Placing the Location Outtakes of Shoah

Abstract

This article considers how to read the location outtakes of Claude Lanzmann’s film Shoah (1985). These film reels — now digitized and available on the United States Holocaust Memorial Museum’s website — often seem to offer little content, and demand different ways of viewing to make any sense of them. I link them to two other types of viewing experience: watching landscapes filmed by Danièle Huillet and Jean-Marie Straub and using online maps such as Google Street View to identify and navigate particular locations. Engaging with the outtakes in this way involves a consideration of the events of the Holocaust indexed by the locations and the images that Lanzmann made of them. It also makes it possible to glean something of the working methods of Lanzmann and his team. Finally, it offers a particular spatial experience in its own right, which, I argue, can serve as a form of Holocaust memorialization.

KEYWORDS:

• Claude Lanzmann
• digitization
• Danièle Huillet
• Jean-Marie Straub
• online maps
• Holocaust
• space and place

Introduction

Claude Lanzmann’s film Shoah is famous for its duration (nine and a half hours), its uncompromising approach (eschewing archive footage) and for the lengthy process of its creation (1974–85). About five years of that time were spent editing the 220 hours of footage into the pattern of the final film, a complex interweaving of voices, faces and places. Among the hundreds of hours of outtakes that were left, mainly from interviews with Holocaust survivors, witnesses and perpetrators, there are about thirty hours of location footage.¹ This material comprises shots not only of camps and their surroundings, but also of the residences and locales of some of Lanzmann’s interviewees, as well as other places visited by the film crew. Unedited, often silent, repetitive and lacking any explanation, these film reels have attracted limited attention, even though place is a central concern of Shoah, and even though they are now easily available to researchers, digitized and with open access on the United States Holocaust Memorial Museum (USHMM)’s website.² In this article, I shall consider the ease of viewing and difficulty of reading this footage to see how it prompts us to think about place in Shoah, and in the Holocaust.

I shall do so by bringing together two very different ways of conceiving of the outtakes and of space and place. The first is to compare it to the radical film-making practice of Danièle Huillet and Jean-Marie Straub. This allows me to situate the outtakes in the context of when they were filmed, and the set of discussions that occurred then, marked by film theory, Deleuzian and Lacanian terminology and references to French high culture, including the poetry of Mallarmé. I will suggest that these discussions of Huillet and Straub provide ways to read Lanzmann’s landscape outtakes, and will focus on the example of the footage of Nisko and Zarzecze (associated with the so-called Nisko-Lublin plan, an early Nazi scheme, partially implemented and soon abandoned, of deporting Jews and Roma to the east). The second is to compare it to the ways in which space can be explored now on the Internet, in the form of online maps. This places the outtakes in the context in which they can now be received. I will reflect on my own use of the outtakes along with Google Maps and Streetview as an example of how space is produced and represented online. Using these resources to pinpoint where an image was recorded offers insights into the making of the film, but also sets the viewer-researcher on a ‘mnemonic journey’ that invites a different kind of reflection on the Holocaust and its spaces.

This admittedly strange-seeming disjunction will allow me to show the complexities as well as possibilities of making sense of the outtakes. As a mode of reading that does not straightforwardly resolve contradictions, it seems to me to be truer to their form, which, as Jennifer Cazenave (2019) argues, enables the ‘coexistence — rather than mutual exclusion — of incompossible worlds’ (47). It also speaks to, and to some extent reproduces, the dichotomy between the poetic and the cartographic visual styles identified by Richard Carter-White (2013) in Lanzmann’s later film Sobibor, October 14, 1943, 4 pm (2001). Carter-White argues that the tension between these two framings of space is resolved through editing so that they have distinct functions and meanings. But the two approaches I take here are more possible routes for a viewer trying to navigate material with few of the signposts furnished by conventional editing. Indeed, viewers do not even have to choose between the two, as the provisional nature of the footage allows them not to have to commit to one way of viewing. This too, I argue, is part of the spatial experience, and memory of space, made possible by the outtakes.

‘Nothing will have taken place but place’: reading Huillet and Straub alongside Shoah’s landscape outtakes

Claude Lanzmann’s work is now much better known than that of fellow French film-makers Danièle Huillet and Jean-Marie Straub.³ At the time that he was recording and editing Shoah, however, the duo received a great deal of attention in philosophically inclined and often politically radical film magazines such as Cahiers du cinéma and Screen. And there remains an interest in their oeuvre among cinema critics and fim studies scholars. Among the latter, possible connections between Shoah and the films of Huillet and Straub have been mooted before, but rarely discussed in depth. Michael Walsh points out the closeness of the approaches taken in Lanzmann’s film and by Huillet/Straub ‘who have made a whole cinema of tours of ordinary-looking hillsides, islands, and other places associated with World War Two Resistance’ (Walsh 2022: 142). Georges Didi-Huberman (2007) also notes the visual resemblances, although he ascribes them not to ‘an aesthetic decision, such as the one Straub [sic] made, but rather […] an ethical constraint’ (114).⁴ D. N. Rodowick (1997) has explored the parallels in the greatest detail, drawing on Gilles Deleuze’s comments on the complex relation of word and image in post-war cinema (the cinema of the ‘time-image’) (Deleuze 1989) in general, and in Huillet and Straub’s work in particular. Rodowick offers readings of Simon Srebrnik’s return to the site of Chełmno, and the beginning of Filip Müller’s testimony, seeing an ‘incommensurable complementarity’ (145) between voice and location which renders memory ‘legible even if it is imperceptible’ (148).

Without their accompanying speech-acts, the Polish farmland of Shoah, or the Italian quarries of Fortini Cani (197[6]) that Deleuze discusses, are empty landscapes. With them the land gains historical depth[.] (149)

In fact, the two film practices exhibit a number of similarities, with both employing long static shots, slow pans of landscape (which are often even given the name ‘le plan straubien’ — the Straubian shot),⁵ silence and lengthy passages read aloud. In Fortini/Cani, as the Italian Jewish communist poet Franco Fortini reads from his text discussing fascism and its legacies, the film provides long takes of the Tuscan hillsides.⁶ In Too Early/Too Late (1981) one sequence alternates between 360° pans and static shots of the French countryside, with empty fields and road signs as recurring motifs, while Danièle Huillet reads out Friedrich Engels’s description of the local social and economic conditions. And aside from the parallels in technique, Huillet/Straub shot a number of their films at the time Lanzmann was working on his film, using some of the same cinematographers: William Lubtchansky and Caroline Champetier worked on both Shoah and Too Early/Too Late.

The language and terms used to describe their approaches can also be strikingly similar. In an interview Lanzmann gave with Cahiers du cinema in 1985, he styled himself a ‘topographer, geographer’ dealing with sites (lieux) where it was ‘a case of insufficient grounds for memory’ (une sorte de non-lieu de mémoire) (1990 [1985]: 406).⁷ Discussing Fortini/Cani in the pages of the same magazine eight years earlier, Jean Narboni (in an essay which is a key reference point for Deleuze) had called ‘Straub [sic]’ a ‘topographer, geographer’ interested in sites of memory (lieux de mémoire) (1977: 9). The intervening time, the lack of overt interest the two sets of film-makers showed in each other’s work, and the fact that the phrase ‘lieu de mémoire’ was popularized by the historian Pierre Nora (1984), mean that direct influence here is hard to prove, but the deployment of the same terminological repertoire is suggestive.

Despite all the above, few scholars have made much of the overlaps between Lanzmann’s techniques and those of Huillet/Straub. One likely reason for this comes from their very different ideological positions. While both are interested in the Second World War and fascism, Huillet and Straub give far more room to orthodox Marxist interpretations, including criticism of Israel, and much less of a sense of the uniqueness of the genocide of Jews. It is also probably right to say that one can overstate the technical similarities. The frequent (albeit sometimes disputed) descriptions of Huillet/Straub as ‘austere’ and ‘rigorous’ may indicate greater consistency and uncompromisingness in their use of techniques (see e.g. Champetier 1980), but also reflect a different affective dimension: asking for a kind of critical absorption from their viewers rather than the visceral reaction Shoah can provoke. Huillet and Straub also put more restrictions on their use of sound. Although they did not rule out the possibility of dubbing a film intelligently, they felt it necessary to film with direct sound to respect the spaces they recorded. Lanzmann on the other hand often filmed locations without sound, and dubbed them later (Straub and Huillet 1985, Vice and Williams 2024a).

However, while these features distinguish such films as Fortini/Cani and Too Early/Too Late from the edited version of Shoah, this is less the case with the outtakes. Although carried out with a great deal of care and thought, Huillet and Straub’s self-imposed restrictions on recording and editing give their films some similarities with unedited footage. Rodowick’s claim that the landscapes of Fortini/Cani mean nothing without the voice-over is complicated by a wordless ten-minute sequence in the film, one that is of great significance for Deleuze and the French film critics he follows. Huillet and Straub are often interested in emptiness and stillness, playing voice-over for only a small part of a shot and then continuing to present a landscape in its own right, such as the northern French villages of Too Early/Too Late. My suggestion, therefore, is that while there is probably more to be said about location in the edited version of Shoah and the work of Huillet and Straub, there is certainly something to be explored in making links between the duo’s films and Lanzmann’s unedited and unused footage. In particular, the rich discourse in Gilles Deleuze’s discussions of landscape in Huillet/Straub (Deleuze 1985, 1989) and his dialogue with such critics as Jean Narboni (1977) and Serge Daney (1979) offers ways to respond to Lanzmann’s location outtakes.

Drawing on Narboni’s reading of Fortini/Cani in particular, Deleuze argues that Huillet/Straub’s landscapes need to be read, not just viewed. Narboni concentrates on the sequence in which Franco Fortini stops reading from his text and the film presents ‘an enormous hole in discourse dug out by the Apuan Alps’. Narboni sees this moment as consistent with the approach of ‘Straub [sic]’ to the discourse of history: words tripping [butant] over stones. Narboni calls these landscapes ‘sites of memory’ (lieux de mémoire), seeing them as creating a moment of silence, something that crosses out (barre) Fortini’s version of history (1979: 9). Place here seems to negate or work against language. The lieu de mémoire (which Deleuze seems to treat as Narboni’s coinage (1985: 318n37)) is a hole in historical discourse, an encounter with the Real. As part of this conception, Narboni makes two explicit references to Mallarmé, and the line in the poem Un coup de dés: ‘rien n’aura eu lieu que le lieu’ — 'nothing will have taken place but place'. Narboni refers to this line because Huillet and Straub were about to film the poem (with Lubtchansky as one of the cinematographers), recited by a number of speakers, in Père Lachaise cemetery, on ground between the monument to the Paris Commune — which is shown — and monuments to WWII deportees — which are not (Huillet and Straub 1977).

Narboni’s reading points to the possibility of a space that is outside or before language: the stones on which words founder, or the space in which words arise. Deleuze acknowledges the importance of silence for Narboni and Serge Daney, suggesting that the spaces are not simply mute, but rather demand an archaeological reading, an active looking that makes connections within and beyond what we see. At times, this includes the voice-over, such as in the following discussion of Huillet/Straub’s films:

Deserted ground is the only thing that can be seen, but this deserted ground is heavy with what lies beneath. […] And if the voice [over] speaks to us of corpses, of the whole lineage of corpses that come to reside underground, at this very moment the slightest quivering of wind on the deserted ground, on the empty space under your eyes, the slightest hollow in this ground — all of this takes on meaning [tout cela prend sens]. (Deleuze 1998: 16–17, translation adapted)

But a link between voice-over and image is not the only possible connection. Deleuze’s fundamental idea of the cinema of the time-image, and that of Huillet and Straub in particular, is that speech and image are able to function apart from each other: ‘heard speech ceases to make seen and be seen’ while ‘the visual image attains to the new readability of things on its own account’ (Deleuze 1989: 245). Huillet and Straub’s landscapes can be described as ‘empty and lacunary’, Deleuze writes, but ‘empty space […] has a fullness in which there is nothing missing’. In these ‘ambiguous landscapes themselves, there is produced a whole “coalescence” of the perceived with the remembered, the imagined, the known’ (244). As Deleuze acknowledges, he draws the phrase ‘coalescence’ from Serge Daney’s essay on Huillet/Straub, where Daney uses it to describe the connections and contradictions between the ‘perceived’ and the ‘known’ for viewers of Fortini/Cani and Toute revolution est un coup de dés (1979: 6). Both films, Daney notes, present landscapes in which dead bodies lie hidden (6). Deleuze expands on Daney’s reading to suggest multiple ways in which to relate the hidden and the seen, including viewers’ memory and imagination as well as knowledge. And this seems to me to be the most appropriate way of drawing on these discussions to read the outtakes of Shoah. I will now turn to an example where I think this applies: the footage of the fields at Zarzecze and the river San, close to Nisko.

Lanzmann returned to the outtakes for Shoah repeatedly in the three decades between its release and his death. But the material he re-edited into other films was almost entirely from interviews, and almost never from the location shots.⁸ The Last of the Unjust (2013), Lanzmann’s film portrait of Benjamin Murmelstein (last, and only surviving, ‘Elder’ of the Theresienstadt ghetto), made use of the many hours of footage filmed with him in 1977. The location outtakes include footage that was almost certainly shot to accompany Murmelstein’s testimony: a few minutes taken in the vicinity of Nisko and Zarzecze.⁹ For The Last of the Unjust, Lanzmann revisited this site in his late eighties, standing at Nisko station and reading a text explaining the place of the ‘Nisko plan’ in the development of the Final Solution. For a few months in late 1939 and early 1940, an area in the south-east of German-occupied Polish territory close to the demarcation line with the Soviet Union was used as a dumping ground for Jews, the first destination for ‘deportation to the East’. Murmelstein tells Lanzmann that he spent about 25 days there, among the Jews taken to an empty field in Zarzecze, on a small hill overlooking the river San. They were instructed by Adolf Eichmann to build dwellings for their SS guards and then for themselves, and informed that they would need to dig their own wells as the existing ones were contaminated with cholera. Some escaped or were driven over the demarcation line about 50 km away.¹⁰ Accompanying Murmelstein and Lanzmann’s words, the sequence of location shots for The Last of the Unjust enacts the former’s account: panning over fields and showing the San River.

It would have been possible for Lanzmann to use voice-overs from interviewees included in Shoah. Richard Glazar and Ruth Elias both briefly mention the deportations to Nisko from Moravská Ostrava (Mährisch Ostrau), which Elias’s father managed to escape, and which Glazar’s father did not.¹¹ But there is no discussion of this plan in Shoah, which is more interested in the last step taken to the Final Solution rather than the evolving and abandoned schemes that preceded it. Lanzmann also decided not to use this location footage in The Last of the Unjust. This decision gives some indication of his thinking about place, and how it could be represented. Aspect ratio, the quality of the film stock and the landscape itself would all have placed these shots at a point in the past: at the same time as his interview with Murmelstein. Even though it was therefore shot closer in time to the events Murmelstein described, and at a season similar to the one in which they took place, Lanzmann chose to film the site again. The location outtakes were not only too late, but also too early to be included. Too late, because they no longer embodied the event. Too early, in that Lanzmann’s approach demanded that the locations appear in the ‘now’ of the film to demonstrate that the Shoah is not in — or of — the past.¹²

Left without any voice-over, although perhaps bringing knowledge of the Nisko plan or of what Murmelstein said about it to Lanzmann, we see that there is very little to see (Figure 1). The footage is very short: a few minutes of film that USHMM has catalogued alongside footage of Kraków (Lanzmann 1979b). All that identifies this landscape are written words: those on the roll of paper (which might even be toilet paper) serving as a film slate (Nisko and the number of the shots 1–5) and those on a road sign north of the village which indicates that Nisko is 4 kilometres away. There are no major features other than the river San: fields, a few trees, a handful of houses and farm buildings, a small hill. In the initial shot, the camera seems to be placed on this hill, looking over fields with the San a mile or two away, gleaming in the sunlight, just below the horizon line. This seems to be the same place that Murmelstein mentioned, and that Lanzmann was to film some forty years later: there are very few other hills in the area. Unlike The Last of the Unjust, where empty fields are presented as the sites where Murmelstein stood and listened and Eichmann stood and spoke, this outtake footage invites the viewer simply to watch this landscape, with much less guidance on how to place action or people in it. The features are mostly filmed at some distance, and difficult to see in any detail. The sky veiled by white cloud and fields thinly covered in snow impart a silvery grey palette to the entire frame. Very little moves beyond one or two dots of traffic in the background, the slight swaying of branches in the wind, a single bird that darts across the frame, and the eddies and ripples in the river. In contrast to the twenty-first-century footage from Zarzecze, there is no sound, and very little sight, of traffic. The camera’s panning (about 60 or 70 degrees) is by far the most significant thing happening (what we might call the greatest narrative element) in these shots.

FIGURE 1 View from Zarzecze over fields and the River San. From Lanzmann (1979b).

Similar fields are included in Shoah, as a site of what seems to be something like emptiness, between the end of the interview with Franz Grassler and the brief interlude when Gertrude Schneider and her mother sing. The emptiness of the fields at Zarzecze (or what can be interpreted as their emptiness) can be made to stand for the inaccessibility of the past, or the silence of the dead. It can also prompt thoughts about what the deportees faced when left here: a site of limited resources chosen precisely because those resources were limited.

But Deleuze’s comments on Huillet and Straub, it seems to me, offer a way into thinking about this footage and its relationship to the history of the Holocaust which does more than simply see what is not there. Deleuze’s point is that the landscape is transformed by what we know took place here. The tiny flickers of movement noticeable in these shots are cognate to those that Deleuze says take on meaning (or sense) in Huillet/Straub’s films. This phrase needs to be seen in the context of his other writings on sense and meaning, in which he resists thinking based on resemblance and correspondence. One of the points is that these movements indicate — and produce meaning from — a difference between the landscape and what has taken place in it.

Following Deleuze, I want to argue that rather than seeing that difference as simply being about absence, it could be read in other terms. One possibility is to think about time, the key term for Deleuze’s thinking about post-war cinema. It is also key to a recurring point of reference in the discussion above, the poetry of Mallarmé and Huillet/Straub’s film of Un coup de dés. While Mallarmé’s lines ‘nothing … will have taken place … but place’ look forward (and then back) to a space without content, the very act of doing so also relies on that space being marked by time. Joshua Ramey sums up Deleuze’s thinking here as follows:

Cinema at its best, for Deleuze, reveals how our actions are haunted by glitches and passivities signaling that our bodies and minds are tied to and extended through discontinuous durations. We live not only the time of the diurnal, but also through tectonic, geophysical, animal, and even world-historical modes of duration that overlap but are discontinuous with our conscious experience of time. (Ramey 2012: 149)¹³

The stirrings of the countryside around Zarzecze suggest that this was an environment, not just an empty space with a name that can be attached to it. This is a space that has its own rhythms, that contrast with the time code ticking away at the bottom of the screen, partly ones given to it by human action and interaction, but also through the seasons, the weather, the way in which it is inhabited by non-human animals, and shaped by the river. The rhythms thus delineated give us a sense not just of the moment in the past. They subtly show the passing of time between the event and now. Lanzmann often wants to collapse that difference, but the sites that he films themselves give a sense of time that is not simply about the moment of the event and the moment of its recall. Meaning here, then, is not simply reconstructing what happened but an awareness of the difference between where we are now and how it was then. And that difference is important for us to acknowledge.

Online space and geolocating: from Treblinka to Tramelan

So far, I have been suggesting ways that an example of footage might be viewable as if it were a film in its own right, drawing on terms used to discuss art cinema contemporaneous with Shoah. Locating it in this context, however, risks ignoring the form in which the footage is now accessible: not as a film projected in a cinema, but as an online streaming file, played on a screen on which many other types of file — text, image or audio, editable or read-only — are available. This is also how film-watching in general can now take place, on a multi-purpose screen at home which enables repeated viewings and pauses to examine single frames. The location outtakes for Shoah, on the other hand, are organized as an archive, catalogued and described in a way aimed at a researcher, with transcripts and translations provided, and some degree of searchability. So, rather than being like a streaming service, curated or tailored to a viewing programme or some sense of a viewer’s interests,¹⁴ these films are presented as a digital research tool, in a similar way to the online video archives of Fortunoff and the Shoah Foundation.

Paul Frosh (2018) discusses the ways that viewing digitized video testimony involves switching between hermeneutic attention, which gives rise to questions such as ‘What do they mean? What knowledge do they impart?’ and operative attention which asks ‘How will this text or image respond? What connections to the world does it enable?’ (360), with both being bodily experiences: the former depending on ‘an act of intense body-consciousness and sensorimotor restraint’ (361) and the latter a form of ‘fidgeting honed to the promise of device responsiveness: sensorimotor restlessness as a system requirement’ (360). Frosh sees both the dangers and possibilities in this set-up, which produces a new situation marked by the ‘ethics of kinaesthetics’ (364–65). Viewers are now embodied and distractable. Todd Presner (2016) too considers the ways that online testimony makes possible a different kind of listening — what he calls distant listening — as an ethical challenge but also an ethical possibility.

Distractibility and distance seem to be even more likely viewer experiences of the location outtakes, which in comparison to the video testimony of a survivor usually have no real story to convey, and make much less of an ethical demand to attend to them, even when the footage is of such sites as Treblinka or Auschwitz. There seems much less to prevent a viewer pausing the film at any point, or switching from it to another online source such as a map, as one of the possible connections. There are also features that are more specific to this location footage that do not come under Frosh’s schema. Whereas sitting and watching video testimony is physically cognate with sitting and listening to a survivor, watching location footage online involves being aware of the placement and movement of a camera, and a body operating it. Interacting physically with that footage — by stopping, replaying, or comparing it to something else — disrupts the flow of the image and the sense of the time and movement within it, and so is much more likely to interfere with one’s awareness of the bodily actions of filming. The viewer can act upon the image to change it, but the means (mouse, screen) and gestures (clicking, scrolling) that are required are so detached from the movements one would make to walk that they de-realize what one sees and make it more impersonal than simply watching the film. With the location outtakes particularly, then, the experience of viewing them online has much more potential to break down the boundaries between the footage itself and the material to which it is compared. Streaming the digitized footage of a Polish country road, or a Swiss motorway, is not very different from following the same route on Google Streetview.

In working with the outtakes on a collaborative project with Sue Vice, I have often compared them to other online resources to be able to locate the places being shown, paying particular attention to road signs (especially when they usefully included distances on them) landmarks and the USHMM’s own cataloguing and description. My motivations for doing so were as a researcher, trying to make some sense of the context of Shoah and the way in which it presented certain sites. But the resources I used (beyond the research time allocated to me as part of my job), were open-access ones that anyone else could employ: especially Google Maps and Google Streetview (resources which are also, it should be said, integrated with the Shoah Foundation VHA video testimony). We started with a set of research questions, but, as I shall show, the experience of working with this material, and the viewer position that I occupied, also demands reflection.

Thinking first of the sites investigated by Lanzmann where killing took place, I started with the footage of Bełżec, the first of the Aktion Reinhard camps in German-occupied Poland, which appears only fleetingly in Shoah (Lanzmann 1979a). An examination of the outtakes alongside the film, and alongside other maps and images of Bełżec allowed me to identify with a reasonable degree of certainty where the shots were taken: a site outside the bounds both of the memorial area, and of the historical camp. Locating the site at which the footage was shot helped demonstrate that although Lanzmann shows himself taking interest in delineating the precise boundaries of camps, especially Sobibór, he also made use of the landscape around them symbolically, the emptiness of the sandpit standing for the absence of the victims’ bodies (Vice and Williams 2021). Closeness and distance from the camps is explored in both geographical and symbolic terms in the location footage. Lanzmann is interested, for example, in the farmer who lives next to Treblinka station, Czesław Borowy, and the location outtakes (Lanzmann 1979f) and online maps helped to locate where that house is, and reinforced the fact that it was some way from the extermination site itself, showing that even Lanzmann’s archetypal bystander was less nearby than a viewer of the film might think.¹⁵

But while the location outtakes are mostly taken in Poland, close to the sites that are central to the film, a significant proportion of them stem from elsewhere, often at some considerable distance. For example, there are three hours of footage that USHMM labels as having been filmed in Germany and Switzerland (Lanzmann 1979c). Some of this is readily identifiable from the catalogue descriptions or the landmarks shown. Here too, using maps where shots were taken helped to indicate what was important or not for Lanzmann (e.g. nothing is made of the fact that the Munich beer hall in which a perpetrator is working lies on the route of the Beer Hall Putsch march) (Vice and Williams 2024b). But it also offered some places that were hard to identify, and puzzling even when located, for example the footage of Switzerland, which includes material filmed in a car travelling on Swiss autoroutes, and on local roads.

Working from road signs and landmarks in the outtakes, and maps and street views online, I discovered that some of these rural routes lay around the Swiss village of Tramelan. They would seem to have been taken to accompany the interview with Maurice Rossel, who is reported to have lived there (Sartoretti 2022).¹⁶ Rossel was the Red Cross representative who visited (but did not enter) Auschwitz, and wrote a report on Theresienstadt, the ghetto that masqueraded as a model settlement for Jews, doing little other than repeating what he was told by German officials. Identifying the location reveals that Lanzmann probably entertained the idea of placing Rossel in his rural setting (Rossel calls himself someone from a village several times in his interview), and may even indicate that he considered using the surveillance van used in the secret filming of perpetrators, as it appears in some of these shots.

Google Streetview was one of the ways that I was able to confirm these locations, because it some cases it is possible to replicate some of the views very closely indeed. (Figures 2 and 3).

FIGURE 2 Treblinka, Poland. Location footage from Lanzmann (1979f). Compare the image of Treblinka village on Google Street View at: https://maps.app.goo.gl/VSRXYp1kTpCY5fN87.

FIGURE 3 Tramelan, Switzerland. Location footage from Lanzmann (1979c). Compare the image from Tramelan at Google Street View at: https://maps.app.goo.gl/3nSpV3AakVq89FgW8.

Being able to map (and to replicate) the locations Lanzmann and his team visited enables a better sense of the material practicalities and the embodied experience of making this journey and the film. It prompts thoughts about what kinds of decisions were made to move through space and to record space. The footage becomes a means to trace the journeys that Lanzmann and his crew took round Europe, and to notice what they were interested in recording. But at some point, it started to feel inadequate to see geolocating shots simply as a way to make sense of how Shoah was filmed. It seemed to me that it was becoming an experience in its own right. It affected how I viewed the outtakes: at times, for example, I found myself clicking on and trying to rotate Lanzmann’s images as if they were 360° shots taken by a Streetview camera. And that experience of using the images as indices of particular locations became not simply a form of mapping but also a kind of meditation on the Holocaust.

Mark Godfrey suggests that the experience of some artworks (that he defines as abstract) addressing the Holocaust is a ‘mnemonic journey’. Referring to Susan Hiller’s J Street Project and the Eisenman Memorial in Berlin, he states:

For Hiller, as for Eisenman, the path of memory is an empty road. The viewer or visitor enters this road literally (in Eisenman’s field) or imaginatively (before Hiller’s photographs and video). They embark upon a mnemonic journey, but their choice to reflect on the history and implications of the Holocaust is only offered, and never determined by the work before them. (Godfrey 2007: 264)

What Godfrey calls emptiness here is not the literal lack of other travellers on this road: the video of and photographs for Hiller’s J Street Project include traffic and pedestrians,¹⁷ and the paths between the stelae of the Eisenman memorial are often occupied by other visitors. Rather, it is the lack of obvious direction to the journey or sights on the way. In what he calls ‘mnemonic environments’ ‘memory is never marshalled or directed towards a specific image or symbol’ (252).

That emptiness is achieved by a similar lack of referentiality in Lanzmann’s outtakes of roads, especially when experienced alongside very similar-looking shots from Google Streetview. Lanzmann called the sites of mass murder ‘non-sites of memory’ (non-lieux de mémoire). These other locations could simply be called ‘non-places’ (non-lieux), a term usually attributed to Marc Augé (1992), although he draws it, while defining it differently, from Michel de Certeau (1990 [1980]; see Conley 2012: 66). For Augé ‘non-lieux’ are locations without identities, relations or history (1992: 100), and specifically include motorways (120). I would suggest that other roads (and possibly most places) mediated by film and online maps are also experienced in this way.¹⁸ Tracing these journeys feels rather affectless and dull, as well as being actually quite weakly embodied. All of these features are a problem for Augé, but for Certeau ‘non-lieux’ offer ways to escape the fixity of ‘lieux’, opening out to the possibilities of movement through ‘espace’ (space). This can even take the form of attention wandering during a train journey. The physical constraints on passengers’ movement, and their being cut off from the landscape that they view and pass through, result in more than gaining ‘optical mastery of space’ at the cost of ‘leaving behind any proper place, losing one’s footing’ (en quittant tout lieu propre, en perdant pied). They also, Certeau argues, permit thoughts of ‘unknown landscapes and […] inner histories’, which sit ‘outside these things but not without them’ (1990 [1980]: 167).¹⁹

Certeau suggests that physical and technical arrangements that seem only to create disconnection can actually help form other tenuous but meaningful links between viewed and imagined landscapes. The experience of the locations of the outtakes as ‘non-places’, caused by accessing them online, can be seen as an opportunity of this kind. The viewer’s virtual, mnemonic journey takes place outside, but not without, the landscapes of Switzerland and Poland. Treblinka and Tramelan as points on this journey are apart from, but not without, each other. Turning them into featureless coordinates on a map, and ‘travelling’ (in virtual form) down the empty roads around and between them, prompt a different kind of awareness of, and ensuing reflections on, the ways in which they are linked to each other.

The process of research, like the process as well as the end-product of film-making, might therefore be called a memorial or aesthetic experience in its own right. This suggests an aesthetics of how we approach the archive, not simply a way of thinking about the archive as a source of information, and not just an aesthetics of archiving, but what we might call a kind of aesthetics of research. This too, can be theorized as a mode of watching. Searching these sites, the obsessiveness that characterizes any research project starts to map onto the kind of obsessiveness that Lanzmann attributes to himself. The very act of researching it is part of what makes this footage into an investigation of the sites of the Holocaust.

Conclusion

Claude Lanzmann claimed to have begun work on his film Shoah with no interest in filming at the sites of mass killing and extermination, because he believed there was nothing to see there (Jeffries 2011). Even after visiting them, and at the end of the decade-long process of filming and editing, that belief does not seem to have changed. Instead, he made sense of his film as doing something with the fact that nothing could be seen. And what was done relied on what was achieved in the editing suite, contrasting the emptiness of the sites that were devoid of traces, identifiable only by their names, with voice-overs speaking of what occurred there.²⁰

The location outtakes pose a challenge, therefore, as they often appear to offer nothing but that emptiness. In the two approaches to them that I have sketched out above, I have suggested ways that do not simply try to fill what seems empty, but to think differently about it. In the first, drawing on contemporary film critics and theorists’ responses to the emptiness and silence of some landscapes filmed by Huillet and Straub, I suggest that the landscape outtakes of Shoah can be read in a similar way. Watching the fields of Zarzecze becomes a way of thinking not just about what happened or the fact that there is nothing to see of what happened, but also the ways in which this environment might have registered those events within a set of different timescales and cycles. In the second, tracing journeys down roads that are empty of any visible link to the past by switching between online footage and online maps flattens space in one way to draw connections in others.

In this article, I have deliberately yoked together two ostensibly very different or even incongruous contexts in which these outtakes might be located. This is partly to show that the differences between them are less marked than might be thought. Watching the outtakes as if they were a film by Huillet and Straub happens alongside and is sometimes influenced by seeing the changing timecode at the bottom of the frame. Viewing the footage taken on different roads alongside Google Street View can be interpreted through the lens of video and photographic art. But it is also to reflect the complexity of the viewer’s position: confronted with material that invites choices of how to act upon — or navigate — it, at the same time as it enables the eschewal of decisions. These are both ways of experiencing space: the feeling of being within somewhere and moving through it, and the feeling of standing back and contemplating it. Both of these experiences involve embodied and reflective elements. The outtakes allow the boundaries between those elements to be particularly permeable. Their complexity and contradiction enable different ways of thinking and feeling about space to coexist. So while it is possible to extrapolate insights from the outtakes to say that the sites of the Holocaust are less easily placeable and locatable than Lanzmann sometimes seems to suggest, or that bystanders can occupy long-distance as well as nearby positions, it is important to recognize that viewing this material need not provide some kind of historical argument, but can offer a particular experience of history. It is only through watching and working with the outtakes and reflecting on the process of doing so, that this experience can take place.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Dominic Williams

Dominic Williams is Assistant Professor in Holocaust and Genocide Studies at Northumbria University, UK. With Nicholas Chare, he has co-edited and co-authored four books in Holocaust Studies, most recently Testimonies of Resistance: Representations of the Auschwitz-Birkenau Sonderkommando (Berghahn Books, 2019, paperback 2023). He is currently co-editing (with Sarah Cushman and Joanne Pettitt) the Routledge Handbook to Auschwitz-Birkenau and (with Nicolas Barnett and Nicholas Chare) The Clinical Witness: Conflict, Catastrophe and Medical Testimony.

Notes

1 The outtakes were created by Claude Lanzmann during the filming of Shoah and are used and cited by permission of the USHMM and Yad Vashem, the Holocaust Martyrs’ and Heroes’ Remembrance Authority, Jerusalem.

2 See Vice (2020) and Vice and Williams (2021, 2024a, 2024b). This is in contrast to the much greater attention afforded to the interview outtakes, including three recent books (Cazenave 2019, McGlothlin et al. 2020, Vice 2021). On place in Shoah see Olin (1997) and Didi-Huberman (2007).

3 I follow the order of names used by the recent Editions Montparnasse DVD series of Huillet and Straub’s films. This is different from ‘Straub/Huillet’ as used by many English-language writers, or the tendency to credit Straub alone or call them ‘les Straub’ among French-language critics. Huillet and Straub used different name orders in the credits for different films, e.g. Fortini/Cani (1976) credits production to ‘Straub-Huillet’, but Too Early/Too Late (1981) puts Danièle Huillet before Jean-Marie Straub.

4 Shoah and Fortini/Cani are also mentioned in passing in Levin (1998: 87) and Marks (2003: 122).

5 Lübecker (2022: 74). The term was coined by Serge Daney (1979), although he defined it more through a set of paradoxical tensions, including the ‘triple resistance’ of ‘texts to bodies, places to texts and bodies to places’ (5). Daney saw Huillet and Straub as seeking truth in such moments of resistance (6).

6 The title of the film is usually written with a slash. There is no title screen in the film itself beyond a shot of the cover of Fortini’s I cani del Sinai. As the book’s title indicates, Fortini’s text was a response to the Six Day War of 1967 between Israel and an alliance of mainly neighbouring Arab states.

7 My translation attempts to capture the legal meaning of ‘non-lieu’ — a case which is unable to proceed for lack of evidence.

8 The sole exception is the earliest ‘supplement’ to Shoah (1985), A Visitor from the Living (1997), which uses location footage from Theresienstadt.

9 The names of these places indicate something of their topography: Nisko: low area, Zarzecze: beyond the river (Bijak 2021). This second one is explained by Murmelstein to Lanzmann.

10 For the Nisko plan see Cesarani (2017: 258–260) and Moser (2012). For the Last of the Unjust see Vice (2015).

11 Lanzmann (1979d: 2 of the transcript, 1979e: 3 of the transcript).

12 The footage of Nisko and Zarzecze filmed for and included in The Last of the Unjust also covers other areas: woods, marshes, a river which stands for the demarcation line between German and Soviet occupying forces, marshes and the nightclub ‘Sydney’. So it may well also be that the earlier outtakes did not provide enough material to cover the interview with Murmelstein.

13 Thanks to Marcel Swiboda for drawing my attention to this book and these specific sentences.

14 Of course, many of the films we now watch, including some of those I have seen by Huillet and Straub, are available in online collections, such as the MUBI season of summer 2019 in the UK.

15 Raul Hilberg credits Lanzmann as inspiring his inclusion of bystanders, and uses Borowy’s words as the epigraph to the section on bystanders in Hilberg (1993: 193).

16 Lanzmann later released an edited version of this interview as a separate film, A Visitor from the Living (1999). On the film see Freudiger (2008).

17 E.g. images 61, 96, 143, 169; 193 in Hiller (2005).

18 The idea of the internet as a non-place is articulated in Bolter and Grusin (1999).

19 Nigel Thrift (2004) and Verena Conley (2012) note that Certeau’s archetype of movement in space is walking. They both focus on the beginning of the chapter ‘Naval et carcéral’ to support their claims that Certeau provides an entirely negative picture of train travel, but as I read it the chapter develops to consider other ways in which movement, and escape from fixity, can take place even while passengers remain in their seats.

20 This of course ignores the fact that traces are identifiable at these places, and have been over the last seventy years, from the bodies and body parts shown in liberation footage, to the finding of fragments of bone in, e.g. Peter Morley’s Kitty: Return to Auschwitz (1979), Guillaume Moscovitz’s Bełżec (2005) and Christophe Cognet’s From Where They Stood (2020). It is also possible to read Shoah as having a rather more complex relationship between a space and memory, especially at such sites as Treblinka, with its memorial, and Auschwitz, with the ruins of its buildings.