Film Studies in Motion: From audiovisual essay to academic research video

Still just a still

Visual embellishment of writing-centered Film Studies started taking off in the 1970s. In order to properly trace logistical hurdles on the one hand, and theoretical, practical, even legal implications on the other, it is essential to reflect on the work of scholars on both sides of the Atlantic. While French film theorist Raymond Bellour was at the forefront of philosophical reflection, American scholars David Bordwell and Kristin Thompson were arguably spearheading practical solutions. To obtain and sustain topical clarity while scrutinizing their works, we will favor thematically clustered discussion over a diachronic overview. First, guided by the work of Bordwell and Thompson, we will trace the practice of including stills in a text from a more or less logistical viewpoint. Then we will take a chronological step back in order to reflect upon the problems that the incorporating of stills initially seemed to solve within Film Studies, yet, in turn, only complicated. This order is imposed due to the fact that Bellour’s investigations link-up with the technological experiments that would follow.

Whereas nowadays anyone can include screen grabs into their writings, this was a rarity up until the 1970s. Back then Film Studies were (still) finding shape, and the possibilities for visual illustration of writings about film were crippled by practical shortcomings and/or financial tolls due to cumbersome frame-grabbing technologies and undeveloped copyright legislation.[16] Kristin Thompson relays that in the 1970s “relatively few authors were using frame enlargements since it took special equipment to photograph such frames: expensive camera attachments, color-balanced light sources, and the expertise to use both” (Thompson 2006). In the event that stills were obtained, the question of licensing remained. This commonly involved bureaucratized communication with production companies and paying fees to copyright holders of the original film print. Therefore, publicity and on-set photos were used more often, which, although they were cheaper and easier to come by, could serve decorative purposes only, as they did not depict scenes as seen in the film and were therefore not specific enough to illustrate actual points of the analysis.
            ​In this light, Amos Vogel’s 1974 book Film as a Subversive Art is a notable standout (Vogel 2005). It boasts mini-essays on over 600 films, illustrated with some 300 stills. The book discusses films that were for the most part shown at the ‘Cinema 16’ screenings, which Vogel organized between 1947 and 1963. These gatherings provided a platform for avant-garde and experimental film, many of which were shown there for either the first or the only time. The book itself was never meant to be an exhaustive account, as it was intended to function more like an overview to direct audiences to these cinematic rarities. Being aware of the limited possibilities of his time, Vogel was grossly unsatisfied with the relative inefficiency of illustration through frame grabs. He aptly noted that “[a] still is not a film. It lacks the dimensions of time and movement – indispensable components of film art – and represents only a small fraction of a single second of a given motion picture,” and that “[w]ords are not the best way to deal with a visual medium” (Vogel 2005, 3), a problem which, as we will show later on, Raymond Bellour would investigate further around the same time. Still, grabbed stills in this collection allowed a single or even, for some, the only glance at films that were hardly or not ever accessible to the general public: the larger part of these frames was taken from films that were banned, censored or even destroyed.[17] Vogel’s book, therefore (with all respect to the writing) owes much of its value to these stills, which he was able to obtain (physically as well as legally) via personal contact with the filmmakers and right-holders.[18]
            ​When in 1977 David Bordwell and Kristin Thompson were preparing for their first edition of Film Art: an Introduction (published in 1979), they too looked for ways to visually support their writings on films. What they had in mind was a use of stills that would tandem with their written close analyses. To accomplish this to a satisfying degree, they not only needed isolated stills as illustrations, but key frames to entire sequences in order to present formal breakdowns of complete scene structures. However, unlike Vogel’s examples, a good deal of their needed material was in the hands of big studios. The studio-generated publicity photos, customarily distributed at the time, were unrepresentative of the actual film, and therefore served little to no tutorial, let alone informative purpose; “they were useless for the sort of close analysis [we] wanted to do[.] Film Art thus became the first (introductory) film text to use frame enlargements extensively” (Thompson 2006).[19] ​Bordwell and Thompson continued to push their practice further with subsequent ‘interactive editions’ (with accompanying tutorial CD-ROM from the book’s 8th edition, from 2006 onwards) and an online video-collaboration with Criterion from 2012 on (Bordwell and Thompson 2012). The authors, as well as their publisher, assumed that the clear educational purpose of the book would automatically suffice as a claim to ‘fair use’, despite the great quantity of frame grabs presented. Similar to Vogel, Bordwell and Thompson approached experimental filmmakers directly, though “more out of courtesy than because we figured we legally needed to” (Thompson 2008), and skipped the studios all together (Thompson 2006 and 2008). According to Thompson, the big studios were oblivious to the goings on of a theoretical counterpart to cinema anyway: “Those of us publishing extensively illustrated books on Ozu or Eisenstein weren’t exactly a concern to the film industry. In fact, studio executives probably didn’t know that we existed and wouldn’t care if they had” (Thompson 2006). The book was published without any conciliation with the majority of right-holders. This juridically daring act did, however, not immediately set a trend; other authors and publishers were more hesitant and stuck to production stills, or paid greatly for official permission to use stills provided by the studios. Later, Thompson was invited to investigate the legality of her approach, which resulted in the Report of the Ad Hoc Committee of the Society for Cinema Studies, ‘Fair Usage Publication of Film Stills’ – a declaration, published in 1993, that provided positive affirmation on the then unregulated practice.
            ​Conversely, even though the advent of DVDs eased the technical means for grabbing stills (by taking screen shots), it also re-problematized the position of scholarly utilization due to laws applied to the technological use of digital video in general. Ultimately, the 1998 Digital Millennium Copyright Act (DMCA) rendered all copying of audiovisual material illegal. Nevertheless, thanks to the newfound practical ease combined with the momentum through which fair use had proven itself viable, many scholars proceeded to rip clips and obtain film stills via alternative channels. Despite the medial change of the source material (from celluloid print to mass produced digital video disc), one of the key requirements of the fair use claim still applied: “Lectures, articles, and books do not damage the studios’ ability to exploit their own films commercially” (Thompson 2006). Still, the boom of footage found online was held back just yet. For some years, official exemption remained pending, until recommendations were made, and authorized in 2006.[20] ​The issue was still not fully resolved, however, as the recommendations were founded upon the proposition that quotes could be taken only from audiovisual materials owned by educational institutions, whereas in practice teachers (and students) would and will often use their personal copy of films for such purposes – if not because of convenience, then due to lack of school budgets and other accessibility issues. After all, the traceability of the original sources used for these still material is next to none, and practices prior to, and since then, have shown that film scholars are in the clear: Thompson (2006) notes that there are no known cases of either title specific investigation nor author/publisher prosecution (Thompson 2008 exemplifies this further). This discussion was revived when uploading video to the Internet became increasingly easy and popular.

Having regarded the inclusion of stills from both legal as well as technological points of view, it is now time to turn to transtextual implications. By this we specifically aim to engage with the dissonance between formulating and rendering film theory, and the process of observing the film itself. This film itself, meaning both the narrative and non-narrative spanning as well as the physical carrier (mainly 35mm celluloid) were first conceptualized as the film ‘object’. Another way of referring to the film is ‘source material’, which is the most logical way to address its role in the video essaying practice (it also considers the fact that film is becoming almost completely ‘object-less’ in times of digital cinema – hard drives notwithstanding). This semantic issue is something we will get back to later on.
            ​In 1973 French film critic and theorist Raymond Bellour published an article where he singled out a particular scene from Howard Hawks’ 1946 The Big Sleep. In the article he embedded twelve key frames – depicting two characters in a car’s interior – corresponding to the twelve shots that make up that scene. The motivation for Bellour’s analysis, and for his approach specifically, was to point out the fact that although Hawks was a director known for setting up his films economically and with clarity, he used twelve shots for a seemingly simple and short dialogue between two people seated in a car. What makes the case interesting, however, are the means Bellour used to reach beyond the constraint of solely written text to illustrate his point. In the article we find twelve frames, presented in the exact succession as they appear in Hawks’ film. In an additional effort to simultaneously ‘decode’ the ‘source material’ (or ‘obtain the object’ – see discussion of Bellour 1975), a diagram was drawn up that charted closeness of framing, status of the frame (moving/still), angle, characters shown in the shot, indication about who spoke, shot duration, and elements of narration taking place within that shot [Figure 3].
From that point on in the analysis, Bellour calls for a parallel reading of his text and the stills, preferably while keeping track of the diagram as well. All in all, in his genuine attempt to let the chosen sequence be ‘visible’, that is to bring the sequence and the reader of the analysis closer together, Bellour in fact added extra hurdles in between the two. The practical impossibility of the proposed reading method ultimately exemplifies the struggle with technological separation of the study and studied, something which Bellour would reflect upon two years later.
            ​In 1975 he wrote ‘The Unattainable Text’, what can now be recognized as a pivotal article. Bellour contemplates “the comparative backwardness of film studies” (Bellour 1975, 19) – the discord of static written text with isolated and incidental frames to represent a multimodal phenomenon unfolding in time such as film. Media scholar Erlend Lavik aptly summarizes Bellour’s discontent concluding that written film analytic texts’ “inability to quote their object has been a long-standing drawback for film critics” (Lavik 2012). It should be noted that Bellour speaks of film in two ways that are of relevance to this study, and call for clarification. In line with our aforementioned categories, firstly, he talks about the film ‘object’, which denotes the (physical) artefact of the film that can be stopped and played back (‘delayed’, as Laura Mulvey [2006] would put it), enlarged, be experienced with all the qualities it inhabits (sights and sound mimicking motion spanning time). In the case where one ‘possesses’ the ‘object’, the film can be accessed at all times (rather than having to restrict one’s analysis on an isolated theatre showing). Secondly, Bellour opens his article with the announcement of discussing film as a ‘text’, “in the sense in which Barthes uses the word,” only to later abstain from the “theoretical labyrinths opened up by the notion of the text” (Bellour 1975, 19). In practice this means that he wishes to problematize the medial properties of film by use of the word ‘text’ in a more phenomenological definition that signifies a compound of (technological) assets, rather than the everyday connotation of words on paper. As Bellour states, “it allows one a full experience of the multiplicity of operations carried out in the work and makes it precisely into a text”, so that “as soon as one studies a work, quotes a fragment of it, one has implicitly taken up a textual perspective [in a] restive and regressive fashion” (20, our emphasis). This operationalization of ‘film’ being a compound of technologies brings us to the heart of the matter why film scholars sought inclusion of stills in the first place, but also why Bellour felt the urge to look beyond that: he underlines his thoughts with a comparison to literary studies, which enjoys “the undivided conformity of the object of study and the means of study”, making it “immediate” since “the quotation is invisible, it is quite naturally absorbed into the page” (ibid). Film analysis, in this regard, could not be performed within the same medium as the source ‘object’, and its quotation thus reduces its medium-specific “physical attributes” (23). That is to say, written descriptions of an audiovisual film sequence cannot, by their very nature, incorporate the celluloid itself, nor feature running film within the written account. Contrarily, the video essay, as we know now, comes very close to being absorbed ‘onto the page’ – in the age of digital intermediates and digital filming (see the conclusion of this chapter) we can take an actual ‘piece’ of the film and super-impose our comments on it, sometimes even using the exact same editing software with which the filmmakers used to make the film.[21] ​Ironically, now that the study of cinema has finally arrived at a point where it is able to communicate with moving images, film has lost most of its original (chemical) physicality: films have become digital streams and are rarely screened from celluloid carriers anymore, if they were not already shot digitally in the first place. Nevertheless, the above holds: practice where a still image is used to represent a moving image that normally unfolds in space and time “reveal[s] a kind of radical inability to assume the textuality of the film” (25). Description of sound eliminates “tone, intensities, timbres, pitches, everything that constitutes the profound solidity of the voice” (23). Within this – technically constrained – historical context, Bellour rightly concludes that “[t]he text of the film is unattainable because it is an unquotable text” (20).
            ​Having regarded the issues that arise when attempting to capture the multimodal instance of film in monomodal writing, Bellour then muses on the possibility of Film Studies improving in tandem with technological advancement. He is convinced about such innovation’s fruition, while at the same time stressing the shortcomings of theoretical exertion if certain possibilities were to become available. In more concrete terms, Bellour is talking about the study of film in an age (remember, we are in 1975) where “at the price of a few changes”, film could achieve “status analogous to that of the book” (19). As said, this would mean that the cinematic ‘object’ and its investigation would be rendered as complementary lexicons of the same ‘text’: within a single medium. Like literary analysis is analogous to the literary text it investigates when regarded from a medial perspective, there would someday be the possibility of studying film through the same audiovisual medium as the film text itself. Bellour presciently notes that

If film studies are still done then, they will undoubtedly be more numerous, more imaginative, more accurate, and above all more enjoyable than the ones we carry out in fear and trembling, threatened continually with dispossession of the object. And yet, curious as it might seem, the situation of the film analyst, even when he does possess the film, any film, will not change in every particular. (ibid, our emphasis)

The quote indicates that Bellour expected problems with what was then only a theoretical concept (one that closely resembles our current video essaying practice), when he stated that “this sudden quotability which film allows to film (...) obviously has its other side” and questioned whether “oral language [will] ever be able to say what written language says? (...) And if not, at the price of what changes?” (26-27).
            ​In effect, scholarly analysis through stills remained the common practice for considerable time to come, albeit dominantly in the Bordwell and Thompsonian way. While a book like Film Art arguably circumvents the aspect of not attaining all-inclusiveness by focusing on one aspect at a time (c.q. mise-en-scene or framing), Bordwell’s 1985 Narration in the Fiction Film sports page-filling scene breakdowns accompanied by large amounts of stills. Similar attempts can be found in, for instance, 1980s Dutch/Flemish film journal Versus, where still-breakdowns take up entire spreads, and extensive(ly annotated) diagrams and graphs illustrate not only plot points and character development, but also the lengths to which one would go to map the dynamic unfolding multimodalities of film in static, monomodal writing [Figure 4].

This page has paths:

This page references: