Restoring Fantasies / (Re-)Creating Realities: Towards a New Dispositif of the Digital Restoration of Reversal Films

(1)

MA Heritage Studies: Preservation & Presentation of the Moving Image

Restoring Fantasies / (Re-)Creating Realities -

Towards a New Dispositif of the Digital

Restoration of Reversal Films

Master’s Thesis Manuel Goetz, 11104732 Supervisor: Mark-Paul Meyer

(2)

2

Acknowledgements

First and foremost, I would like to thank Mark-Paul Meyer for his inexhaustible help and support throughout the entire research process. I further thank Eef Masson and Giovanna Fossati for making possible the study film archiving in the first place, and their continuous academic support throughout the course. I owe gratitude to Gerard de Haan, Frank Bruinsma, Alan Marcus, Albert Edgar and Brian Pritchard for their advice on technical issues. Thank you Tulta, Nadja, Jasper and Nick for the community, your comradeship and spiritual guidance.

Bill Brand has been a great mentor, beyond film archiving. Will Sweeney, Elena Rossi-Snook, Steve Cossman and Ron Magliozzi contributed substantially to the coming into being of this work, with their willingness to share knowledge and expertise. Thank you, Monica, Dessane and Ted for conversations about immensity (and political correctness). Thanks to Janneke, for invigorating lunch breaks, and to Raoul for his trust. Thank you, Max, for the late-night scanning session.

Above all, my deepest gratitude belongs to my family and friends, for putting everything in the right place. Thank you, Steffi, for my idea book and its completion. You will see it filled with blood, sweat, tears and Käferbohnensalat.

(3)

3

List of Figures

Figure 1. Scan of a negative color image and the same image with the luminosity

inverted. 5

Figure 2. Scan of a reversal slide. 5

Figure 3. The characteristic curves of a color negative film (Eastman EXR 50D), and color reversal film (Kodak Kodachrome 64), as retrieved from

Kodak data sheets. 23

Figure 4. Illustration of Sharma’s concept of a profile connection space. 27 Figure 5: Illustration of digital restoration workflow with the overlooked field of

excessive metadata. 51

Figure 6. The workflow of the restoration of American Dreams (lost and found). 57 Figure 7. An E-6-processed reversal image and a cross-processed (C-41) reversal

image. 57

Figure 8. Original American Dreams frame of faded reversal element and graded scan

after color correction. 58

Figure 9. Setup of conducted experiment. 68

Figure 10. Sensitivity of Kodachrome. From left to right: blue layer, green layer,

red layer 69

Figure 11. The slide used for the experiment. 69

Figure 12. The red and green scans, before any correction applied. 70 Figure 13. The black-and-white color topography image of the green and red layers. 71 Figure 14. The black-and-white records of green tinted magenta and red tinted cyan. 71 Figure 15. Comparison of experimental outcome image and original starting slide. 72

(5)

5

1 Introduction

“Projecting a color reversal original induces a superb high or, if you’re lucky, hallucinations. [...] There are other mediums better suited to graphic imagery, but the sensual qualities of color reversal are what free my imagination."1

Figure 1. Scan of a negative color image and the same image with the luminosity inverted

Figure 2. Scan of a reversal slide. Consider the different color patterns in comparison to the inverted negative.

1.1 Reversal Film

With the emergence of photographic film, mechanical reproducibility became not only a potential but in essence a necessity for the technology to fall into place. Throughout much of its history, photochemical photography and motion picture film were negative/positive

(6)

6

systems: the chemicals coated on the surface of the piece of film exposed to light reversed the luminosity and color spectrum of the imprinted object and in order to gain a positive image, these recordings needed to be printed (the act of transferring images on negative film onto positive stock) or otherwise shone onto a new, second material. Besides this method, a different process emerged at the beginning of the 20th century, one that was about to allow photographers (and film makers) to easily and cheaply produce direct positive photographic images. 2 This technology, referred to as slide film, positive film, or reversal film produces a positive, projectable image on the same photosensitive layer of the film that was previously exposed to the light of the captured scene in exposure. Further printing or enlargement on positive photosensitive surfaces (such as photographic paper) was no longer necessary for viewing. As motion picture film, reversal stocks have almost exclusively been assembled into 8mm, Super-8 and 16mm configurations; formats generally referred to as 'small gauge', with 16mm being located somewhere between small and large gauge film. This thesis intends to contribute to the field of the restoration of these materials.

1.2 Film Restoration after the ‘Digital Rollout’

The institutions involved in the film heritage sector have a shared interest: to preserve and carry on the film artifacts of their holdings and their associated viewing experiences, of the past and the present, into the future. This is done either passively, in that the medium’s carriers are stored under certain conditions that are thought to be ideal for a maximum longevity, or actively, in interfering with processes of deterioration and decay, in order to restore a certain previous state of the object.3 The term 'restoration', a procedure within active preservation, encompasses the creation and juxtaposition of two different materials, at the

2_{The first practical and fairly successful process was Autochrome, developed by the brothers Lumière in 1907.} See: Coe, Brian. Color Photography. The First Hundred Years. Vol. 1. Sussex: Ash&Grant, 1978. p. 52f.

3_{UNESCO. United Nations Educational, Scientific and Cultural Organization. Audiovisual Archiving.}

Philosophy and Principles. By Ray Edmondson. Third ed. Bangkok: UNESCO, 2016. Web. 20 Dec. 2016. p.

(7)

7

beginning of the investigation, the so called 'original' and at the end, when the new version of this original is released.4 In archive practice, the artifact created in the restoration is

understood in reference to the original, which functions as both the starting point and final benchmark for any restoration project.5 It is aimed in this thesis to bring forward an

understanding of the ways in which the two products are compared, with the specific case of reversal motion picture films. Ultimately, it intends the formulation of a critique of these current modes, accompanied by suggestions for possible revisions.

Film and its associated industries have been facing the medium’s biggest change in history, with film production and distribution turning almost in its entirety towards digital

technologies, within only a few years.6 For film production, this process firstly brought about the development of what is coined the 'Digital Intermediate' workflow, which found its peak use before the transition to all-digital workflows started, about one decade ago.7 In this process, the film is shot on analog film, the original camera negatives digitized, the digital files then edited, corrected, retouched, etc., and ultimately recorded back to film for cinema distribution, or output to a different distribution format if no cinematic release is intended.8 It is this processes that film archives appropriated heavily for their use in the recent past, because the overall majority of their holdings is photochemical film, while most of the restoration work today is taking place digitally. The implementation of these practices took place in institutions with long histories of analog film preservation and was to considerable

4_{Read, Paul, and Mark-Paul Meyer. Restoration of motion picture film. 1st ed. Vol. 1. London:} Butterworth-Heinemann, 2000. p. 1.

5_{Edmondson, Audiovisual Archiving. p. vi.}

6_{The experimental and partial use of digital cinema technology, in particular for big-budgeted Hollywood} production started in the 1980s, or even a bit earlier, with Francis Ford Coppola's Apocalypse Now in 1979, which, while shot entirely on analog film, reveals a leaning towards what was about to come; Coppola had the digital turn manifested in his revised Apocalypse Now Redux. However, essentially it was the late 1990s and 2000s, which saw the profound changes happen. It is argued, that the process was finished roughly in 2010. See: Fossati, Giovanna. From Grain to Pixel The Archival Life of Film in Transition. N.p.: Amsterdam U Press, 2011. p. 34ff.

7_{Belton, John. "Painting by the Numbers: The Digital Intermediate." Film Quarterly 61.3 (2008). p. 58.} 8_{Fossati, From Grain to Pixel. p. 35.}

(8)

8

extent a result of the powers of outside forces, such as the Digital Cinema Initiatives.9 This employment met existing practices that have been developed at the institutions over long periods of time, resulting in specialized, concentrated expertise. The so-called 'digital rollout' brought with it the sudden availability of post-production tools for film restoration means, scrutinizing extant positions, practices and theories of archival institutions. These

mechanisms required a sudden production of new knowledge and skills.

Among the archivists, the new archival techniques stifled hope to more effectively and efficiently being able to tackle preservation and, in particular, restoration issues that before the digital turn entailed very labor and time intensive procedures. While this proposition has certainly seen the promises of this technological change implemented in some aspects of archival work, in particular in new medium access models, due to the speed and far-reaching extent of the recent repositioning, archivists could not yet develop an expertise comparable to the analog era. This is particularly applicable when it comes to its influence on proceedings in restoration, where a specialized niche application of digital means has not yet been

developed, as will be illustrated with the case studies presented in this paper. The

practitioners saw themselves exposed to a variety of unique difficulties generated by (and inherent to) this new technology.

I am writing this paper to contribute to the understanding of this specification and help delineate the requirements and potentials of the new situation, in order to overcome prevalent problems in restoration practice. I will do so in pursuing alternative potentials of digital technologies and developing strategies for the archives to utilize these. I aim to outline reasons for why it is precarious to employ these new and ever accelerating technologies

9_{The DCI is a union of the "big six" Hollywood studios (Disney, Twentieth Century Fox, Paramount, Sony} Pictures, Universal and Warner Brothers), formed in 2005. In 2007 they released a revised version of their initial first statement, codifying the guidelines of the conversion to digital cinema production and distribution. This included a list of the favored presentation codecs and carriers. The initiative is considered to stand

emblematically, as the final momentary turnover, for the Hollywood industry abolishing their film production on photochemical film. See: Fossati, From Grain to Pixel. p. 58.

(9)

9

uncritically and solely in their indented uses. I suggest employing comparable methods to those that have a long-lasting tradition, including practices of appropriation, experimentation and adaption. Considering the current position after the digital-rollout, the field has come to a point in time when it is fruitful to explore such situations, as the most hectic moments of this upheaval seem to have been surpassed, allowing for contemplation and reconsideration of what has happened. I believe there is enormous potential to be found in the tracing back and comparison of the habitus of digital technologies to practices of earlier analog

experimentation.10

The advent of digital technologies in the film restoration field coincided with the production of a new kind of professionalism, rooted in academia.11 It is now after film as an industrial commodity has abolished its purpose to serve as a commercial mass product, when two generations meet in the archive: one that has known film as an entertainment device, technically based on analog photochemical processes, and one that is informed about this past, but is also accustomed to film's functions beyond and outside these logics. This paper explores the potentials for change that arise in this constellation, marrying existing

knowledge about the analog medium with proposed potential of digital technologies, from the perspective of newly established archival theory.

1.3 The Restoration Dispositif

This paper seeks for insight about film restoration practices in utilizing a comparability and hermeneutics of the crafts of filmmaking and film restoration as disciplines of interpretation

10_{The term habitus was shaped by French philosopher Pierre Bourdieau, who defined it as a total sum of a} human’s social actions, preferences and manners. See: Lizardo, Omar . "The cognitive origins of Bourdieu's Habitus." Journal for the Theory of Social Behaviour 34.4 (2004). Web. 19 May 2017. p. 375. The habitus of a technology is thus the entirety of its use, its application and intended relations to other machines and the human operator.

11_{Several academic courses were created around that time, for instance the Selznick School of Film} Preservation at the University of Rochester, N.Y. in 2005, or the Moving Image Archive Studies at the University of California, Los Angeles in 2002.

(10)

10

of an outer world.12 What is for the film maker the interpretation of a reality surrounding him/her is for the restorer the interpretation of a similar reality, the presence of an artifact, the film original.13 The conception of the restorer in this paper will therefore be understood as operating along similar trajectories, in that s/he too works with a given set of preconditions that have to be deciphered and translated into a new medium. Filmmaking is thus the first discipline extrinsic to film restoration that the author will appropriate for his discussion about film restoration.14

In a film lifecycle, at all stages processes of understanding as interpretation can be found: in the initial capturing of a scene onto a carrier, the interpretation of the incoming array of light by the film emulsion, in post-production, where certain elements of this capture are enhanced and others repelled, and finally in projection from a carrier, which is in return deciphered by an audience. Analogous, after the film's initial exploitation, in the archive the restorer is responsible for a film's translation into a new form. At the end of every archival process stands the human agency that interprets both the film original as well as the restored version and weights them up against each other. None of these processes is freed of individual judgment. Every result of a man/machine interaction, be that in the making or restoring of a

12_{Hermeneutics means a practice that investigates the interpretation and understanding of texts through the use} of symbols. It operates with and utilizes for a delineation of the understanding the elements the text at hand operates with. See: Mir, Raza A., Hugh Willmott, and Michelle Greenwood. The Routledge companion to

philosophy in organization studies. London: Routledge, Taylor & Francis Group, 2016. p. 113.

13_{It is an annotation worth alluding, that both these practices, filmmaking and film restoration, in the time of} their onset have been mistaken for duplications of reality, presumably creating identical copies. Widely known is the example of the early documentarist, who scared his audience with the perceived reality manifesting on the film screen: the train approaching the station in the film narrative was perceived as a train breaking the

boundaries of the silk canvas and actually approaching the audience in the presentation hall. Likewise,

nowadays (in the early days of digital film restoration) the discussion if it is appropriate to display digital copies of historical films without mentioning that this is not the original, is omnipresent. See, e.g.: Horwath,

Alexander. Film: Was bleibt? Sylvia Winkelmayer im Gespräch mit Alexander Horwath. Vienna: Austrian Film Museum, 2017. p. 4f.

14_{It is, of course, slightly condemnatory, to consider filmmaking an occupation alien to film restoration, as the} output product of the first becomes the starting product of the second. Also, quite simply, in the heydays of independent analog filmmaking, many film makers (even though that term did not exist back then) were also film restorers. Thus it is surprising that these two disciplines have hardly been thought of in the same realm. Therefore, and for matters of argument, my paper will strive for resemblance of filmmaking and film restoration. Further external disciplines shall be consulted in this paper and will be presented in subsequent sections.

(11)

11

film calls for acts of intrusion that are highly affected by the standpoint of the practitioner in terms of knowledge, skills, background (institutional frame, aim of project) and more. This is why I believe it is fruitful to analyze the film restoration craft with the same set of principles that the field of film studies facilitates for analyses of film expressions. From there it follows that in line of this argument this paper will conceptualize the film restorer as an auteur.15

As flat interpretations of reality, photochemical films are both tangible (as materials) and intangible (as performance). Its tangible character means that film is translatable into

descriptive, mechanical systems. As an art form, film, in the moment of projection becomes a perceived reality in its own right, transcending a descriptive language. These two figurative concepts operate intertwined within what might be framed as the dispositif of film

restoration.16 The restoration of film takes place in a field of powers rooted in the basic principles of film technology; the medium has been imagined at all times with a dependency on external machines for the fulfillment of its indented function (its apparatus). Filmic capture, development, production, handling and display all rely on machined transcoding. At the same time, film encompasses a performative character and is an ephemeral expression that manifests in the brains of the viewers in a display enabled and performed by machines. This relies on human interpretation, as only humans are able to experience and relate individual expression, and make sense of the performative act. This act, while elicited by machines, takes place outside of them in a second space (the physical space of the cinema and the abstract space of the mind). This is the very concept of projection. For Jean-Luis

15_{In a filmmaking context, the term auteur, leading back to the French Nouvelle Vague and the film critics} André Bazin and Alexandre Astruc, means a director who has the full control of all creative aspects of the making process. All decisions culminate in the single person of the film maker, who has full control and responsibility of the final film product. See: "Auteur theory." Encyclopaedia Britannica. 1998.

16_{Dispositif, first appropriated as a sociological concept by French philosopher Michael Foucault describes the} presupposed and unspoken conventions within a certain discourse, which result in corresponding social actions. The term puts in its center the power structures that are at play in the process of forming these discourses, and studies resulting from this term investigate the distribution of power and authority. In the broadest sense,

dispositif describes the habitus of a society. See: Deleuze, Gilles. "What is a Dispositif?" Michel Foucault Philosopher. Hempstead: Harvester Wheatsheaf, 1992. 159-68.

(12)

12

Baudry, the cinema apparatus, the appareil de base, the technology and actions needed to produce a film include the projection dispositif – according to him, the projection event is the final stage of the fulfillment of the appareil de base, and includes the human agency, the spectator.17

The dispositif of film restoration functions in an analogue setting: its quest is to create appropriate technical substitutes to its material originals and enable performative features comparable to them. As Fossati outlines, when preparing her discussion about the different conceptualizations of film archives: “[In the film as dispositif framework] film identity becomes a variable that realizes itself only within a dispositif, a situation if you wish, where the film meets its user. From the perspective of the archive this is certainly an interesting approach as it allows for a different way to look at films, namely, as dynamic objects where the material and conceptual artifacts are bound together.”18 This conception places the act of film exhibition in the foreground.19 For the case of this investigation, I will now look at the implications of this notion when applied on the work of the film restorer.

As machines cannot experience ephemeral expression (because it is felt rather than

empirically understood, in the aforementioned second space),20 this process will always only be an endeavor of subjective approximation of the restorer. What is thus at hand is the interplay of technical algorithms and an individualized, situational appreciation. As outlined, the interplay between a technological apparatus and a human operator is crucial for the entire medium at all times. This paper seeks for modes of film restoration that do not aim for a translation of their filmic 'nature' for a second interpretation outside its allocated machine space, tailored to the person of the film restorer. Instead it looks for and aims to enable

17_{Baudry, Jean-Luis. L'Effet cinéma. Paris: Éditions Albatros, 1978.} 18_{Fossati, From Grain to Pixel. p. 127.}

19

Ibid. p. 126.

20_{Referring back to an earlier explanation of hermeneutics, this means that a hermeneutical analysis of the data} production of a machine would be an entirely different study, with different tools than an analysis of human grammatical understanding in a hermeneutical light.

(13)

13

situations that do not entail a necessity for the decoding of immanent machine language. It raises the question whether the technological properties of the film restoration craft are not to be employed differently - without translating or decoding their being - for a second, human interpretation.

It is this area of conflict that both the film maker and film restorer have to navigate: Does our perception of film depend on any form of materiality and what happens to the subtleties of a specific expression if rendered into another medium? Film restoration as a theoretical concept is underpinning and resulting in the medium's technological manifestation. The restorer is arranging the various versions and elements into a revisited original, with the help of machines, and occupies the role of an intermediary between expression, the material this expression is captured with, and its decoding mechanisms. This is why the film restorer in this paper will be also conceptualized as a negotiator; as negotiator of an object's (imagined) previous lives, between its various elements and generations, between the film original(s), its machine interpretations and associated performative characters and the audiences in the present. The restorer as negotiator aims for the culmination of these arrangements, and for finding implication in an object's material and immaterial characters.21

For the restoration dispositif, the twofold positioning of the restorer as auteur and as negotiator of worlds means the following. Analog films enter a different medium in digitization, a new technological and cultural world. Recreating something in a different medium that is identical to the original is impossible.22 The human agency in person of the film restorer converges with the gap that resides in the translation phase of analog to digital, in obtaining a subjective position at every moment of interaction with the material. As I will show, in particular the case of color restoration can illustrate these ontological differences,

21_{Pescetelli, Marco. The Art of Not Forgetting. Towards a Practical Hermeneutics of Film Restoration. Thesis.} University College London, 2010. p. 301.

(14)

14

which account for not only insufficient translatability but also reveal the need for a renewed appraisal of the relationship of these two mediums. I will argue that this inherent divergence calls for strategies in moving image presentation that communicate and educate about these distinctions, in order for an institution to contribute to a critical thinking in a visually pervaded society. While it can be claimed that the all-encompassing incompatibility of the two mediums inevitably accounts for a loss in translation, a different perspective, one that should find attention here enables a perspective that sees the frictions not in the light of any form of qualitative inferiority or superiority but as a convergence whose reception simply exceeds and transgresses beyond subtleties of the initial norm input. It is the intent of this paper to locate this access value.

The concept of the film restorer as an auteur is needed to attain an autonomous position, one that acknowledges subjectivity and a personal handwriting with which the materials are arranged, decoded and remediated, according to an idea derived from an original object. This then enables the restorer as a negotiator to acquire authority, to achieve a decisive position within the gap of analog-to-digital translation models, to navigate between presupposed notions of medium specificity and to acquire potentials from the different remediations of the object at hand. I want to extract and delineate new possibilities for both analog and digital technologies based on the belief in their intrinsic disparity. Going with Ross Lipman, whose writings this work owes great appreciation to, I believe there is the need to overcome the digital/analog dichotomy that seems to allow only strong passions or rejection.23 It is the aim to develop a stance where one can appreciate the forms themselves, as such, with their unique characters and qualities.24 The way I strive to achieve this position is through developing a research model that incorporates technical and aesthetical concerns, the physical and

23

Lipman, Ross. "In Search of Sight-Specific Cinema." The Moving Image 12.1 (Spring 2012). p. 100. 24_Ibid.

(15)

15

conceptual side of film.25 The conjunction of the models of the restorer as auteur and negotiator thus allows for this renewed appraisal of digital technology within the film restoration dispositif, in that it repositions the practitioner outside the notions of an image provider, creator of content, which are capitalist conceptualizations that automatically subject the practice to industrial, commercial and consumerist dichotomies harmful to the

responsibilities of film heritage institutions.

In summary, this thesis aims to find out how digital technologies transform an established film restoration dispositif and where to locate the potentials of this new situation - in order to expand upon a notion of authenticity, in particular with regard to the color reproduction of reversal films.

25_{Lipman, Ross. "Technical Aesthetics in the Preservation of Film Art." Big As Life: An American History of} 8mm Films. New York and San Francisco: Museum of Modern Art/San Francisco Cinematheque, 1998. p. 88.

(16)

16

2 Methodology

"I have described our current situation as a ‘bifurcated path’. […] [T]he first of these two strands […] appears to be expanding, thus inviting substantial hope as far as the future of the discipline is concerned. We could call it the scientific or historical-materialist approach. It is in dialogue with theory and scholarly tradition but also engages, on a very concrete level, with the material qualities of archival and museum holdings."26

In this chapter, I will illuminate the system of investigation that this paper draws upon in order to investigate the subject of research as brought forward in the preceding chapter. At that point, I have already introduced the restorer as auteur and the restorer as negotiator concepts, which form an ideology driving the methodology.

In line with the quote at the fore of this chapter, this paper will make use of three major pillars in its methodology: the advent of a new archival scholarship and its adjacent discourses, tools from the digital humanities derived from the natural sciences, and film restoration theory. As my paper is aiming for a reconsideration of the application of digital tools within the film restoration craft, it in return intends to suggest a new methodology for the field. Chapters four and five will introduce and utilize this new methodology, which includes disciplines such as media archaeology27, imaging science28, hermeneutics, as well as classical arts restoration scholarship. The paper will draw from the restoration of colors

26

Horwath, Alexander. "The Old Life. Reframing Film "Restoration": Some Notes." Journal of Film

Preservation 96 (April 2017, p. 28.

27_{Media Archaeology is a relatively new academic practice (or rather a loose field than a defined academic} methodology, leaving its mark also in arts practice and media production, for instance) that does refuse

progress-oriented, straightforward narratives of the historical development of various forms of media and rather investigates this overlap, incorporation and incongruencies of media history, as well as forms that failed or were commercially unsuccessful. Important publications defining the field include: Hui Kyong Chun, Wendy, and Thomas Keenan, eds. New media, old media: a history and theory reader. New York: Routledge, 2015., Parikka, Jussi. What is media archaeology. Cambridge: Polity Press, 2012., Natale, Simone. "Understanding Media Archaeology." Canadian Journal of Communication 37.3 (2012): 523-27.

(17)

17

reversal films its specific case studies because I believe this type of material highlights many difficulties of the restoration craft exemplified through its technological nature, while at the same time offering a fruitful point of entry into the following positioning, which I regard intrinsic to the film medium and film archival work.

I will consider film restoration as a craft operating for and within realms of the arts, industry and science. I think it is necessary to consciously investigate its practices from an

interdisciplinary position, considering that the craft itself borrows its tools from various disciplines located in the three areas just identified. I anticipate from this position to be able to analyze current proceedings of film restoration, in achieving a vantage point to be able to criticize and formulate changes to these practices, from a respective position.

2.1 Why Color?

Color, functions as a gateway for such an investigation in that it precisely occupies the fields of art, industry and science. Color perception is highly subjective – it depends on a viewer's past experiences as well as biological preconditions.29 It is intangible (in observation) and measurable (in the laboratory). Color occupies the field of the empiric as well as the field of the speculative. Hence it inhabits the same discursive field as film technology, as brought forward in chapter one. This constellation allows the drawing of analogies from the

discussion of color restoration to the power structures of a film industrial discourse, including the archive as a field of power.30 This is a trajectory the investigation here aims to follow.

The verbal cannot sufficiently describe the aesthetic. While this is the basis for the

appreciation of the visual arts (and film) as artistic expressions, it reroutes a purely scientific approach towards these products. For me, and for the case of this paper, this calls for a hybrid

29

Flueckiger, Barbara. "Color and Subjectivity in Film." Subjectivity across Media. Interdisciplinary and

Transmedial Perspectives. New York: Routledge, 2016. p. 147.

30_{Derrida, Jacques. "Archive Fever (transcribed seminar)." Ed. C. Hamilton, V. Harris, M. Pickover, G. Reid,} R. Saleh, and J. Taylor. Refiguring the Archive. Berlin: Kluwer Academic Publishers, 2002. 38-78. Print.

(18)

18

model, of an entanglement of scientific and artistic means, following the notion that

"[s]cience is both the basis of a laboratory’s standards and the background for its experiments in art."31

The majority of problems related to the restoration (and storage) of reversal films is to be attributed to technical factors intrinsic to the material. Considering that both traditional photochemical workflows as well as newer digital restoration techniques have been designed intentionally for the negative/positive system, there needs to be a search for the potentials of these technologies when it comes to the archival treatment of reversal film. This situation, first of all, requires asking technical questions. Second, it calls for the search for potentials of technologies outside their proposed uses. Third, with the newly established tools of

scholarship at hand, the digital predominance can now be interrogated from revived

perspectives. And last but not least, the prevalent situation demands a look beyond the realms of film archiving theory, as its major players, the technologies employed, have been initially developed and first used elsewhere. Hence, it is needed to reframe what these discussions reveal, into a new, interdisciplinary structure that integrates questions and concerns from the fields of materiality perception, audience reception, philosophy, ontology and more.

In order to assess the practical feasibility of the proposed model for current film archival practice, the paper will bring forward examples of digital reversal film restoration It will do so in presenting both cases that employ digital technologies 'traditionally' (considering that it is difficult to speak of traditions in this rather recent development, traditionally here refers to a use that accustomed as best practice over these few years), and examples that aim to utilize an experimental approach towards its original source. In doing so the case studies should

31

Paletz, Gabriel M. "The Finesse of the Film Lab: A Report from a Week at Haghefilm." The Moving

Image 6.1 (Spring 2006). p. 12. He refers in his paper to film laboratories. However, these formulations can also

be applied to the laboratory as conceptual model, as an over-encompassing term for an institutionalized situation of research and investigation.

(19)

19

exemplify some of the problems previously outlined as well as allow to be analyzed according to potentials for theoretical implementation of my proposed framework. It is

sought to approach the case studies from a viewpoint as suggested here, and examine possible advantages and disadvantages with respect to the cases at hand. These observations should then fuel formulations about revisions and additions of the proposed model, in reference to and in dialogue with ongoing scholarly research.

Further, I will present an experiment of the author's own, employing the concept formulated along the progress of this paper. This will be enriched with the examples from actual practice, and evaluated according to its success and feasibility for professional use.

In summary, my paper will contribute the following:

● I will map the practices of reversal film restoration that are in place today, both photochemical and digital, and sketch current problems that are encountered in these processes. In particular, I will examine the conversion from analog to digital

restoration and the new conceptual notions this brought about. I will bring forward how these problems have been tackled in traditional archival work and suggest points of attack for comparable techniques of current and upcoming technological

developments.

● I will suggest changes to the digital film restoration dispositif, aiming for a more appropriate and more exhaustive application of the utilized technologies. I will do so in re-examining strong-held beliefs regarding digital technologies and their function in restoration.

● These observations will provide a clearer understanding of the craft of restoration through an evaluation of the argument from authenticity, which in a traditional

(20)

20

discourse heavily depends on a restoration's qualities of aesthetical reproducibility. I will examine if, facing the eventual disappearance of reversal film stocks, the central role of the original 'look' is in fact crucial. In chapter five a hypothetical experiment will be introduced that suggests a way to foil film industrial logics of discontinuation of certain materials, with the particular example of the Kodachrome film stock. ● Finally, the restoration case studies will be examined through aspects suggested by

the proposed framework. This will evaluate the framework's feasibility under existing preconditions, while pointing to existing conceptual differences in current practices and my proposed model. I will analyze its advantages and disadvantages and, building upon them, formulate possible further additions and recontextualizations for possible future work.

(21)

21

3 Current Reversal Film Restoration Practice

Making way for proposed solutions for the current problems of digital reversal film restoration, this chapter aims to provide an overview of problems arising in archival work with reversal films. This shall be no in depth analysis, but a contextualization of why this topic calls to be extended upon.

The author starts from a growing scholarly recognition of reversal films, as well as a more widespread public appraisal of these forms of film. One main reason for this is recent trends to scholarly discuss amateur films, which almost always, if shot on film, were made with reversal film.32 What seems to be missing is a link from theoretical insight to the new modus operandi digital technology brought about. This is essential because for these types of film, digitalization often means the ultimate and only preservation effort, as photochemical

duplication is (hardly) no longer feasible, which is to be attested in the following. Reasons for why film heritage institutions continually struggle with the technical aspects of the

preservation and restoration of reversal films shall be outlined subsequently. I am arguing that while there is a growing scholarly recognition of reversal materials in the archive, the appropriate means of their technical implementation, and in fact their practical realization, are often dragging behind, or missing. Thus, this paper will focus on the necessities for a wider ranging inclusion of reversal materials, taking into consideration and elaborating on new professional developments. In order to provide a possible solution to the issues at hand, one has to rethink current practices in both, their sheer technical condition and the underlying

32_{Academic publications that illustrate this attention towards the study of amateur film are, among others:} Zimmermann, Patricia R. Reel families: a social history of amateur film. Bloomington, IN: Indiana U Press, 1995., Fox, Broderick. Rethinking the amateur acts of media production in the digital age. Los Angeles, CA: U, 2004., Wasson, Haidee. Museum movies: the museum of modern art and the birth of art cinema. Berkeley: U of California Press, 2005., Fuentes, Maria. Orphan Film: Definition, Value and the Archive. Thesis. University of Amsterdam, 2010.

(22)

22

ideology. As I will show, it is neither truly possible, nor appropriate for institutional practice, to think these two aspects separately.

3.1 Analog Restoration

To begin with, I will examine the possibilities of analog restoration of reversal films that are open for film archives in the present tense. Prevalent problems resulting from the image's contrast are probably the most illustrative and most pressuring, when it comes to this aspect of archival work. They are the prime technical reason (second only to economic imperatives) for why nowadays many film archives prefer a digital route over a photochemical.33 In a conventional positive/negative printing process the contrast levels of the final release print are achieved primarily in the printing and grading of the positive print, and not during the exposure of the original camera negative. The negative intends to flatten out the recorded light, to capture a low contrast, in order to record as much information as possible (a ‘flat’ negative records a lot more light than visible for the human eye) and leave room for the final grade, which then pushes certain luminances, resulting in a more tonal, and therefore more visually appearing image. Conventional positive print stock has a median gamma (a measuring unit for an image's luminance values and indicator of its contrast, with a higher gamma meaning a greater difference between the darkest and brightest values, hence a heightened contrast) of about 1.5.34 In printing from negative stocks, which usually have a gamma of 0.6, this results in prints containing a gamma of about 1.0.35 Reversal stocks too have a gamma of 1.5, however because they are receiving their brightness information not from a 'flattened' negative, but from the incoming 'real' light, they automatically depict luminance values with a corresponding gamma of 1.5, hence they have a heightened contrast

33_{An ethical reasoning is that digitization of 8mm films is considered to reflect better the original characteristics} than a blow-up onto other analog formats (an option of photochemical duplication probably most relevant for small-gauge film) would do.

34_{Meyer, Mark-Paul. E-Mail Conversation. Message to the author. 13 Apr. 2017. E-mail.}

(23)

23

as compared to prints obtained from a negative/positive process. There is no possibility to drastically influence this in the recording situation – and no intent to alter the chemical structure of the emulsion either, as the aim of reversal materials was and is to provide a singular, projectable image that corresponds to a certain visual appearance.

Figure 3. The characteristic curves of a color negative film (Eastman EXR 50D, left), and color reversal film (Kodak Kodachrome 64, right), as retrieved from Kodak data sheets.36

While this might seem miniscule at first, this condition complicates the duplication process due to the comparable loss of information in the reversal originals. The heightened contrast in reversal films is achieved at the expense of luminance subtleties. Any further printing of reversal positive film onto something other than another reversal element, the creation of an intermediate stage, a so called 'intermediate negative' is necessary. Taking into account that this new negative is obtained from a print with a heightened contrast, a considerable amount

36_{"KODACHROME 64 and 200 Films." TECHNICAL DATA / COLOR REVERSAL FILM. Eastman Kodak}

Company, June 2009. Web. 21 June 2017. <http://125px.com/docs/film/kodak/e88-2009_06.pdf>, "EASTMAN EXR 50D Film / 5245, 7245." TECHNICAL DATA. Eastman Kodak Company, May 2003. Web. 21 June 2017. <http://www.kodak.com/KodakGCG/uploadedfiles/motion/H-1-5245t.pdf>. As color films are sensitive in different degrees to the red, green and blue spectrums of light, for matters of comparison it is fruitful to examine the individual sensitivity curves for each color layer. The reason for why the negative curves ascent in the opposite direction is due to the fact that these materials translate the incoming light spectrum into its correlating complementary colors, while reversal film directly create a positive image.

(24)

24

of information will be lost, as it does not receive the same amount of light information as it would have in exposure to natural conditions. Hence, any analog restoration product of a reversal original is inferior in quality to the original.37 This is why in many cases such efforts have been done only for archival storage purposes, without the production of a new

projection element.

3.2 Digital Restoration - Scanning in the Rift between Analog and Digital

Preservation efforts in the digital realm, naturally also have to face distinctive problems. Early on into the research of this project, it came clear that the major problem for the all-digital or hybrid restoration of color reversal films is to be found in the digitization stage, the scanning process ,both in terms of its technical improvement (which will come inevitably), but even more so in terms of its appropriate assessment and use.38 Before going into detail of the specific problems that occur there, I will present a reflection on the contextual breadth of the scanning issues.

The process of scanning, which mirrors the analog stage of printing, is the most invasive intrusion into the restoration dispositif. It creates a facsimile, a new product that is referred to as copy, duplication, imitation, or reproduction. It is the moment information is retrieved from an original source and transformed and embellished into a new product. Following form

37

A few negative stocks have been designed specifically for the copy of reversal positives, such as Kodak's Eastman 5251. This stock was also used for the duplication of tinted and toned films. See: Read, Paul. "‘Unnatural Colours’: An introduction to colouring techniques in silent era movies." Film History: An

International Journal 21.1 (2009). p. 25. Fuji's F-64D has provided good results when copied from reversals,

while it was not originally designed for this purpose. See: "Interview with Gerard de Haan (Haghefilm)." Personal interview. 15 Mar. 2016. However, neither of those achieved sufficient (meaning comparable to the negative/positive process) contrast levels. Kodak also did produce positive camera reversal stocks that were intended not for immediate projection but further copy, with Ektachrome Commercial (ECO 7252) being the most prominent. See: Read, and Meyer. Restoration of motion picture film. p. 161. This is characteristic of the widespread appraisal of the look of reversal prints in projection. See: Frye, Brian. "The accidental

preservationist: an interview with Bill Brand." Film History: An International Journal 15.2 (2003). p. 217. It is clear that generally no positive film stock, current or past, has been sufficiently able to duplicate reversal originals. While any printing process means a loss of photographic information, this is particularly severe in the duplication of reversal films.

(25)

25

this intervention, the further work of the restorer is nothing but erasing digital benchmarks and artifacts, resulting from the capture, to reach a stage that can be compared to the original viewing experience; hence: to make the digital disappear. The end of the process sees two images that allow for comparison. This is why the scanning procedure is the seismic shift in the restoration process, the step that calls for the most attention. There are two major issues ever present in the scanning of reversal film, which intertwine technological limitations with theoretical sentiment. These are the appropriate representation of color spaces in the digital realm and the previously introduced issue regarding the heightened contrast in the original prints, which, in the digital realm, causes different problems. Both of these issues are only technical problems at first glance and point towards a larger discussion: the calibration of an institution's entire digital infrastructure. The fundamental origin of both problems refer back to what has been stated previously; the limited compatibility of modern film scanners with reversal films, as they were initially calibrated for the dominant film forms, negative and print stocks only. To begin with, I will address the issue of color spaces.

In order to approach this problem in full breath, it is beneficial to understand the difference between additive color systems and subtractive color systems. Additive color systems add and mix different light wavelengths, mostly corresponding to the red, green and blue sectors of visible light, in order to achieve nuances in the other colors resulting from the degree of overlap of these three primaries. In film practice this meant for both capture and projection that the light rays have to go through rotating filters of red, green and blue to achieve the impression of a color film.39 In 1914 a different system was developed, which similarly recorded light going through filters of red, green and blue but captured the respective

'secondary color', which is cyan, magenta and yellow.40 As this result is achieved in filtering

39_{Flueckiger, Barbara. "Temporal synthesis (rotary filters)." Timeline of Historical Film Colors.} 40_Ibid.

(26)

26

out a certain degree of the incoming light, this model is called 'subtractive'.41 In contemporary reversal stocks, the filters are incorporated in the film carrier, while in the past they were also built in the recording device. In the history of motion picture film, few examples

experimented with additive color systems, most of them in creating specialized capturing and projection apparatuses (such as the rotary filter of the early Kinemacolor system).42 The vast majority, including all reversal stocks, used and use subtractive methods to record and depict color. In contrast, digital image capture technologies, most frequently (with the exception of early experimental digital cameras) make use of an additive color system. The translation of a subtractive color sample into an additive model is not innately lossy, however, it is argued that it is prone to a loss of color information, if not properly encoded through a central profile connection space, which functions as a common denominator in the communication of the individual machines in an image processing environment (such as camera, computer, printer, or scanner, computer, film laser, for example) in incorporating the output device's color profiles into the transcoding of the file in question.43 'Color space' refers to the spectrum of colors (the gamut) a device can depict, and is the integral part of Sharma's device profiles.44 Especially with the advent of digital technology (but already with television), the amount of and the difference among color spaces increased rapidly. The profile connection space that Sharma introduces is a digital color space that is informed about and communicates between both the color space (profile) of the image producing device (a film scanner, for instance), as well as in the later process with the color profile of the output device (for example a film laser, or a hard drive to be put into a DCP-projector).45 Workflows incorporating a central connection space are very rarely found in common film scanning practice; often the captured

41

Hincha, Richard. "CRISIS IN CELLULOID: COLOR FADING AND FILM BASE DETERIORATION." Archival Issues 17.2 (1992). p. 127.

42_Ibid. 43

Sharma, Abhay. "Understanding Color Management." International Psychogeriatric Association

Bulletin (March/April 2005). p. 17. 44_{Ibid, p. 20.}

(27)

27

scans are stored directly on the computer, without rerouting the images through other processors.

Figure 4. Illustration of Sharma’s concept of a profile connection space.46

The biggest challenge for the restoration of color film is that color perception is relative.47 In theory, a scanner's ideal function is to translate the color values of a film sample, that it attains in photographic a projection, into values responding to the three color channels of any visual image, be that on film, in video, or the human sensory apparatus.48 What makes this issue precarious and therefore worthy of consideration is the subjective perception at play. In particular the relation of the color 'appearance' as result of human perception as opposed to filmic/digital representation is one infused with experience, mood and memory, among other things and lacks a congener to a central conversion agency, which here offered a gateway into the discussion. In color restoration (and film grading), technicians employ a twofold approach of merging scientific data of measured colors (depicted through histograms, for instance)

46_{Ibid, p. 17.} 47

Flueckiger, Barbara. "Bridging the Gap between Analogue Film History and Digital Technology." 1st INTERNATIONAL CONFERENCE COLOUR IN FILM. Friends House, London. 03 Mar. 2016. Lecture.

48_{See also: Croci, Simone, Tunç Ozan Aydin, Nikolce Stefanoski, Markus Gross, and Aljosa Smolic.}

(28)

28

with subjective impression. Already in 1949, Hollywood cinematographer John Alton stated that "[n]egatives should be timed first for mathematically correct density, second for feeling, mood. Neither one alone is sufficient. It is an ideal combination of both that makes a good print."49 Paul Read and Mark-Paul Meyer state that quality control of film laboratory procedures "is more objective if the characteristics of this process can be converted into numbers."50 This procedure stands emblematically for the nature of film as being a

mechanized interpretation of a subjective gaze upon reality. Such practices have been in place at film laboratories for a long time, dating back to when the restoration of films was not part of their profile. At the film laboratory, a common example of the incorporation of scientific tools in the subjective craft of color grading is through so called LAD's, Laboratory Aim Density Systems, which compare the density levels of a test wedge exposed in the laboratory before the production of a new film element to a sample provided by the stock manufacturer, presumably perfectly exposed and developed, in order to calibrate the lab infrastructure.51 Paletz explains how this system is further used for grading the final prints:

“Like film developing, the process of grading, or timing, reveals how science supports industry and art. Haghefilm [a photochemical and digital laboratory in Amsterdam with a specific focus on archival jobs] uses a scale of 1 to 50, with one applied to low density, or low-contrast, images of fifty for high-density, or high-contrast, pictures. [...] The negative filmstrip of the LAD has as essential a place in grading as the sensitometric wedge has in developing.[...] When printed onto any positive with the average printer lights (25-25-25, midway between 1 and 50), the LAD should produce a film with color densities prescribed by Kodak, representing the medium densities for every kind of stock. [...] Both the

49_{Paletz, The Finesse of a Film Lab, p. 12.}

50_{Read, and Meyer. Restoration of motion picture film. p. 105.} 51_{Paletz, The Finesse of a Film Lab, p. 10f.}

(29)

29

sensitometer and the LAD system create scientific targets for producing the same results no matter the stock or condition of the machine. [...] [L]ab's technicians view a film image as both a set of standardized values and as a palette to paint particular effects. Science is both the basis of a laboratory’s standards and the background for its experiments in art."52

Of course, from a film maker's perspective, this reveals that somebody else, namely aforesaid lab technician, actually plays a crucial role in the look of the final film. Stan Brakhage, referencing early video intermediate procedures, noted that the actual color hues seen by a viewer are actually the result of a subjective “know-twiddling of an unknown.”53 It is clear that while in the post-production of current features, the cinematographer and/or director are consulted for the color grade, in the case of historical film this is often impossible and at times even undesirable, as a film maker might want to correct ‘mistakes’ that occurred in the initial production, and such practices would go against any form of historical accuracy and authenticity. Such and similar observations speak to the disparity between a film maker and his/her expected aesthetics and the scientific mandate of a film lab, which, as we now know, is not only scientific after all.

When talking about subjective color interpretation, it might be fruitful to establish a link to human perception. It is clear that not every human sees in the same way. Likewise, we do not have a blueprint or scale that we can refer to when discussion notions of individual

perception. Hence, it could be argued that, for the case that is under investigation here, the restoration of color becomes an interpretation of memory.

Following along from there and linking thoughts presented in the previous two paragraphs, Rosenthal et al., in referring to an experimental conceptual model of the human visual

52_{Ibid, p. 11f. Haghefilm is a film lab in Amsterdam that has a strong restoration focus.} 53_{Lipman, In Search of Sight-Specific Cinema, p. 54.}

(30)

30

apparatus, define singular decoding mechanism in our eyes, which are, according to them, not taken into account in the development of current models of digital technology.54 They argue that, and this expands Sharma's concept, no digital color space can accurately reproduce the colors generated through the human visionary system, first because of the already mentioned different representational spaces but also because the human visionary system does not work exclusively along the temporal principles of light (light waves), but might rather use the frequency character of the incoming light (the photons), to process and carry on

information.55 This is not the minor detail it seems to be, as all our imaging technology has been designed according to the wave characteristics of light, while its nature as photons, as particles – a discovery of quantum physics - is not considered. Rosenthal et al. argue that a recentering on the particle-character of light in technological research might be beneficial for the development of imaging systems with a heightened degree of realism for the spectator (such as virtual reality systems), as the incorporation of all details of the functionality of the human eye, theoretically, allow for the production of images that are indistinguishable from reality.56 One reason for why this has not happened on a wider scale yet is clearly also because the traditional conception of light as a wave has a technological and philosophical head start of about 300 years. This shows that while many imaging systems are designed with the human visionary system in mind; the machines always only imitate one conceptual model of our vision. Current research in the natural sciences might potentially bring to the fore radically different concepts (as was done in quantum physics).

54

Rosenthal, Eric, Richard Jay Solomon, and Clark Johnson. "Waves vs. Photons: A new view of how we detect light and color." (2004): Creative Technology, LLC, 18 June 2004. p. 2ff

55_{Ibid, p. 4ff. Visible light reaches the eye in photon bundles that reveal properties of both time-dependent} waves and elemental particles. The authors of the quoted article argue that one aspect of this duality has been widely ignored in the conception of optical instruments, because it is generally assumed that the human brain infers the information from the incoming visible light in decoding its wave character.

(31)

31

Hence, the belief that one image system is more ‘real’, defined as being conceptually closer to the human body, and thus more authentic (in whatever definition), is irresponsible. For a discussion about film restoration, this means that it has to start with fundamental questions about our sensory apparatus. While for the authors of the quoted paper argue that, the presented misconceptions are the reason for the lack of realism of current digital images,57 I do, in this context, take a step back in wording and speak about a certain visual independence of all image-producing technology. This paper does not intend to prefer one model of vision over the other, but it advocates a detachment of notions of realism and authenticity in images from a single concept of the human vision.58 The path it explores intends no fixed conception.

Returning now to the profound implications in practice of problems of color restoration, and making the bridge to the proposed second major problem, one illustrative example of the abstract relationship of subjective vision and scientific principles within a film technological and archival discussion is the Callier effect. It describes the change in contrast of the same image when viewed under a direct light source, and under diffused light. Most digital capturing devices make use of diffuse light sources, meaning they direct light rays that are distributed equally over certain scanning area (for example the image area of a 35mm film frame) onto the sample material, as compared to directed light rays (also called specular light), which for example film projectors make use of.59 The alternating contrast levels in color film substantially alters the colors recorded, because the difference in the reflective angle of the light rays recorded by the sensor alters their wavelength, hence their color appearance. These deviations have to be either automatically corrected by machine algorithms or manually fixed by the operator. Other problems related to contrast are very

57_{Ibid. p. 2.}

58_{In similar discussions about the realism of photographic systems, debates about whether analog film is} conceptually closer to the human vision because of the shared use of subtractive color production were very common. Also here, this paper will not participate in such conversations.

59_{Flueckiger, Barbara. "Color Analysis for the Digital Restoration of Das Cabinet des Dr. Caligari." The Moving} Image 15.1 (Spring 2015). p. 30.

(32)

32

similar to those occurring in the analog realm. Like a new internegative struck from a reversal print will not have a low contrast like a camera negative, the digital scan too can only reduce the contrast of the input image to a certain degree.

3.3 Further Issues

Besides commonplace problems with color change and heightened contrast, there are a few further issues that the restorer encounters in the process of digitizing reversal originals. Many scanning processes are characterized by the creation of excessive elements, elements not found in the original sample ('noise'). What has been mentioned in reports of past projects is, among others, the creation of digital artifacts in face of an original's coarse grain structure.60 In the case where intermediate stages have been created, in the past, the scanning of those, too, did not achieve comparable results, and – where possible – the reversal original has been used for further efforts.61 Archives, beyond that, struggle with varying levels of material stability of reversal films.62 Other issues that arise on a constant basis are a particular color change in wet gate scanning63, the instability of the film frames due to damaged perforations, or imprecise camera shutter and gate mechanisms, jumping frame lines64, the difficulty of cropping a round-edged film frame into a rectangular digital image, the changing exposure and the related decisions of setting the scanner's black and white levels, as well as the overall

60_{Lepore, Matteo. "Lost and Found. Restoring James Benning's AMERICAN DREAMS (LOST AND}

FOUND)." Work/s in Progress: Digital Film Restoration Within Archives. Vienna: Synema, 2013. p. 163 61_{Read, An introduction to colouring techniques in silent era movies, p. 27.}

62_{While Kodak's Kodachrome, the most widespread reversal film stock proved to be a highly stable film carrier} that, although not without minor signs of degradation, stays relatively unaltered for many years if stored under appropriate conditions, other stocks are not. Color fading is one of the major issues not only for print film but for reversal film alike, and considering the absence of a camera negative that would allow for comparison; this loss could remain undetected or is only retraceable due to experience on the side of the film archivist. Black and white films are, on overall terms, much more stable than color films. It is believed, that a vast majority of noncommercial and non-narrative filmmaking (among them home movies) has vanished. See: "Editorial: FILM PRESERVATION: A Critical Symposium." Cinéaste 36.4 (Fall 2011). p. 1.

63_{"Interview with Gerard de Haan (Haghefilm)." For instance can be seen a fade to green in some Kodachrome} originals.

64_{Interestingly, the correction of the problems arousing from imprecise transportation of the film through the} gate of the scanner is often corrected live, meaning during the recording, which evokes an original projection, where the projectionist constantly had to tweak the frame line and focus.

(33)

33

instability of the machines and their frequent need for service and repair. In all this, we have not even touched the complex issue of magnetic sound recording, the type of sound most prominent on reversal carriers. While none of these issues do arise from color translation, they do account for decisions for or against the digitization or restoration of particular film examples in question. Such are also to be found in the following description.

Besides problems of technical nature, it is crucial to consider the array of issues with reversal films that exceed its mechanical foundation. As a result of the huge following among

independent film makers and moving image amateurs, most film archives today hold an immense amount of reversal films, making cataloging a very labor intensive task. In many cases the film makers have died or are otherwise not available for consultation, leaving behind orphaned films that lack contextualization. This makes the tracing and subsequent production of copyright and ownership metadata in many cases very difficult. What is more is that films shot on reversal in many cases were personal memories, anthropological records, rather than 'film'. Their mass of uniform expression is striking; the family on Christmas Eve, the children's first bicycle ride, a visit to the countryside, and sex, are tropes that reoccur constantly.65 Hence, a selective preservation (if one deems sheer digitization as a preservation effort) can be criticized as unauthentic, as it misses the representation of a mass.

Correspondingly, the original viewing dispositif was hardly outside the space of family and friends; a fact that clashes with the public and educational agencies archives usually aim to fulfill.66 A further, more general complexity for film preservation is the uncertain future of analog film stock production. Today there is no existing film stock on the market that is

65_{This aspect was explored in the Huis van Alijn exhibition last year. Homeless Movies. 26 June 2016.} Exhibition. Huis van Alijn, Gent.

66_{Exhibitions of amateur films and home movies are still scarce and rarely do these works execute a role other} than supporting other materials. This is the case in art museums and galleries just as in film museums. See: Meyer, Mark-Paul. "Conversation in Vijfhuizen." Personal interview. 30 Oct. 2015, Wasson, Museum Movies. Two recent examples of attempts to contextualize home movies in a public space were conducted at the

Museum Limburg in Venlo, The Netherlands, and at Huis van Allein, in Gent, Belgium. See: A century of home

Restoring Fantasies / (Re-)Creating Realities: Towards a New Dispositif of the Digital Restoration of Reversal Films