils that intended primarily to display colours as an attraction in itself. 26 Later on the approach changed progressively, but decidedly: when stencil colouring was technically improved and elevated by a more marketable and ambitious brand name, Pathécolor, the palette clearly veered toward the pastel and the naturalistic intent came to the foreground. 27 This change in function and chromatic characteristics (as gradual in time as evident in the results) echoes the almost contemporary process of taming colours, when colour experts strove to eliminate the jarring juxtaposition and too bright hues which then became restricted to lower class imagery. 28 Was stencil colouring s new palette chosen to please (or at least not to unsettle) a middle class that cinema deliberately tried to appeal to? If this is at least partly true, then one wonders about realism, the other component in the new use of stencil colouring. Beginning in the late 1900s and throughout the teens, Pathécolor was increasingly employed (and extensively advertised) for its ability to simulate natural colours. 29 With this purpose in mind it was exploited mostly in non-fiction films and period pieces, with undoubtedly amazing results: its flat dyes acquired depth, realistic shades and modulation when carefully applied onto the grey scale of the black-and-white picture. In the magnificent golden fur of the panther cruelly slaughtered in Chasse à la panthère, for example, as well as in the foliage, the illusion of witnessing natural colour is so complete that after a screening one is sure to have seen the blood although there is none, and the blood on the panther is actually black (orthochromatic negative film), not red. 30 Inevitably, Pathécolor was not, nor could it be, a form of natural colour. On the contrary, its strength, fascination and interest lay in the slight, sometimes unnoticeable shift it operated between the real and the figurative, and this was doubled by the ambiguity of those extra-carefully and consciously applied colours pretending to be just recorded colours. The intention was to play hide-and-seek with the figurative, while realism was pretended, although that shift and that displacement were always to be felt. But in Pathé s A Car Trip in the Pyrenées there are red and pink stones, and green and blue water, which seem incredibly unnaturalistic: the idea of people having fun with colour is really important, Uricchio recalls. 31 Inherently ambiguous (and thus fascinating) as Pathécolor s results might be, Pathé felt the need to proclaim the realism of its colours in a clear attempt to respond not only to competitors, but more generally to a call for natural colours that was rather pervasive in the literature of the time. Natural colour in film is of course both ontologically and technically impossible, and this is also what makes it interesting and allows cinema to use it as a language. Technologies designed for (mechanical, chemical, or digital) reproduction of colour might feature a stronger or weaker component of indexicality; they might display a more or less effective synchronicity of colours in space (the rose is red, and its stem is green) and even more importantly in time (the ever changing hues of a sunset), the one advantage natural colours have compared to applied colours). But they inevitably fail to be natural (despite what any Kodak ad might say). A convincing colour synchronicity might be displayed as an attraction in itself (as Kinemacolor or Chronochrome did, despite their crippling technical limitations), but natural colour is really only successful when its use

9 Colours, audiences, and (dis)continuity in the cinema of the second period FILM HISTORY Vol. 21 Issue 1 (2009) 75 becomes complex and self-conscious enough to take advantage of its unnatural component, thus turning into an effective narrative and figurative device. Once the technical and ontological impossibility of such a thing as natural colour is revealed, the ideological intent behind the recurrent call for chromatic realism found in the literature of the time is exposed and can be interpreted as part of a more general strategy to neutralize the use of colour in early cinema. Invoking and evoking natural colours implies condemning as unnatural applied colours that could more easily be turned into a danger for civilization. Those early films, which exploited their sensual, metaphorical, aesthetical, and emotional component so extensively, did not suffer from their lack of naturalism, but rather, as Tom Gunning so clearly points out, [T]he intense hues and sharp juxtaposition of these colours place them in the tradition of emotionally effective and possibly dangerous colours so bewailed by the guardians of genteel culture. They function as attention-grabbing attractions and incitements to fantasy, rather than harmonious and subtle appeals to an established aesthetic order, or carefully observed images of nature. 32 An effective way to normalize such a disorderly use of colour as displayed by early cinema, was to force it into the more orderly realm of realism oh, those carefully observed images of nature! in an attempt to restrain it from being experienced as a power in itself, rather than simply a secondary quality of objects. 33 When applied colours are perceived as more intense than reality, then realism is called to the rescue in order to force cinema (and its colours) back to its original Edenic (as in sin-free and imaginary) state of pure, neutral technical reproduction. Constrained by the boundaries of realism, cinema colours could thus be normalized by reverting to the ideologically and emotionally neutral function of being a scientific, mechanical, and consequently neutralized representation of the colours of nature. As Heide Schlüpmann points out in reference to the similarly ideological debate about unnatural colours vs. realistic black-and-white, to pull cinema away from colours had to do with disregarding the audience and the place of cinema in the public sphere: withdrawing to the scientific and technical side of film, withdrawing into the private sphere. 34 No wonder that audiences resisted, for a while, at least, both normalizations: natural colours and black-and-white. Interestingly, at approximately the same time that anonymous stencils become Pathécolor and took up the role of natural colour, stencil colouring and tinting and toning part their ways, most probably because the latter had established a different set of narrative and figurative functions for themselves that would not fit into Pathécolor s realistic agenda. Not that they were never to appear together again. But when this happened the results were completely different from those of La poule aux oeufs d or, for example, where they shared the same level as narrative tools (blue tone for the night, yellow highlights for the golden eggs, etc.) as well as functioning as attractions in creating or contributing to the representation of the fantastic. In the cinema of the second period, when they co-exist, they do so as independent colour schemes that interact and interplay while remaining separate, and this interaction almost inevitably produces rather exceptional results, as Maudite soit la guerre demonstrates so well. Maudite soit la guerre is undoubtedly one of the great films of the second period, and here I can t but refer to (and borrow from) Eric de Kuyper s extensive and inspiring discussion of the film and its use of colour, and of Alfred Machin s works in general. In Maudite, Machin draws a clear line between the two colour processes: war scenes are almost invariably tinted red, while stencil colouring is used extensively in the rest of the film. Clearly, the effect is a visual differentiation of the two dramatic poles of the film: The contrast between the biting monochrome red and the always rather honeyed atmosphere created by colour stenciling emphasizes the tension between the horrors of war and domestic bliss. But that s not all. The use of colour makes it easier to grasp this romantic and subtly sentimental dimension, without compromising its suggestive tone. 35 In addition to this basic function, the colours used in the stenciled section are also carefully chosen. And yet again, as we should by now be used to, they are all somehow subdued: Machin uses with subtlety and refinement the palette of colours available to him, which is rather limited in itself: soft greens, salmon

10 76 FILM HISTORY Vol. 21 Issue 1 (2009) Nicola Mazzanti pinks, faded ochres and a blue verging on grey-green. The colours used in the key villa scenes (in the repeated exterior and garden shots), i.e. greens and ochres, are also present in the interior shots, in which light and dark ochres predominate. 36 Within this chromatic orchestration of understated pastel colours, which at first sight could be taken for a plainly realistic representation of the villa s garden, one colour conjures a whole different set of meanings. It is elevated to another level of meaning due to its own characteristics and its relation to the other colours of the film: In the exterior scenes, one colour clearly dominates: the salmon pink of the geraniums. This colour tint practically leaps out against the green background of the garden! The same salmon pink is taken up in an even more emphatic shade in the tiled roof of the villa, where it contrasts with the light ochre of the walls. 37 The geraniums as such do not play any role in the film; they are neither referred to nor picked up by any of the characters. It is not about the flowers and their colour. Rather, it is about the colour that brings out the flowers, and by its tonality helps identify them as geraniums: This presence of geraniums, made miraculously visible by means of colour, brings me again to the theme of the film, though this time in a different register: to a domesticity that has become impossible and a love that is destroyed by war. And so the geraniums add yet another more subtle nuance. 38 And as in many other films of the time, each colour always works in relation to the others (or their absence). So in Maudite, the use of colour is also about the balance and the inter-relation of that salmon pink with the other colours, colours that let s not forget are always chosen and applied. So, in short, through the insistent touches of salmon pink in the geraniums and because their nearness to the ground, we see the earth which is green and ochre. This latter combination of colour contrasts vividly with the reddening sky (the scenes of aerial combat are given in a red tint wash). Colour adds to the charm of Maudite soit la guerre in a decisive but also impalpable way. And it is certainly not used purely for decoration. 39 Duplication, restoration, colours: some caveats Turning to the films in search of answers, as in the case of Malombra or Maudite soit la guerre, is both rewarding and challenging. The endeavor is made undoubtedly easier by the large number of coloured films that have been preserved or restored in colour by many archives around the world in the golden age of silent cinema restoration, from the late 1980s through the 1990s. But as Uricchio warns us, if we shift our focus away from discourse and production practices to the films themselves, we face an array of difficulties. 40 In fact, not unlike any document in any other discipline, films pose their own set of questions and challenges that deserve consideration. The key issue, as obvious as often disregarded when it comes to films, is that any document must be questioned before being accepted as a faithful witness. Is it an original, unaltered document or, if it is a copy, does it retain all the information and the characteristics that are relevant in the context at hand? This rings particularly true in the case of cinema, where what we access are prints resulting from either mass production or duplication. 41 The film production process has always been designed with the intent of allowing intentional variants and versions (languages, censorship, distribution strategies, etc.). It was also prone to produce unintentional variants due to inconsistencies in the techniques employed. Finally, its products (the films) suffer from the sheer passing of time. In the specific case of film colour, duplication (as part of a preservation or restoration process) implies the use of modern techniques and materials (such as colour film stock) that, not being consistent with those originally used, inevitably modify the characteristics of the film image to a more or less significant extent. 42 Not to mention, of course, that to access a print in fact means to experience the projection of a print, with all the consequences that this has on the actual perception of the film in this case of its colours and of their inter-relation and balance. 43 Although this is not the place for an in-depth discussion of how the methods of textual and material analysis (as per the traditions of philology and fine arts restoration) are applied to film, a concise review of some basic issues particularly relevant in the context of a discourse on colour might be useful. First,

11 Colours, audiences, and (dis)continuity in the cinema of the second period FILM HISTORY Vol. 21 Issue 1 (2009) 77 we are faced with the problem of assessing the relation between the print at hand and the original film: that is, if it is a version or a variant (clearly, this remains a key issue whether we access an original or a restored print). 44 Seeking clarity and conciseness, and accepting a great deal of simplification as necessary within this context, a version can be defined as the form a text (a film) acquires after deliberate modifications take place in the course of its life. In other words, a version is created by somebody (producer, distributor, director, censor, restorer, etc.) at a given point in the timeline of the film s history, and usually a family or group of film elements (negatives and their derived prints, or prints alone) testify to it. In short, a version always implies a deliberate choice and concerns a number of elements (although theoretically it can refer to only one element, as in the case of independent or avant-garde works). A variant differs from a version in that it applies to only one element (usually a print), and it can be the result of either a deliberate act or an accident. While a thorough discussion of the complexities in the issue of versions and variants cannot take place here, some key points should be mentioned. First, variants can derive from accidents. For example, parts of a print can decompose, some or all of the colours can fade or decay, inconsistencies or errors can occur in the production process. Secondly, variants can derive from the deliberate act of a collector or an archivist. For example, it was common practice in many archives to collate different incomplete prints to produce a longer, more complete one, thus creating variants that provided a complete storyline, but that did in fact mix and intertwine different versions. When this happened with prints deriving from versions utilizing different colour schemes, real chromatic Frankenstein s monsters were created. When such a variant is produced in an archive and is subsequently duplicated and distributed, it does in fact originate a version of its own, in the sense that it produces one or more generations of prints. Later on, these archive-originated versions can be combined to produce a new restored version, which is the summation of all the modifications stratified in the various original versions, archiveoriginated version, and variants that were used to create it. This is often the case when restoration projects imply a reconstruction from multiple sources. As a side remark, I feel obliged to underscore the fact that while versions and variants are a source of complexity in the study of cinema, their differences and their history provide invaluable information, as the earlier discussion about different colour schemes applied to different release versions demonstrates. In short, it is critical that versions and variants are made accessible for study and research. 45 Once the origin of a print is assessed (after we have ascertained if it is a variant or a version, and how it relates to the original film for example, concerning its colour scheme), it is crucial that the shortcomings of colour duplication and the effects they have on the images are taken into careful consideration whenever a restored or preserved print is accessed (as is most commonly the case). In very broad terms, in the analog domain there are two processes that are routinely employed to duplicate an original coloured print: colour internegative, and the so-called Desmetcolor process. 46 It is a well known fact that these processes fail to faithfully reproduce the original colours; a lesser known, but even more troubling, fact is that some colours are reproduced more faithfully than others, with the result that the overall chromatic balance (within the frame when hand or stencil colouring is used, or between shots when different colours and processes are applied) is distorted to an extent that has led some authors (such as Paul Read and Mark Paul Meyer) to define these processes as simulations rather than duplications. In the end, each process has its own specific shortcomings. The key drawback of colour internegative duplication (predominant in the analog domain and the only one available to duplicate stencil coloured films) is that it tends to reduce, and sometimes completely eliminate, the difference between tinting and toning, due to the sensitometric characteristics of the duplicating material. Its effect is that most tints tend to appear as tones, thus making it almost impossible to fully discern this key articulation in the use of colours in the silents. 47 On the other hand, while the Desmetcolor process allows a clear differentiation between tints, tones and their combinations, the fact that it recreates the colours in the printing process, rather than duplicating them from the original, has two serious consequences. 48 First, throughout the film and within every shot and frame, the colours appear to be more consistent in a Desmetcolor print than in the original, since they are applied with modern, more precise printing techniques that do not reproduce the colours with their typical inconsistencies (due to technical

12 78 FILM HISTORY Vol. 21 Issue 1 (2009) Nicola Mazzanti limitations or subsequent fading or decay). Second, the process allows the restorer to freely change the colour applied to a shot. 49 Obviously, this is a great advantage in the context of restoration (as when parts of a film exist only in black-and-white, or derive from different versions using different colour schemes), but it leaves the scholar with a doubt that risks undermining an analysis of the colours: Am I watching the film in its original colours, or am I watching a restorer s interpretation? To some extent, the latter is almost invariably true. 50 In conclusion, no current duplication process ensures the correct colour reproduction of an original silent print. Consequently, whenever possible, the original elements should be accessed and studied, and if this is not possible it is critical to ascertain which duplication process was used, and to account for the distortions it has inevitably introduced. 51 A safer corpus to analyze: the Film d Arte Italiana Daring and complex as it might be, venturing into a screening room to be confronted with the films appears to be the most effective way (or perhaps the only way) to try and understand the way applied colours were used in the first three decades of cinema. In fact, although we have found clues hinting at how applied colours built a specially strong and complex relationship with their audiences, we find it extremely difficult if not impossible to locate primary sources defining this relationship. Furthermore, while we have sufficient information about the techniques used at the time (we know how films were coloured) and we are fully aware that applied colours, as opposed to natural colours, were always the result of a deliberate choice, we largely don t know whose choice this was, and what considerations it was based on. We sense that the intent was driven by many diverse, concurrent and intertwined elements (realistic, narrative, figurative, plastic ), as well as by an interwoven fabric of relations between colours in other films and in other media, and by a context of notions, conventions, common feelings (that is, meanings) which cinema shared with its spectators. Still, we see this as a largely uncharted territory (hic sunt leones aut colores). Besides, if we visit a library searching for answers in the literature of the time (books, handbooks, trade journals) we are left unsatisfied. Almost invariably, whenever we find references to the function of applied colours, we meet with very uninspired and uninspiring attempts to force it into reassuring canons either naturalistic (blue for night, yellow for daylight or was it green?), or psychological (red for passion, pink for romance or was it the other way around?), and this fails to be really convincing and satisfying. On the one hand, the normative effort is too open, and on the other, it does not account for the complexity of the effects and the variety of functions shown in the films, particularly in the cinema of the second period when colour was used in ever more complex and differentiated ways. In view of all the previous considerations regarding the difficulties of analyzing silent films in terms of their use of applied colours (instability in the narrative structure and image characteristics, distortions introduced by decay, errors, human interventions and duplication), an ideally safe zone from which to start our research would be one consisting of works of which we know the original form in terms of narrative structure (mise en scène, editing, intertitling, etc.) and colour scheme(s), and these should also be documented by sources other than the surviving prints. Furthermore, for our analysis to be significant, these requirements should not only be met by one isolated film, but by a coherent corpus of works of which the production and distribution history are known (for example, several films produced by one company, or in a given period of time, or country). Such safe zones are hard to find, and we are too often forced to work with isolated examples and/or prints for which we cannot, then, determine with certainty to what degree their characteristics are consistent with the original intent. Fortunately, although rare, such safe areas do exist. One such promising area of research is that of the Film d Arte Italiana (FAI), which was the object of a complex, multi-year restoration project by the Cineteca di Bologna (in collaboration with the Cinémathèque Française and other archives) that led to the preservation and restoration of a vast number of films. From the perspective of studying the use of colour, FAI s production offers a number of advantages. First, FAI films were mostly produced in the key period between 1910 and Second, a significant number of films survived: of the 154 titles produced between 1909 and 1917, seventy-seven are conserved in some form, either as negatives or positive prints. Although we do not have information for many of the films colour schemes (many are negatives of stencil coloured films, others have been preserved only in black-and-white), reliable informa-

13 Colours, audiences, and (dis)continuity in the cinema of the second period FILM HISTORY Vol. 21 Issue 1 (2009) 79 tion is available for at least some sixty titles. In some cases, we derived the colour schemes from extant positive prints conserved in many archives around the world (FAI s films were widely distributed by Pathé). When only negatives survived, as in the large collection at the Cinémathèque Française, most still retained the original information regarding the colour and the colouring process to apply to each shot. In other words, not only do we have the colour schemes, but we derive them from first-hand unambiguous information dating back to the original production. Also, as we know from FAI s history, all its films were processed (negatives developed and positive prints manufactured) in Pathé s laboratories near Paris, and this gives us a rare first-hand glimpse into the production and colouring workflow of one of the most important production companies at the time, highly renowned for its colouring processes. Film d Arte Italiana was incorporated on 2 March 1909 as a key component of the larger Pathé strategy to strengthen its presence and position in the Italian market against growing competition from local companies (Ambrosio, Itala, Aquila Film and Cines). In case the name is not telling enough, the company s act of incorporation provides further evidence of the role Pathé played in the life of the new studio. Charles Pathé was a shareholder, member of the board of directors, and president of the company; the largest shareholder (with one third of the shares) was that same André Laffitte who had founded Le Film d Art almost exactly twelve months earlier. Laffitte would also guarantee the contracts with both Pathé and Film d Art. The former would supply film stock as well as services to process and sell FAI s negatives, while the latter was to provide artistic and technical guidance, and allow the use of its (translated) brand name. In short, without the contribution of Pathé and Film d Art, the company could not be founded, as is openly admitted in the act of incorporation. 52 In just a few years, FAI became an important component of the Italian film scene. After a short initial phase in which its production strategy would imitate that of Film d Art, FAI would soon release a varied range of genres: comedies, bourgeois dramas, costume films and adaptations from classics (Schiller, Shakespeare, Manzoni, Dumas fils ) featuring original Italian scenery. FAI s success was also secured by its two most active directors and script-writers, Ugo Falena and Gerolamo Lo Savio (later joined by Eugenio Perego, Baldassarre Negroni, Ubaldo Pittei and others), and by a score of actors, a clever mix of stars from the legitimate theater and such newcomers as Francesca Bertini, Vittoria Lepanto, Ermete Novelli, Giovanni Pezzinga, and later Maddalena Céliat, Guido Brignone, Hesperia, Stacia Napierkowska, and Ettore Berti. In its first year FAI produced only four films, but soon after production increased steadily (20 films in 1910, 23 in 1912, 33 in 1914) until the company was hit by the war. In 1915 it released only seven titles; a brief recovery in 1916 (13 films) was then followed by a final crisis with only six titles produced in In the meantime, Pathé was also facing the effects of the war and decided to disinvest its Italian branch. In 1918, FAI s remaining shareholders realized that without Pathé Cinéma s support, there was no future; they sold the company, its studios and its brand to Tiber Film, thus ending FAI s history as an independent entity (although another eleven films were produced under its name until it finally disappeared in 1924). Most FAI negatives have come down to us in their original state, small rolls containing mostly one shot, sometimes more [Fig. 1]. Invariably, each roll is protected by a short leader bearing the data necessary for colouring and positive cutting: title of the film, number of the shot, length of the shot in feet and frames, and a code consisting of letters and numbers to identify the colour. It is hard to imagine how the production process at Pathé could operate with such scant information written on only a few inches of clear film. In fact, at the time FAI s films were produced the process of manufacturing coloured positives was extremely complex. An operator would pick up the small rolls of negative and print them. After printing, he would hand the exposed positives to another worker who would take them to the processing department. There, the positives were manually wound on wooden racks, processed (by immersing them by hand into developing and fixing baths, the processing time being evaluated by sight ), washed, rinsed and assembled together on large drums to be dried. After drying, the developed shots had to be separated again, rewound (so that the short identification leader would stick out), and finally taken to the colouring department. There, by carefully following the instructions on the leaders, the small rolls were sorted out by colour and colouring process [Fig. 2]. Again, each roll would be wound on a rack, immersed in one or more tinting baths, washed, rinsed, dried, and rewound. Whenever a tint and tone com-

14 80 FILM HISTORY Vol. 21 Issue 1 (2009) Nicola Mazzanti Fig. 1. Film d Arte Italiana negatives in small rolls, as they were originally arranged. Each roll contains a few shots. bination was called for, the whole procedure had to be repeated twice. Shots requiring stencil colouring would have to meet their corresponding stencils (one per colour) and run through a stencil-colouring machine as many times as there were colours to be applied. Finally, the small rolls of positive film, now fully processed and coloured, were rewound and separated again, divided by title, and delivered to the right positive cutter, the one person, among dozens in the factory, who was in charge of splicing together a print of that specific title. In this final stage, the positive cutter would splice all the film s shots in the right sequence, and add the intertitles appropriate for the version he or she was working on (intertitles that, in most cases, the positive cutter could neither read nor understand because they were often in a foreign language). 53 Such a workflow would already be an organizational and technical nightmare if applied to only one film at a time. But the Pathé laboratories would process several prints of several different titles, each of them in several versions, all at the same time and all based on the same scant information written on that short piece of clear film. Without a great deal of precision, reliability and efficiency, such a complex workflow had no chance to function. When it came to the restoration of the FAI titles that had survived only as negatives the majority the key challenge was to break a colour code for which there was no Rosetta stone (such as a written document, or a coloured positive print matching an existing negative). Having such a large corpus of films available, the process was time-consuming but not extremely complicated. 54 The interpretation of the code allows one to know the exact colour and the colouring process intended for each shot. Thus, the colour scheme of most films could be reconstructed exactly and without any possible doubt about the real intent, since the information does not derive from a print that could have been manipulated or whose colours could have decayed or could contain errors in the colouring process. Unfortunately, the code does not help to determine the exact characteristics of a specific colour: we may know for a fact when a sepia tone was to be used, or when two different blue tints were planned, but we cannot be sure what type of blue or sepia was intended. FAI s (and Pathé s) code is relatively straightforward. Toning is indicated by capital letters alone or accompanied by small letters as qualifiers (for Fig. 2. A short length of leader from a Film d Arte Italiana negative indicating title, shot number, and shot length. Colour information for tinting or toning is absent because stencil-colouring was intended here.