Remixing Rose Hobart

Creator's Statement

This essay explores how videographic approaches might help to unearth latent meanings within Joseph Cornell’s collage film Rose Hobart. The video is divided into four parts: an introduction and three movements.  

The introductory section takes Cornell’s own working method as a starting point, explicitly asking in text and voiceover how a remixing of Cornell’s film—just as Cornell remixed East of Borneo (George Melford, 1931) and as-yet-undetermined found footage—might reveal something hidden or latent within it. The visual track in the introduction presents a kind of symbolic representation of Rose Hobart as a nexus of separate texts, as footage from Cornell’s film and footage from East of Borneo intersect both spatially and temporally.  

The first movement offers a meditation, through voiceover and multiscreen work, on the various translations of Rose Hobart from its 16mm original to digital form. The most widely-available versions of the film introduce a number of artifacts, variations, and repetitions that videographic analysis must be aware of and contend with. This movement attempts to make those artifacts visible through freeze frames and multiscreen. However, this section also troubles the idea of an “authentic” version of the film, and points out how even imperfect digital transfers can draw our attention to Cornell’s careful montage, the materiality of the source material, and the importance of the second-hand to the more latent aspects of Rose Hobart’s aesthetic. In conveying my own critical reception of these different DVD transfers, I tried to gesture toward some of the limitations of videographic approaches—reliant as they are on commercial digital transfers—for texts that operate at the level of the frame, while at the same time thinking about how digital limitations might be reworked as critical strengths.  

The second movement juxtaposes a selection of shots from Rose Hobart with their correspondent shot in East of Borneo, as they appear in the latter, to videographically illustrate the logic of Cornell’s extraction. Cornell carefully cuts around minute gestures, original splices, and even dialogue in order to include or exclude very specific moments from Borneo. This extractive logic is based almost entirely around characters and their behavior—not only around fetishism of Hobart, as received readings tend to emphasize. The lack of voiceover in this section was an attempt to convey this idea without forcing it; the characters’ voices seem to guide Cornell’s cutting. It was also meant to break up the heavy use of voiceover in the first and third sections in a more lyrical interlude.  

The final gesture emphasizes Cornell’s extremely precise editing in Rose Hobart, particularly his extraction or leaving in of single frames. The most striking example, a single frame of an Austronesian man, opens up the possibility of entirely new readings of Rose Hobart in relation to colonialism, orientalism, and the work of Claude Levi-Strauss. The famous shot of a pebble dropping—probably the most enigmatic shot in Rose Hobart, and the only one he repeats—takes on new meanings in relation to this frame and the almost-retitling of the film as Tristes Tropiques. The last set of multiscreen shots in the piece evokes the lachrymal imagery that this frame allows us to read as latent in Rose Hobart.   Fundamentally, my approach in this piece, heavily reliant on both voiceover and multiscreen, was an attempt to guide the viewer through my own videographically-enabled reading while at the same time resisting imposing it on top of the inherent and beautiful mystery of Rose Hobart’s surrealism. This was fundamental to my decision to leave voiceover out of the second gesture. Finally, through the somewhat ominous soundtrack that underlies the introduction, second, and third movements, I sought to evoke the latent, buried imagery in much of Cornell’s work.

-Derek Long

In this piece, Derek Long explores the particular value of the video essay as a means of analysing found footage or appropriated film. In the first part it addresses the problem of provenance of Joseph Cornell’s 1936 re-configuration of East of Borneo (Melford, 1931), which exists in several significantly distinct DVD versions. When we descend to the level of frame by frame analysis, Long demonstrates, we must consider which version we are analysing, and what variables have been modified in the act of preservation. In a film that involves many arguably minimal interferences on the part of the artist, the need for fine-toothed analysis must be tempered by an awareness of alterations that creep in not through the authorial intent but through time, contingency and the vicissitudes of the archive and exhibition in the era of digital display and distribution. Indeed, it is an accidental intervention of the preservationist that brings Long back to his least favourite copy of the film, the Magic Films of Joseph Cornell version released in 2004, which, because of its poor frame-sync during the digital scan, now produces strange ghosts, that suggestively (though misleadingly) appear as flickering spectres of the frames that have been excised by Cornell.

In part II of the video, Long begins the real work of comparing the re-edited shot sequences from Rose Hobart with the unaltered original sequences from East of Borneo, showing us precisely where Cornell has made his cuts and splices. Much like Cornell's apparently simple interventions into the body of East of Borneo, Long performs a conceptually simple, but structurally and formally complex manoeuvre, offering a direct visual comparison of corresponding sections. We see what was excised, what was kept, and what was left behind, down to the frame, and we begin to see how a logic of association is developed. A rationale begins to manifest through a witnessing of what-could-have-been. Why this shot and not the one that follows it? Why cut the shot at that precise point? Why retain a brief flicker of another figure/movement, glaringly obvious at the end of a shot, when we know how freely and easily all trace of it might be removed entirely? Through Long’s method we see more clearly and precisely how a new language is born of the old, how the text is rewritten. We also begin to see the chronological and structural reconfigurations that Cornell has effected, while also gaining a greater sense of the jarring re-contexualisations that simultaneously strip meaning and create openness, that nevertheless produce hermeneutic vectors, directions for travel, indicators of thought, affect, atmosphere, animus. This is the video essay as philology, as a tracing of the etymological roots of shots, of the textual and contextual history of recombined images. 

The video essay ends with a specific and fascinating discovery: a real ghost. At the junction of two shots Long identifies a single frame that reveals a hidden character, a mysterious other. The figure appears to be a remnant of the ethnographic footage that Cornell used to source his billowing palm trees. This momentary intrusion of a figure from the other source could be written off as accidental, especially as it flashes by too quickly for the audience to see during a normal screening. But Cornell's artisanal engagement with the material of the film itself would indicate that his knowledge of the film he produced pertained at the level of the frame, especially the frames at either end of his splices. The intrusion of this indigenous figure also complicates Cornell’s use of the outsourced palm trees, as the figure resonates so closely with East of Borneo’s exoticised indigenous bodies. This final point opens up a host of fascinating questions.

In my own work on this film I have emphasized the way that Cornell uses a number of strategies to induce illegibility in the text, to break through the encrusted significatory structures of the original film. However, what Long shows us here is something more directed, more closely resembling a considered critical re-writing of the narrative. It hinges around the artist’s obsessive return to certain images, but through Long’s forensic approach we can pick a path through the dense oneiric atmospherics of Rose Hobart, on order to see more clearly the structural and conceptual plan at work behind the construction of Cornell’s film. This video essay seems to me to demonstrate that the form can provide an especially potent mode of analysis for appropriated and found footage works. 

As I'm putting together my thoughts on how to properly review this video, I'm finding myself pulling short of writing a polished review and instead offering fragments (perhaps following the lead of both the video and its object of inquiry). On the one hand, I admire the different analytical approaches and lines of inquiry and the deliberate inconclusiveness of it (fitting for such an indeterminate work as Rose Hobart). On the other hand, I wonder if those same qualities are diverting from what could actually be a more fully realized version of any one of the video's three chapters, as currently they feel less of a satisfying whole than fragments that each could be developed further. Here I will point out what I feel are the strongest or most crucial moments of each part:

Part I: Pointing out the differences of the two digital prints commercially available helps us recognize the inherent lack of "authority" or "conclusiveness" in the analysis, given the instability of the source materials. This feels like the closest among the three chapters to being a fully developed analysis.

Part II: I admire the bold break from the first part to perform a voiceless, purely visual side by side analysis. At the same time, I can't help but find this part begging to be played through a second time with Derek offering an account of what he discovered through this experiment, as the conclusions a viewer might make aren't so obvious. What exactly is the "logic of extraction"? Can the question be answered of what made Cornell decide to extract what he did and leave the rest out? Or lacking an authoritative answer, what are the most productive modes of speculation that we might gain having witnessed the side by side comparison? (for me there seems to be a fixation on close-ups and gestures in Cornell's extractions).

Part III: This section might be the most thematically important but also to my mind the least developed section. The pointing out of the inserted frames of the indigenous figure not found in the original East of Borneo is absolutely brilliant and leads to many questions... but the follow through is more a tease of a possibility of a critique of colonialism that doesn't feel satisfying enough even as a tease. The passing link to Lévi-Strauss is also fascinating, and opens a door for a more sustained engagement with Tristes Tropiques (what else is in the book that could resonate with Cornell's film?), but as it is we get just the first sentence from the book, and a concluding question that feels more like the opening to a much longer inquiry ("For Joseph Cornell, what makes the tropics sad?"). That last line projects an even richer inquiry extending much further beyond the video's ending.

In a most positive light, I see the three chapters as charting a progression from an initial but important comparison of source materials, to a formal analysis of Cornell's extractions, to the beginnings of a deeper ideological inquiry. As it goes on, the video digs deeper and deeper and obtains new tools and insights along the way. Perhaps if the video were framed more explicitly as a progression (and one that's still in progress), so that the chapters could be read more linearly than laterally (i.e., here are three different ways into the film), I'd find the overall effect more satisfying. (One part that I'm not convinced by at all is the section from 5:28 to 5:50. Given all that's been said previously about the instability of the prints being worked with, I'm not convinced that the missing frames being spotlighted in this section can be conclusively marked as "what Cornell is leaving out," much less ascribe those omissions entirely to authorial intention [maybe the print he worked with had frames missing to begin with?]).

I would pose the question to Derek whether he might consider working further with parts II or III in light of the feedback I gave for each (I acknowledge that doing so would lengthen the video). The second, perhaps more useful method, would be to frame the video more explicitly as a succession of approaches that build on each other in terms of burrowing deeper into the film's meaning (that is, if Long agrees with this reading). But most of all, I'm concerned with the ending and where it leaves us as viewers -- this question of Cornell's relationship with colonialism certainly deserves a little more than the peeks we have so far...