If a film scholar has to analyze a film, s/he commonly uses a home video edition of such film. When the sources are official one should not worry about the faithfulness of the given copy towards the original version of the film. However, publishers not necessarily bother offering the user a reliable copy of the film, even when they state that there has been a restoration and a remastering aimed at preserving the purity of the film. For many cinephiles this is no concern, although there is a niche of aficionados that take a lot of trouble over it, even more than scholars, whose neglect of the matter is more serious (cf. Enticknap 2004). While we take care of precisely indicating the bibliographical sources in our academic papers we do not do the same for films. This was comprehensible when films could be seen only or mainly in theater projections, in some archive with the moviola, or through a television broadcast. However, we should indicate the specific home video edition we are using, so that the reader can verify.

The matter I am dealing with has to do with film philology, which consist of the study of the relationship between a given copy of a film and its original version. Often home video editions present alterations in the quality of images and sounds which depend on an imprecise “restoration” work or on deliberate intervention by the filmmakers or the publishers. When we have to consider the stylistic patterning of the film, as regards cinematography (specifically the use of color), production design, costumes, sound editing and mixing, the philological problems of the copy become crucial. This article does not intend to offer a theory of philology or of restoration, but rather some methodological reflections for a more aware search and analysis of the copy of a film.

The concept of original version of a film is complicated (Venturini 2006: 22; Fossati 2011: 117–23; Catanese 2013: 54–8), but I believe that it is usually acceptable to speak of an original version — and hence of faithfulness towards it — in reference to the version originally distributed in accordance with the choices of the filmmakers. It is this that we normally expect to find in a home video copy, and that constitutes the object of analysis. Even film restoration scholars, who justly express the ontological problem of the original version, make a common-sense use of it. Catanese states that

we must reckon problems of film philology and aesthetics: if “improving” the visual or aural output of a film is an erroneous procedure, the restoration must take the oeuvre back to its original state and not adding a new modern aesthetic (2013: 101, my translation and emphasis).

Likewise, Fossati writes that

Film restoration is based on the best possible simulation of the original film artifact (where original is something in between the material artifact, as it has survived, and the idea of what it originally looked like), carried out using different technologies (2011: 142, my emphasis);

among these technologies there are the digital ones, which are able “to simulate analog reproduction media” (2011: 140).

The ideal situation is that of watching a copy of a film in its original format, but this is not always possible and would not automatically be an assurance. Nevertheless, a video version consisting of the digitization of a film can be a valid surrogate of it, provided that the qualities of the original film are respected: the aspect ratio, the grain, the color grading, the sound. Catanese affirms that both in the case of film prints and of DCP — to which I would add home video — “the restoration it is not limited to the intervention of repairing the source-copy, but involves the migration of the filmic text to another medium, which possibly preserves analogous or homologous features compared to the original one” (2013: 116).

Those who accomplish home video editions not always care about the philological respect towards the film. A common practice is that of the digital noise reduction used to diminish the grain, supposing that the consumers want to see “clean” images — i.e., the grain is treated as noise. Another problem is that of the color grading, which is sometimes carried out without bothering to respect the film’s original color scheme but rather to adjust it to current trends. Soundtracks are often remixed in order to adapt them to today’s standards and conventions (e.g., from mono or Dolby SVA to 5.1). By doing this the re-recording mixing style is altered, and sometimes there is even the introduction of new sound effects to make the sound more palatable. The digital restoration for home video, as well as the analog restoration of film elements, should bring the film back to its original appearance, so to preserve its features and make it available as it was originally made. Every other intervention, albeit legitimate, should be presented as an alteration of the original film.

As Enticknap (2004) notes, the case of Vertigo (1959) is emblematic of such matter. Restorers Harris and Katz distinguished between preservation and restoration, explaining that they took care of preserving the original materials before restoring. However the concept of restoration is used here ambiguously: Harris and Katz made a punctilious research to present the film visually as Hitchcock intended, nevertheless they recreated the soundtrack in 6-channel DTS, generating one of the most loathsome instances of film sound restoration ever, to such an extent that movie buffs caused Universal to exclude it from the Blu-ray Disc editions of the film, which rather contain the original mono (although unfortunately only for the US market) and a new, less radical 5.1 mix. If we search the Internet for information about the color grading and the audio of this edition we can find discussions that could be inspirational for us film scholars (Home Theater Forum 2012). Therefore, one who wants to analyze the film using this version should be aware of the philological status quaestionis — at least selecting the mono audio in the menu of the BD. The same can be said for Psycho (1960), whose BD editions include both the original mono and a 5.1 remix. Though the engineers who made the remix, in a featurette contained therein, explain with the usual ambiguous rhetoric that they faithfully improved the soundtrack, we must prefer the original mono in order to experience the film as closely as possible to the original version.

Even when the filmmakers, above all the director, take part in the restoration process, the problems are not lacking. It is legitimate that an artist modifies his/her work after some time, and that s/he offers a new version of it. However, the latter should coexist with the previous one, rather than replacing it, because for historical reasons and for research purposes it is necessary to access the film as it was originally made and distributed, which is usually how people saw it during the years.

The case of Thief (1981) is noteworthy. In the United States we find the prestigious Criterion BD edition (first published in 2014), which has excellent reviews (Atanasov 2013; Tooze 2015). The cover indicates that it contains director Michael Mann’s cut, in a director-approved transfer. The booklet states:

Director Michael Mann’s original 35mm answer print was used as a color reference, and Mann supervised and approved the entire transfer. … The original stereo soundtrack was remastered to 5.1 surround at 24-bit from 35mm 4-track magnetic audio stems, and approved by Mann [cited in Atanasov 2013].

At least two questions arise: (1) did the director approve this version according to a philological principle or to his current artistic will?; (2) what is the purpose for which we watch the film? To answer the first question we should focus on two aspects. One concerns the fact that the indication “director’s cut” suggests that this version is different from the one originally distributed and that has been seen by audiences since 1981. In fact there are some differences in the editing (MajoraZZ 2015); moreover, Mann had already changed the film for the 1995 Laserdisc edition (Criterion). Thus this is the third version of the film (Nield 2014); Mann has legitimately favored his present intentions rather than philology. The second aspect consists in the verification that the quality of images and sounds is consistent with that of the original version, which should be guaranteed by the words “original answer print” and “original stereo soundtrack”. Watching the film we immediately notice a teal tint, while all of the previous masters presented a considerably warmer color scheme. Since Mann is not new to changes to his films, the suspect is that the statement apropos the color grading is misleading. Besides, a 5.1 soundtrack cannot sound as the original Dolby SVA one, whose four channels were encoded through a matrix. Even if this 5.1 version consists of a transfer of the original DME tracks, we are deprived of the chance to hear the film in the closest possible way to the original, i.e., applying a Dolby Pro Logic decoder to a 2-channel matrix audio. In light of all this, maybe we should prefer the acclaimed Arrow BD edition (Atanasov 2015; Tooze 2015), distributed in 2015 in the UK in only 3000 units, which includes the Criterion master (this time also with the 2-channel audio) and a MGM master of the “theatrical version” (Nield 2014). Arrow, a publisher that embraces a cult philosophy more than the highbrow cinephile one that is proper to Criterion, knew that its customers would have loved to own the film in its original version along with the new one (this edition is now out of print). Regarding the color grading, based on the memory of those who saw the film at the time, on the careful observation of the Criterion version, on the knowledge of the history of film style and techniques, and on the comparison with previous masters, aficionados tend to conclude that this version is more faithful to the original one albeit presenting typical digital alterations (DVD Talk 2014).

As for the second matter, what is the purpose for which we watch the film? If we need to study the dramaturgy of noir films of the 80s, or Caan’s acting style, or editing, then philological problems concerning colors and stereophony are negligible. However, if our study takes into account the production design, the cinematography, or the re-recording mixing style of the period, then the problem is substantial. Lastly, if we want to study Mann’s style, the new version must be taken into account, provided that we consider the differences with the original one — and with the 1995 one, for that matter.

In sum, it seems that some publishers, technicians and filmmakers prefer to better presenting a film rather than to best presenting it in a philological sense, with the highest fidelity towards its original features. Those who use home video editions for scholarly purposes should always undertake a more or less thorough philological research, at least to clarify possible problems. I believe that we should first of all look for information about the content and the technical specifications of the film in its original version, starting from the Internet Movie Database (https://www.imdb.com/). Then we should check how much the edition we use is affected by the usual interventions, such as digital noise reduction and alterations in the color grading, in the sound editing and mixing. Internet sources are essential: we can refer to review sites such as DVDBeaver or Blu-ray.com, and DVD Talk (the last two containing a user forum as well), and other web sites and forums of aficionados who, based on my experience, are often much aware and scrupulous, such as OriginalTrilogy,¹ Fanres (Fan Restoration Forum),² Home Theater Forum,³ AVS Forum,⁴ Bad Audio on Blu-ray.⁵ In any case, one should indicate in the text the details of the edition that is used, so that the reader can verify. The information required is the kind of medium (e.g., DVD or BD), the publisher’s name, the year and place of publication or the name of the streaming service or the URL; some code indicating the edition may be specified as well (e.g., the catalogue number or the ISBN). This is particularly needed for video essays that contain literal quotations of picture and sound of preexisting works (cf. Grant 2010), which are the audiovisual correlative of the quotations of written texts that are carefully referenced in academic papers.

Videographic analysis for home video philological studies has an at least twofold advantage over the usual comparisons found on review sites, such as the well-known ones by DVDBeaver (which are nevertheless useful): first, moving images can be included — instead of stills only — which offers a much more effective support; second, sound can be reproduced as well. Depending on what is at issue, film excerpts from the different versions can be juxtaposed in sequence or through split-screen, or can be reproduced alternating with one another. The following video essay, containing a philological analysis of three home video editions of The Shining (1980), is intended as a case study to demonstrate my argument.