Read an Excerpt

Silent Cinema and the Politics of Space

Indiana University Press

Location, "Location"

On the Plausibility of Place Substitution

Mark B. Sandberg

Hidden behind the casual use of studio terms such as "location scout" and "shot on location" are histories of film practice that reveal ongoing, productive tensions inherent in the idea of cinematic place. On the one hand, the filmic medium conveys a strong impression of specific location because of the photographic image's indexical properties, which inspire confidence in the verifiability of an original shooting location. On the other hand, the inherent mobility and portability of the camera and cinema editing's powers of formal juxtaposition combine to untether the image from any securely specific sense of originary place. One might sum up this contradiction by saying that although we can be confident that we are seeing some place filmed by the camera, the cinematic image itself can never make us completely sure of the actual place; the possibilities for substitution and misidentification are endless. The commoditization and interchangeability of shooting locations are a direct result of this central ambiguity. Add to that the convincing impressions of fictional location made possible by an array of cinematic sleights of hand (animation, miniature models, and increasingly, digital imaging), and one quickly sees the need for an approach to cinematic place that acknowledges the wide variability of the concept.

Thinking carefully about such issues might help to open up the habitual terminology of cinematic location to more genealogical perspectives. The questions that arise from the recognition of the historical contingency of current practices include: At what points in film history have audiences cared about the actual location of filming? When did it become a mentionable, promotional asset for studios to claim that a film was shot "on location," and what contrastive prior practices gave that claim its value? When does the identification of actual shooting locations become an expected piece of information in the end credits of films, and why? At what point does the investigation of location discrepancy (the disparity between actual and ostensible locations) become a pleasurable form of fan discourse in its own right? What is the history of these geographical-analytic modes of film spectating?

A recent example will help set the stage. In the 2007 film The Kite Runner, there are two diegetic locations established for the film, one of which happens to be very remote from my current writing location (Kabul, Afghanistan) and one that is quite close (the San Francisco Bay Area). As each of the two locations is introduced in the film, a conventionally superimposed title locates the image in the usual authoritative way. As it turns out, however, the scenes marked "Kabul, Afghanistan 1978" were not actually filmed there, but were shot on location in three different filming locations in China instead. Here is what a New York Times reporter wrote about one of the location choices while the film was still in production in December of 2006:

The production team spent three months researching locations, giving little thought to Afghanistan itself, for obvious reasons, as they drew up an initial list of 20 countries and deliberated on which one would get them closest to Afghanistan's look. The possibilities ranged from India to Morocco to South Africa, but E. Bennett Walsh, who oversaw the search, said the conversations kept returning to Kashgar, a place that few people in Hollywood had ever heard of and where no Western film had ever been made.

Kashgar is a small city in far western China, actually not far from the Afghanistan border, which partially explains its potential to duplicate "Afghanistan's look," as the reporter puts it. While the substitution of China for Afghanistan might initially sound quite incongruous, when one sees this relationship on a map the differences seem less stark. Both of the cities in question are actually in the foothills of the Pamir mountain range (Kabul to the west, Kashgar to the east). The distinction between the two places, in other words, might not be so much geographically essential as it is politically arbitrary, since topographically speaking, there might not be much of a "substitution" at work here at all. Since Kashgar is culturally speaking also a Muslim community like Kabul, the discrepancy between the real and ostensible locations has more to do with preconceptions of "China" than with anything else.

One of the film's producers, Rebecca Yeldham, revealed the complexity of the location simulation in commentary included on the DVD:

We filmed in China, but where we filmed in China isn't really China ... and the old town of Kashgar feels like a city, as we were told, a city in Afghanistan that is now long forgotten, so it was a perfect place for filming. In fact many of the Afghans who again traveled through the Xinjiang-Kashgar portion of our filming felt like they were home, and it was a very surreal experience in the movie through their eyes because the cities that we shot in are strangely locked in time.

Her observation helps to isolate a central issue about location shooting: what constitutes substitution? If two places belong to a continuous landscape or some other larger entity, does it make sense to use the word "substitution" at all? Or does using the term "substitution" only come into play because of the existence of a perceived border or boundary, a clash of some sort that must be ignored or covered over in order for the process to be successful? And further: does the image itself betray that difference? How much can one rely on something resistant in the image itself, some visual ontological tie to actual location, and to what degree is this instead a function of supplementary labeling and the viewer's verification abilities?

The Times reporter continues his description of The Kite Runner' s location shooting in this way:

Beyond the cooperation of authorities and the availability of highly skilled filmmakers in China, Kashgar was always the best fit in terms of appearance, beginning with a diverse but overwhelmingly Muslim population and a countryside that plausibly resembles Afghanistan. "In some locations you are limited to working small, little corners, whereas here you can shoot 100 yards down the road," Mr. Walsh said. "The streets of this city are just dripping with production value. All you have to do is change the signs."

The phrase "dripping with production value" reveals the economic calculation involved in place substitution—the bet that the visual similarities of the substitute will help cut corners on either set building or travel, or perhaps circumvent political obstacles to filming in the place the fiction is located. The reporter's final claim is also interesting, since it reinforces a definition of cinematic location as nonessential; his implication is that for the purposes of filming, the distinctiveness of place is assumed to reside in a thin, superficially visual layer of culture that can be taken on and off like clothing. In this case, it seems that film's long-standing exploitation of façade aesthetics has colored assumptions about cinematic location as well.

All of these considerations become more resonant, of course, when seen against the central metaphor of the film: the "kite runner" is a young boy with an inerrant sense of location, a living GPS system really—he is the one who can run down a kite when the string breaks and arrive first at the spot of its landing. At the same time, the film production itself was forced off location by the political realities in contemporary Afghanistan, presumably because it was too difficult to film there. Like the unmoored kite, the production's affiliations with place were quite untethered. As screenwriter David Benioff puts it at one point in the DVD commentary:

So you have a book written by an Afghan-American, adapted by a New Yorker, directed by a Swiss man, one of the producers is Australian, being shot in Beijing with the kids from Kabul, Baba from Iran, other actors from England and France, incredibly international, all gathered together to shoot this movie set in Afghanistan. It's just kind of bizarre, but ... but you know, and, it ... it worked.

Since "no Western film had ever been made" in Kashgar, and since the Afghanistan imagined in 1978 might be assumed to be a lost place anyway, there was probably a calculation that the substitution of Kashgar for Kabul would not be noticed by the likely audience of the film. And in fact, although I am told by friends who have traveled in Afghanistan that they did notice something slightly "off" in these location shots, for me Kashgar made an unproblematic Kabul. I suspect the same was true for a majority of those who saw the film. Benioff even boasts:

One of the really gratifying things that we've heard just recently was that the movie was shown to the Afghan diplomats including the ambassador, I think, to the United Nations and, you know, several officials, all of whom grew up in Kabul, all of whom lived there during the seventies, and they couldn't believe that [director] Marc [Forster] was Swiss, you know, they thought it was so accurate—first of all they couldn't believe it wasn't shot in Kabul because it looked exactly the way they remembered it.

From this comment we see that place substitution is a kind of calculated risk for the film producers about the degree of specific geographic knowledge resident in the audience and about how much that will matter to the reception of the film, a wager that paid off well in this case since even some viewers with insider knowledge found the substitution plausible.

The film's other location suggests the possibility of alternative reactions. When I see the opening scene identified as "San Francisco, California 2000," I recognize the actual location to be a particular grassy hill down at César Chávez Park near the Berkeley Marina where the annual kite festival is held. Those of us who know that place well enough and intimately enough—that is to say, locally enough—might even object to the caption here, since for us this is not San Francisco at all, but more precisely the Berkeley Marina with a view of San Francisco. To those who have never been to Berkeley or experienced the distinctions between the city and the East Bay, this will likely matter as little as the relationship between Kashgar and Kabul, since the image performs adequately as the visual invocation of "San Francisco," just as Kashgar did for Kabul. It is the deictic aspect of this location recovery, however—encapsulated in the shock of recognition that reveals the "there" of the film to be the "here" of my daily life—that is worth emphasizing. The free substitution of shooting locations for fictional locations by film producers will inevitably collide at times with local site-specific knowledge, but if it does so within an acceptable range of plausibility for most viewers, it will pass unnoticed.

Of course, our current viewing culture of DVD special features and PIP (Picture-in-Picture) bonus views creates a niche for those spectators who want their location-recovery information delivered in tandem with the fiction itself. In effect, the availability of these features can potentially turn all viewers into "kite runners" able to recover the specific shooting locations of films cut loose by the practice of place substitution and international distribution. Although DVD commentaries have proliferated only recently, the analytic function of the behind-the-scenes, on-location peek into the filming process belongs to a much-older analytic fan discourse that has long taken pleasure in uncovering the "truth" of the filmmaking process. Early on, we can find that impulse in photographs of film sets that reveal façade-like architectural structures, or the actual way that a stunt was created, or the unstaged physical environment just beyond the framed image. The existence of these "revealing" perspectives as a parallel pleasure to that of immersive viewing was an early fact of film spectatorship. The impulse continued with the proliferation of fan magazines, the eventual rise of "making-of" documentaries as a genre, and more recently, with the analytic apparatus that makes up the special features of today's DVDs. Tellingly, this kind of viewing pleasure is now also fostered by sites such as the Internet Movie Database (IMDb), which provide in their "trivia" section this kind of detailed information about real production locations as well, though derived in many cases from a symbiotic relationship with the DVD commentary, the source of much of the insider location information. Although various desires motivate fans to look behind the curtain of the film production process, one recurring attraction throughout film history has involved the sorting of real from fictional location: the pleasure in finding out that although it looks like it was shot in Afghanistan, it was really shot in China. We might call this "kite-runner spectatorship."

This geographic settling of accounts does not appeal to all viewers, of course, and it is certainly reasonable to assume that public interest in location discrepancies would wax and wane throughout cinema's history. Today, for example, it is easy to think of many factors in the current media landscape that might motivate public interest in knowing more about cinematic practices of place substitution, including the explosion of virtual geographic access online through various kinds of mapping software and the general proliferation of GPS technology (including the ability to geo-tag photos and videos). These and other media developments have arguably created a particular consciousness of the location and dislocation of the image, not only in contemporary films but in those of the past as well.

I readily acknowledge that the questions that interest me in what follows are driven in large part by my own current cultural context of early-twenty-first-century place sensitivity. Current questions make it possible to see new things about the past, however, and it is in this spirit that I turn now to a case study of the use of location in early Danish silent film. The gaps between then and now, between there and here, present themselves as an interesting challenge for reconstructing what mattered to the Danish producers and their European audiences of the early 1910s. Some of the questions I have posed can only be answered through suggestive, fragmentary evidence or when triangulated from surviving textual materials, but it seems to me that the only way to answer questions about shifting historical horizons and audience expectations is in fact through the accumulation of specific case studies. The one offered here is both significant and evocative, but will eventually be most useful in comparison with others.

My reasons for choosing this material are several. First, the Danish film industry was unusually precocious in the years before World War I, so it gives us an early example of successful studio production at a time when some of the initial shifts in thinking about location are particularly visible. The Nordisk Film Company, founded in 1906, became one of the world's leading studios by 1913, in that year second only to the French company Pathé Frères in terms of film production. But the size and output of the company were not so large as to make the trends unobservable. The Danish film industry also played a leading role internationally in the development of the multi-reel feature film around 1910–1911, so the development of a sustained sense of fictional place proves acutely interesting in this cinematic tradition from an early stage. Furthermore, the entrepreneur behind the Nordisk Film Company, Ole Olsen, was a clever businessman who came to own a vast distribution and cinema network, not just in Denmark but throughout Germany as well. His savvy international marketing strategy included a calculation that prewar audiences would prefer a placeless product, one that traveled freely between markets, unburdened by obvious cultural signifiers. In the time and place of early Danish cinema, this was accomplished through a combination of place substitution and the erasure of site specificity from the fictional worlds of the films.

What I am calling the "erasure" of "site specificity" counters recent thinking about "small-nation filmmaking," as theorized by Mette Hjort in her analysis of contemporary global cinema. Her analysis of the "small nation" as a category of film production includes a common set of strategies that are necessarily deployed today in countries with small domestic markets. One of the primary strategies of small-nation filmmaking, according to Hjort, is to make a virtue of locality, to sell cultural and geographic difference in an appealing way to a niche market. This allows small-scale production to compete in the larger distribution networks with a unique product. The early Danish studio, with the flamboyant Ole Olsen at its head, initially pursued quite a different strategy: namely, to compete from a small-nation position on the same turf as larger film-producing countries like France, Britain, Italy, and the United States by creating a "placeless" product. In the years before World War I, it should be emphasized, none of Nordisk's competitors had yet developed to the degree that Hollywood would later do, so in this period before 1914, there were possibilities for Danish cinema that would be closed off after the war. The reasons suggested for that sudden decline have included both aesthetic and economic causal factors, but one of its effects included a shift in thinking about cinematic location.

---

Excerpted from Silent Cinema and the Politics of Space by Jennifer M. Bean, Anupama Kapse, Laura Horak. Copyright © 2014 Indiana University Press. Excerpted by permission of Indiana University Press.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.

---