Abstract
In 2014, the British Imperial War Museum (IWM) contracted New Zealand-based filmmaker, Peter Jackson, to use their audio and video archives to create a media-based memorial to the men who served in World War 1. The documentary film, They Shall Not Grow Old (TSNGO), released in 2018, was the product of this collaboration. Jackson took on the project to better understand his own grandfather’s experience as a soldier at the Battle of the Somme. Weta Digital Studios, founded by Jackson, converted the standard WWI newsreel footage into a product that aligned to a modern audience’s perceptual sense of truth. Weta Digital redrew and colored each frame, a process that is strikingly similar to CGI animation. Curiously, Jackson, a filmmaker whose career has hinged on his ability to collaborate with CGI animators, does not describe his new historical film treatment as animation. This article first argues that TSNGO is following in the footsteps of previous CGI animated films, in particular those that have re-edited historical footage, but more importantly asks: why would Jackson prefer to keep the word ‘animation’ out of the discussion about his new historical documentary? The answer to this question leads us to a critical discussion about how animation has become both the preserver and ‘re-imaginer’ of existing historical archives.
Keywords
Find out about your own families. The only message I’d say at the end of the day is forget the politics. Forget the reasons why the war started or finished. Just find out if you had a great grandfather or a great great uncle. Find out because you’ll be amazed about the stories you can unfold about your own family, your own DNA. You are carrying their DNA. If you are not interested now I guarantee you’ll be surprised. Just dig deep. Dig deep. (Peter Jackson at the 2018 London Film Festival)
Digging deep and filling gaps
In 2018, Peter Jackson released They Shall Not Grow Old (TSNGO), a joint project with the Imperial War Museum (IWM) and the BBC to commemorate the centennial anniversary of World War I. The popular filmmaker was initially approached by the IWM in 2014 to use its WWI footage in a ‘fresh and original way’ (Jackson, 2018e), and they presented Jackson with 100 hours of archival videos and 600 hours of audio interviews. What is most striking about TSNGO was the subsequent restoration and retreatment of the footage by Jackson and his New Zealand–based special effects company, Weta Digital, into full color, running at 24 frames per second (fps), as opposed to its original 10–18 fps. To a modern audience and a majority of critics, Jackson had created a more realistic depiction of World War I than we had ever seen, and he would comment in the DVD featured interview that he was ‘“absolutely amazed”. . . that the jerky figures in the archival footage suddenly became “human”’ (Jackson, 2018a). Peter Jackson had, quite literally, filled in the gaps (frames) that had been lost to time.
Jackson also realized the project would become one of personal import, as his grandfather, William Jackson, enlisted in the Second Battalion South Wales Borderers in 1910 when he was a 20-year-old. He would go on to fight at Gallipoli where he was injured by a German machine gunner. The injury would send William Jackson home to marry Peter’s grandmother, Harriet. William Jackson died in 1940 without meeting his grandson. Though Jackson was told the highlights of his grandfather’s war experience, including that William had fought alongside author JRR Tolkien at the Battle of the Somme, this project made him realize there were gaps in his grandfather’s story he felt compelled to fill. He states: I was always thinking ‘well my grandfather would have experienced all this. This would have been what he ate. This would have been how he slept.’ So it was actually great in the lack of personal information that I’ve got on him, I was able to paint a picture in my head of what his life was like. (Joseph, 2018)
The result was undoubtedly a remarkable piece of archival curation, and a film that Jackson described as an effort to understand his grandfather’s experiences more closely (Jackson, 2018c). However, I will contend throughout this article that we are no closer to understanding the lives of these soldiers than if the IWM had simply presented the archival material in their original form. What Jackson has done, with a hybrid of CGI and traditional animation techniques, is to help us feel as if we know these stories better by packaging them in a format that appeals to our understanding of contemporary digital aesthetics in relation to perceptual realism, while at the same time maintaining the familiar Eurocentric bias of previous WWI master narratives.
All technological marvels aside, Jackson ultimately undertook a problematically positivist historical project: to fully and authentically convey all Western Front WWI soldiers’ experiences through a relatively small sample. His strategy for developing such a project was to dig into the primary sources provided by the IWM: hundreds of hours of interviews and video footage as well as material archives (uniforms, mess kits, artillery, tanks, etc.) from the period. The process was a means to reconstruct an ‘authentic’ representation of what the experience of WWI was like for the masses of common soldiers like William Jackson and JRR Tolkien. Yet the idea that a ‘common’ soldier’s WWI experience could be captured for a single sitting is preposterous, though Jackson doubles down on the claim that soldiers’ experiences on the Western front were strikingly similar. He stated at the 2018 British Film Festival: What was interesting was that, with 120 soldiers, their experience was pretty common. No matter where they were, no matter what year they were in, they had a very shared experience. The film is like 120 voices telling a common story. Every human story. (Jackson, 2018b)
After months of repairing the IWM footage, Jackson commented that the restoration had recaptured the humanity of the soldiers, claiming that ‘they come to life again. They are not hidden behind this technological fuzz. They are suddenly real people’ (Jackson, 2018b). Jackson would have you believe that his technique is getting his audience to an interpretation which is much closer to a singular historical truth, which is not his or that of the IWM, but the soldiers’ truth, told through their own words and with their own images captured a hundred years ago. Contemporary historians’ most severe criticism of TSNGO was that the individuality of the 120 white men was blended to tell ‘every human story’. Compounding this issue, the men being interviewed are not the same men captured on film, and none of the voices are credited. Though Jackson claims he has restored the soldiers’ humanity, he surely neglected to maintain their individuality.
This article will first take a look at interviews with Jackson leading up to the event of TSNGO’s premiere, noting a curious avoidance of talking about the animation process involved in reconstructing the historical footage, one that parallels the production of classic cel animation. Second, I will provide a brief recap of the relationship between CGI animation and ‘perceptual realism’ as articulated by Stephen Prince (1996). Third, I will draw on the work of contemporary historians arguing that the documentary only portrays experiences of white European men, thus contextualizing the problematic nature of telling a ‘common’ story in a period that values diversity of the individual experience over grand narratives. Finally, I will consider why Jackson needed to distance his project from the term ‘animation’, as any connection with the term would be understood as implying a form of artistic deception.
Mysterious black boxes
The exploitation of technology to alter our visual reality is nothing new. Even before cinema, Johann Wolfgang von Goethe wrote about the deceptions that obscured one from a ‘true’ perception, inspiring anatomist and physiologist Jan Purkinje to explore the persistence and modulation of afterimages (see Cavero et al., 2017). Their work became the philosophical and practical foundations for how an observer can create the illusion of movement through a series of quickly changing still frames. Since then, new technologies regularly dazzle audiences in making increasingly complex spectacles feel true. Today, optical truth has become a question of aesthetics rather than the empirically provable concept that Purkinje sought. The human eye can perceive up to 60 frames per second, 7,000,000 colors and with a turn of the head observe far beyond the 16:9 aspect ratio we’ve become accustomed to in modern film. Yet, present media conventions would have us believe that 24–30 frames a second, a limited color palate and a 2D surface is ideal for portraying a subject ‘realistically’. In essence, it is an audience’s relationship to its current technology that dictates an aesthetic sense of perceptual truth and not the technology alone. Jackson’s performative choices made in TSNGO stood out to critics and scholars as a significant turning point in the treatment of archival footage for documentary film making. Jackson is certainly not the first to use animation for documentary footage. Since the release of Waltz with Bashir (Ari Folman, 2008), the use of animation has been hotly contested by scholars for the apparent distance of the drawn sign from its original referent (see Honess Roe, 2013). Waltz with Bashir remains quite different from TSNGO as it uses animation when photographic images are not available in a process that Annabelle Honess Roe calls ‘non-mimetic substitution’, where animation is ‘a medium that has the potential to express meaning through its aesthetic realization’ (p. 226). Nea Ehrlich (2020: 260) similarly notes that Bashir takes advantage of virtual culture’s ability to discern truth within animation, claiming this use of animation ‘acts as real-time visualization of computer-mediated actions’, requiring us to question our definitions of animation and reconsider ‘its link to the realities depicted and the documentary value of animated representation’.
Though I have yet to find any footage or text of Jackson outright dismissing TSNGO as a work of animation, he is particularly coy when the subject of drawing frames is introduced. Take, for instance, this promotional interview from Fox 5, Washington DC:
You make new frames right?
The computer does it yeah.
Wow!
And I don’t even know how it does it. Please don’t ask me. But somehow it takes the frame before and the frame after and it creates a whole new frame that then gives you a feeling of a smooth movement once you’re back to 24 [frames a second] again. (McCarthy, 2018)
Mysterious black box descriptions of this technology aside, Jackson goes a long way to avoid calling his work in TSNGO animation, though he tells us the computer redraws the missing frames. My research has uncovered that the process is far more complex than ‘the computer does it, yeah’, but I am left with the question: why the need for such evasion? Jackson’s recoloring, shadowing, relighting, and frame recalibration have essentially redrawn the story around a human reference, similar to the collaboration that happens between an animator and subject in the Rotoscope process. The answer perhaps lies in the way Jackson negotiates the relationship between the referent and its original sign. Where Bashir sought to animate the unphotographed in its act of visualization, TSNGO looks to overwrite the original index of the IWM footage and, according to Jackson, somehow moves a simulacrum closer to its original through further reproduction.
Until recently, documentary aesthetics dictated a strong separation between archival footage and an imagined scene. In such events where recreated scenes were used, it was once (and perhaps still is in certain cases) an ethical standard to use the subtitle ‘based on real events’. Conversely, CGI aesthetics seek to close the gaps between animation and an audience’s perception of their known world. Andrew Darley (2000: 82) comments upon CGI’s contentious relationship with traditional notions of perceptual truth with his concept of ‘second order realism’. Darley states: Attempts – such as those focused upon here [Toy Story] – to imitate and simulate, are at the farthest remove from traditional notions of representation. They displace and demote questions of reference and meaning (or signification) substituting instead a preoccupation with means and the image (the signifier itself) as a site or object of fascination: a kind of collapsing of aesthetic concerns into the search for a solution to a technical problem. (p. 82)
As TSNGO falls away from ‘traditional notions of representation’, I believe Jackson traded the importance of the WWI soldier’s story for the preoccupation with the perceptual realism of his doctored footage. Through my analysis of TSNGO as animation, I wish to reconsider animation’s link to reality in the unique way that Weta Digital’s ‘black box’ technology has allowed us. Before I can address this line of inquiry, however, a much simpler question must be answered.
Is it animation?
To this question, I would like to defer to Stephen Prince (2012: 102) when he claimed the following in his text Digital Visual Effects in Cinema: ‘indeed all digital performances are to some degree the work of animation.’ Prince delineates animation into two camps: characters completely realized through animation, as in WALL-E (Stanton et al., 2008), and characters that are the result of an animator’s collaboration with live actors, as in Gollum from Jackson’s Lord of the Rings (2001). In the second case, a live actor provides necessary motion cues via the motion-capture process, which are the skeleton upon which the animator builds a completed character. Jackson’s Weta Digital was engaged in a similar practice of reconstructing characters from visual cues provided by the traces of skeletons and flesh in the IWM’s archival film collection.
A 2019 interview given at the University of California by Eliot Traver, TSNGO’s assistant editor, alongside Peter Jackson’s (2018) interview at the London Film Festival, have provided several intriguing insights about their take on visual truth in documentary storytelling. They both reminisced about the joy they had in gathering as much World War I memorabilia as they could, from uniforms, patches, buttons, and boots to tanks and live ammunition. Jackson encouraged the team to use the physical archive to inform artists, who would draw individual frames, creating ‘key images’ that would inform the build outward. Here Travers (2019) speaks about the process involved in redrawing the moments we see in the film: A poor roto-artist would draw around every single tiny little thing that we want to create massive shapes with. We use, to a degree, some proprietary software to anticipate where the motion and the velocity of these shapes are going to go as things move across the screen. In a lot of cases, in order to get them to stick, manual intervention is required to keep the matte where they need to be. Then those get used to house the color we use.
Travers supplies some rather fascinating evidence that links his work to that of the animator. Here ‘a poor roto-artist’ draws all the shapes (though the more accurate description may be ‘traces and guesses what’s missing’) that form a soldier’s details. His buttons, patches, boot eyelets and laces are redrawn into a matte that creates an outline of everything on one single frame. This is fed through ‘proprietary software’ that anticipates the motion (direction on an X, Y and Z axis) and velocity of that matte. Finally, other artists manually position the matte. Though the term ‘proprietary software’ beckons as a mysterious black box, the animation of these matte shapes is a collaboration between a machine algorithm and human aesthetic. In other words, the process contains all the hallmarks of the foundational relationship between human and machine that is characteristic of computer-generated animation.
A British Film Institute (BFI) interview with Jackson at the 2018 London Film Festival did, however, reveal an intriguing line he draws between his work and CG animation. An interviewer asks ‘What does the future hold? One day are we going to be interacting with these people?’ Jackson responds, ‘Once you start getting into that you start getting into CGI and animation and you’re not using the real guys anymore’ (2018b). Clearly he does not see what he is doing in TSNGO as a form of animation or CGI, despite the fact that a majority of his footage is redrawn with the assistance of a computer. The dilemma seems to be an inversion of the quest for ‘realness’ in animation without the human interference, which has been explored by Yacov Freedman (2012) in relation to motion capture technology, and in the context of digital forms of ‘drawing’ by Birgitta Hosea (2010). In both scholarly examples, humanity encroaching upon animation and animation sneaking into documentaries, a dilemma occurs when an audience is incapable of distinguishing between the two. Though I would argue the ethical minefield of the former has far more political and social implications beyond the simple deception of an audience, Jackson’s statement ‘you’re not using real guys anymore’ raises the question, ‘when does the moving image of a man captured a hundred years ago stop becoming a “real guy?” And was it ever a “real guy?”’
Deleuze and Peirce on the real
Gilles Deleuze (1989) tackled this problem of the ‘real’ during his discussion of André Bazin’s article on Italian neo-realism. According to Deleuze, Bazin believed the following to be true in cinema: The real was no longer represented or reproduced but ‘aimed at.’ Instead of representing an already deciphered real, neo-realism always aimed at an already ambiguous, to be deciphered; real . . . neo-realism therefore invented a new type of image, which Bazin suggested calling ‘fact image’. (p. 1)
According to Deleuze, the new definition of the real as represented in cinema was not ultimately defined by this ‘additional reality’ created by the arrangement of moving pictures; rather it was the result of the ‘mental’ work done to assign meaning to the new arrangement of movement images, perceptions, actions, and affects. Though CGI is a technical intervention where Italian neo-realism was a social one, this concept of an ‘additional reality’ is curiously similar to Darley’s definition of ‘second order realism’, and therefore quite useful to the discussion of TSNGO’s identity as animated.
In Cinema 2: The Time-Image (1989), Deleuze was referring to the trend of abandoning the montage for a shot sequence, but this definition of the ‘real’ as a series of ‘fact images’ that were to be aimed at reality suits us quite nicely when discussing reality as the intersection of technology and contemporary aesthetics. Today, represented reality appears to cinema goers as frames moving between 24–30 fps on a screen with a 16:9 aspect ratio. Nothing about these parameters are intrinsically ‘real’. Rather, they approximate that which we perceive as real in our own sensory-typical experience. Deleuze defined the normal experience of movement in film as that which approximates our personal sensory experiences of movement, therefore, he stated, ‘a movement that avoids centering in whatever way, is as such abnormal, aberrant’ (p. 36). To answer our previous question: the man on the screen stops being ‘real’ when the technology capturing him falls outside our culturally accepted parameters of the real that happens to be in fashion today. Thus a 10-fps image with a 4:3 aspect ratio may have looked authentic in 1917, a period where these specifications were standard for news media, but it would not look ‘real’ today until it conformed to our new aesthetic standard for realism. There is another interpretation of Jackson’s ‘real guy’ statement. Perhaps he was referring to the ‘real guys’ as the editors who are adding the realistic touch-ups to the original footage. If this is the case, Jackson would seem to value human contributions to the editing process over CGI. We do know that Travers’ ‘proprietary software’ is doing more than storing the imagery an editing artist creates. It is generating imagery based upon the way it ‘anticipates’ motion.
Deconstructing the way Jackson distinguishes his work apart from the signifiers of ‘animation’ and ‘CGI’ may hold valuable insight about one filmmaker’s line between historic truth and an artistic overstep, between ‘real guys’ and cartoons. For Jackson, sensory truth lies in the material object. He collected authentic uniforms from the era, and fired 18-pounder guns with its original ammunition, using his personal .303 rifle to record the ‘authentic’ sound for the film (Hurwitz, 2019: 34). Foley artists recorded people walking in period hobnail boots to replicate the sound of soldiers walking. Jackson also consulted linguists and accent experts while using software to detect the phonemes being created from soldiers’ lip formations to correctly identify not only what a soldier was saying on the silent film, but what accent he was using. Documentary filmmaking practiced by Jackson is a process steeped in reverence for trace. Material evidence is the foundation for the entire TSNGO project. Its original idea was conceived by a museum, a dealer in material artifacts, after all. Animation, on the other hand, is popularly known for making large artistic leaps that foreground authorial expressivity over object accuracy. Though the art and technicalities of animation have changed significantly over the decades, popular association with animation as cartoons and spectacle do not provide much assurance in the business of non-fiction.
The legacy of Prince’s perceptual realism
Long before Prince would release his authoritative text on Digital Visual Effects in Cinema (2012), his interest in computer generated animation was piqued when Forrest Gump (dir. Zemeckis, 1994) and Jurassic Park (M Crichton and D Koepp, 1993) were offering digitally mediated images that challenged our notions of cinematic realism. Prince’s (1996) article, ‘True lies: Perceptual realism, digital images, and film theory’ documents the watershed moment in film theory where existing technologies forced the recognition that we could now see ‘unreal’ images (from dinosaurs in a frame with humans to Forrest Gump shaking hands with John F Kennedy) that are understood as perceptually realistic. Prince would comment in his article, ‘unreal images have never before seemed so real’ (p. 35). Reflecting upon his work, Prince (2010: 20) would define the term ‘perceptual realism’ in the following manner: I identified a digital basis for realism in cinema in terms of what I called ‘perceptual realism,’ which was the replication via digital means of context clues designating a three-dimensional world. These cues include information sources about the size and positioning of objects in space, their texturing and apparent density of detail, the behavior of light as it interacts with the physical world, principles of motion and anatomy, and the physics involved in dynamic systems such as water, clouds, and fire.
Prince pointed out that the CGI digital imaging and motion blurring allowed Gump to appear as a credible inhabitant of his filmic environment, ensuring that the present-day actor (Hanks) would appear as if he was an inhabitant in the world of the past footage. As we consider how important it was that CGI aligned with the known media world for Robert Zemeckis, recall the emphasis Jackson had put on collecting all the WW1 memorabilia he could get his hands on. Every khaki, button, patch, and shoelace had to match the one he saw in his present reality. For Jackson, material objects unlock historical truth. For Zemeckis, it is the visual qualities of historic media and ability to simulate photography itself that unlocks time. The only real ‘tell’ as to Forrest Gump’s digital identity occurs in scenes where characters are made to say words absent from the original recorded footage. In one scene where Gump appears on the Dick Cavett show, Forrest Gump unwittingly inspires the lyrics for ‘Imagine’ with John Lennon by recounting his trip to China. Yet, in conversation with Gump, Cavett and Lennon’s mouths twist around their phonemes in ways that are mechanical and a bit off from the way humans form sounds. Technical limitations aside, Forrest Gump (dir. R Zemeckis, 1994) was responsible for blurring the distinction between animation and live action cinema for decades to come.
Prince inspired a new generation of scholars to look deeper into the concept of perceptual realism as CGI began to depict worlds that most audiences would never have experienced personally, though would feel real in the way they match pre-CGI cinematic composition and human movement. Lisa Purse (2018) has commented upon CGI’s longstanding tradition of creating seamless composited images which seem to carry a sense of ‘depthlessness’ because of their undifferentiated digital surfaces. If CGI animators were to create believable footage from a previous era it seems they would want to embrace all the textural qualities intrinsic with the analogue film of the day. Audiences, however, became used to the ‘scrubbed, hygienic, disinfected images’ appearance of CG objects and Purse recognized digital media’s threat to ‘overwrite’ lived experience. She states, ‘Yet the “material gravity” of human existence, agency, and self-expression is, for those who can afford it (Schradie, 2011), now thoroughly situated within, and lived through the screens and surfaces of an expanding media ecology’ (Purse, 2018: 153). Lisa Bode (2018) similarly notices that, as audiences are exposed to more perceptually real CGI imagery, they become less inclined to believe anything they see. She states, ‘A broader viewing culture of incredulity has reemerged due to a combination of production publicity, cult viewing of bad cinema, online forums, editorial photoshopping, and image hoaxes’ (p. 1). Quite similar to the early cinema audiences of the 1910s and 1920s, contemporary audiences, saturated with ‘fake’ media images, will habitually disbelieve any spectacular image at first glance.
Zachary Samuel Gottesman (2018) also looks to past animation practices to explain sociological and materialist phenomena in today’s CGI audiences. He states: ‘the melding of the cinematic, animatic, and digital through CGI forces us to think about a technology that has been largely forgotten in the history of animation: the rotoscope’ (p. 193). Gottesman uses the Japanese anime Aku no Hana and its entirely rotoscoped cast from real human models to analyze the repression of the Japanese working class. He states, ‘I trace the rotoscopic uncanny through the history of capitalism and posit that the return of the repressed and the non-Being of hauntology comes out of the very nature of neo-liberal capitalism’ (p. 193). It should not escape the reader’s attention that the traces of real soldiers on the IWM footage were drawn over with likenesses we are told are more like the original than the archived footage. Though I would not go so far as to posit TSGNO as another example of neo-liberal capitalism’s repression of the common man, we are being sold on a new perceptual realism haunted by the traces within its original media.
Reactions to They Shall Not Grow Old: Recapturing humanity and stereotypes
Mark Kermode (2018) said in a Guardian review of TSNGO, ‘Jackson has done something quite remarkable: using twenty-first-century technology to put the humanity back into old movie stock. The result is utterly breathtaking.’ To refer to the sum of an increased frame rate, recoloring and rematting as putting ‘the humanity back into old movie stock’ is perhaps just as breathtaking of a philosophical jump as the technological leap that went into repackaging these images. Kermode’s thought that humanity is present somewhere in the celluloid waiting for technology to find it once again is loaded with all sorts of fascinating questions about the intersections of humanity and the archive. But what lies at the heart of the discussion is a cultural understanding that humanity can be projected in film, and the more deftly the technology is used, the more it speaks to our perceptual sense reality.
A brief glance at Kermode’s article can reveal word choices attributing Jackson’s alterations as a restoration of truth and humanity. Kermode notes the film’s ‘impressive natural hues’, and its ‘interstitial frames that recapture the rhythms of life’, all working together to bring ‘unexpected life and laughter’. Kermode also suggests the film is revealing some aesthetic truth that somehow lay dormant in aging footage when he suggests that ‘the technology has somehow pierced the surface of the film, causing (virtual?) memories to come pouring out . . . How profoundly truthful it all felt.’ (Note: the insertion of the term ‘virtual’ was Kermode’s, not mine. It appears that Kermode is also anticipating the ability for this technology to manufacture a new reality or Deleuze’s ‘additional reality’ with this archival footage, as well as the ambivalence around Jackson’s film as partly animated.)
Jackson has made several public statements admitting to trying to endear himself to a common audience through the ethos evoked from a soldier sharing his personal experiences, while simultaneously distancing himself from the perceived pedanticism of historians, stating ‘I want it to be less about the history of war and more about the personal experience.’ Another BFI interview captures Jackson stating, ‘Historians can certainly go through the archives, which these guys [the IWM] actually had and had certainly pieced together the commands version, the political version of the war. This film is actually what the soldiers experienced’ (Jackson, 2018b). Jackson has never attempted to discredit historical viewpoints; actually he holds their accounts with great reverence, but he claims to be doing something that has yet to be done. Jackson seeks to allow the audience to live the life of a soldier, to know only what the soldier knew, see only what the soldier saw and lived within the very small bubble of understanding a soldier had of the socio-political moment in which he found himself. 1 Considering Jackson’s mission for the film, accuracy of color, lighting, textures and movement from the soldiers’ perspective are paramount to the storytelling process.
James Taub (2020) made the following observation in his review of TSNGO: Many First World War documentaries focus on poetry and the doom and gloom of the trenches, but They Shall Not Grow Old breaks from that mode. By relying on the veterans themselves for all of the narration, everyday aspects of training and life at the front become quite humorous. A soldier’s life then, as well as today, is not always terror and strife. In the end, they were still a group of young men going about what young men throughout history have done. The comical nature of many of their experiences enlisting, in training, and in everyday army life makes the viewing experience even more of a connection to the past. It feels as though the viewer is sitting down with the veterans themselves as they tell stories of their youth.
Historians agree with film critics in building a consensus that the footage is certainly enticing and lively; however, that criticism is double-edged. Historians voice concern when a Eurocentric voice is full of empathy-inducing spectacle because it only helps to bolster the dominant take on WWI, one that is fraught with hegemonic misdeeds. Susan Grayzel (2019: 1784) comments in a roundtable issue of the American Historical Review (AHR), ‘The film is certainly a technological marvel, and the attention to the process of capturing accurate colors and sounds is remarkable, but it does not help us understand the story of the war beyond the familiar stereotypes.’ Apart from the contention on the film’s value, popular critics and history academics agree that the technology has brought archival subjects to life more than ever before, but where does this illusion come from? It cannot be present in the still images of each frame. Kermode and Grayzel have surely seen colorized still pictures from the period and had those evoked the same feeling, it seems likely they would have mentioned it. So even though our critics mention the life bringing color, lighting and textures, the real illusion must occur when all these features are put into motion. The final section of this analysis will focus on the techniques Jackson has used to move TSNGO one step closer to achieving the holy grail of computer animation: photographic quality realism put into motion.
Elements of classic animation in They Shall Not Grow Old
Shadow and light
When Jackson’s technique works well, it creates a connection with the audience that may have the power to change collective memory. Santanu Das (2019: 1772) remarks that ‘Jackson exquisitely blurs the boundaries between the original and the digital, the bodily and the prosthetic, between memory and perception.’ Some of the most striking moments are the closeups of soldiers, smiling at the camera (us) with missing and black teeth. They turn their head to profile and the peach of their cheeks brightens as they turn to direct sunlight. Another soldier wearing a ‘Brodie’ helmet tips his head down, casting a shadow over his face. Were this footage black-and-white, the face would surely be obscured, but Jackson has given a broader range of gradients sliding down the soldier’s face. Donald Crafton (2013: 183) notes a similar quality embedded in animation from the 1930s in his book Shadow of a Mouse: Performance, Belief, and World-Making in Animation, noting that ‘There is no better way to chart how mechanics, form, and content aligned to reconceptualize performance space in the 1930s than by observing the treatment of shadows . . . they are essential for creating the illusion of depth.’ Shadow is instrumental in unifying the foreground and background, but seeing texture within those shadows used to only be the territory of live action film. It was not until the birth of CGI that animated character features could efficiently present themselves from behind shadows. Jackson is surely keen to the affect created within these moments, evident by his playing closeups at three-quarter speed to ensure we catch it too.
Optics and haptics (the seen and felt)
Panning and zooming on still pictures was first used in documentary filmmaking by the National Film Board of Canada’s City of Gold (1957). Ken Burns would later use the effect prodigiously in his documentary work and give credit to City of Gold for the motion effect in an interview for Weber Studies (Vause, 2006). The ‘Ken Burns Effect’, coined by Steve Jobs in Apple’s (2003) update of iMovie, creates motion (i.e. life) where the original archive provided none (Balakrishnan, 2016). Das (2019: 1778) also highlights another kind of effect-pairing in the film, noting that ‘Jackson’s twin art of zooming in and colorization thrusts us into subcutaneous haptics – we wince, we tighten.’ Crafton (2013: 183) similarly writes about how Walt Disney would bring in Don Graham, a professor of the Chouinard Art Institute (later Cal Arts), to speak to his animators often about the necessity of a tactile (haptic) understanding of their environment. Graham would stress how the illusion of motion would begin with the drawing of the environment itself. Real knowledge of the environment was a dual understanding of these haptics and optics, the felt and seen. Jackson’s reclamation of the IWM’s footage is able to provide us with a sense of both the physical and textural environment. Bodies in the foreground are pulled out of the background through use of shadow and highlights. As a camera pans, the dimensionality of a space becomes apparent as shapes will shift in ways consistent with 3D space. Close shots reveal the texture of clothing, landscape, and faces. One gets the feeling that running a hand over a uniform’s fabric, the barbed wire of No Man’s Land, the trampled grass of a field or the bloodied bandage of a field dressing would produce entirely different tactile sensations. Though this same footage was cleaned chemically prior to Weta Digital’s treatment, the 10-fps, starkly contrasted black-and-white film of the archive lacked the depth of composition we can see in TSNGO. This layering of textural features may, however, not have even been present when the footage was newly developed. Inserting these textures took the painstaking concentration of a frame-by-frame animator, and it is this addition to the original footage that should be called into question. Altering an archive, then suggesting the footage is a truer version than the original, certainly blurs the boundaries between cultural memory and individual perception.
Movement
When it comes to the illusion of realistic motion for a contemporary audience, nothing is as important to the animator and cinematographer as frame rate. Jackson (2018d) himself spoke about the process of changing the frame rate from the archival footage to a standard 24 fps. The first film cameras relied on an operator to turn a crank which would determine the rate each frame was exposed to light. Jackson suggested that war photographers were particularly inclined to turn the crank faster in moments of excitement and danger. This human element would result in footage that would vary in frame rates from 10–18 per second. Jackson admits here that the process of standardizing to 24 fps was more about feel than any technical process, ‘What you do is take a stab at it. You say, ok this is about 14 frames a seconds, so you slow it down to get 24. It’s just trial and error shot by shot.’
Nicholas Hiley (1993[1915]) republished an insight into the training for cameramen in 1915 with Pathé’s ‘Hints to Newsfilm Cameramen’. One Pathé hint read, ‘Where speed is the dominant factor turn slowly, about half speed. Failure to do this is unpardonable. Remember then that you must shut down your iris diaphragm so as not to over-expose.’ The reasoning for slowing down the exposure rate does seem sound. A camera operator would catch the blur of a fast-moving object, such as an explosion or a falling body, if they were to slow down their cranking. Though Jackson was correct that his frame rate speeds did vary, his analysis that this was because of the operator’s biomechanics is purely speculative. Here we have an example of where solid training for war journalism would have produced the same affect Jackson attributes to operator error.
The effectiveness of resetting the frame rate is most noticeable in places where Weta Digital could not entirely fix the footage, during the moments where the speed of an action was too quick for the camera to catch. Smoke, in particular, creates an interesting effect when made to conform from 10 fps to 24 fps, as it is constantly evolving in its billows by the millisecond. To fix the frame rate, Jackson’s team would speed up or slow down the rate, then add duplicate frames in order to fill out the required rate. A frame of duplicated smoke does not take into account its naturally dynamic form, the result is a ghostly apparition. Whenever a man puffs on a cigarette, or an artillery shell explodes, the moment feels out of sync with the rest of the filmic universe. The aberrant behavior of smoke on reconditioned film draws attention to its own artificiality. Another point where the action outpaces the frame rate is when a soldier plays with a frog. One moment the soldier lifts the frog, plays with it and then he drops it accidentally. Had it been filmed in 24 fps we would have watched it fall to the ground over a series of a few frames. Jackson’s film, however, shows the frog doing a disappearing act, vanishing from the man’s hands, then reappearing on the ground.
One final moment that serves as evidence to the spell Jackson’s technology casts on the IWM footage occurs when a man trips into a ditch as he marches along the road. His body quickly becomes digitized, momentarily broken into bytes of data as he falls down the embankment, his body moving quicker than the frame rate can catch. Just as quickly, he is reassembled as he works his way out of the ditch. The bit-formed frame looks alien to the filmic universe as he struggles to regain his balance. A moment later he is recomposed and seems at place again in the environment. This moment, for me, was nothing short of a Brechtian Verfremdungseffekt, distancing me from these men. Nothing in my research suggests Jackson purposely wants to jar his audience with the discontinuity resulting from a pairing of technologies a century apart. The object lesson here, however unintentional, reminds me that I am watching footage Jackson purposely manipulated to make us experience the war from the soldiers’ perspective. Bertolt Brecht would have called this distancing effect a moment of clarity where the bargain we have made, the exchange of rationality for sympathy (thinking for feeling), is laid bare.
These distancing moments are also a clear sign of where an audience can see the metaphorical line Jackson was drawing when it came to reanimating his subjects. He would recolor, retexture, relight, and change the frame rate but then pull back from redrawing frames that were missing necessary information. If anyone were to argue that Jackson’s film is not animation, this is where they would lay the cornerstone of their argument. In moments where Jackson could have rotoscoped an object, reshaped it and inserted it into the next frame, he did not. It is a curious bridge not to cross when so many other archival traces are altered, the most conspicuous addition being sound.
One final note: Sound
TSNGO’s foley work was possibly one of the processes that was most authentic to the spirit of the project that valued trace (material objects) over deixis (interpreted context). Where the visual imagery was the product of a human–computer collaboration, the foley sound effects often came from original materials used during the war. The entire film is narrated by actual WWI soldiers, included as part of the IWM’s audio collection. The other collection that added to TSNGO’s authenticity was Jackson’s personal collection of WWI artifacts, including weapons, personal gear, uniforms, and vehicles. The TSNGO sound designer, Brett Burge, stated, ‘He [Jackson] has a collection of props and historical items that he was happy for us to record, for just that reason, which is usually not the case. Here, it was, “No, no, that’s what they’re there for”’ (Hurwitz, 2019: 32). The sound teams were responsible for all aural ambience (boots marching on dirt roads, clanking kits, mooing cows, etc.) as well as the specific shot-by-shot momentary effects. This would include mine explosions, rolling tanks, gunfire and recreated dialogue based on research from certified lip readers. The work was substantial, considering there was no sound from any of the original footage. All of it had to be inferred based upon what was occurring in the video footage.
Perhaps this soundscape, even more than the proprietary visual software, was the largest gap filling Jackson demanded. Jessica Meyer (2019) comments in the AHR Roundtable that WWI was metaphorically known as a ‘silent’ war because of the fact that the men who returned home would suffer their shell-shocked memories in silence, yet Jackson’s film rebels against that metaphor. Meyer states: It remains, then, for the final element of sound in the film, the underlying track of noise rather than speech, to demonstrate how this form of technical enhancement of film records can help to challenge the myth of the silence of the First World War. By including a complex noise-scape, the soundtrack makes audible not only the dominant artillery fire recorded and remembered by so many servicemen but also the full array of noises created by non-human actors and inanimate participants – the jingle of harness and the creak of gun carriages, horses snorting and birds singing, the crash of tiles slipping from a roof minutes after the shock of a shell explosion. (p. 1791)
What is most striking is the transparency exercised by Jackson and his team when it comes to their sound work. They are more than generous in giving us exact archival models and methods for recording. Unlike the mysterious black boxes that somehow redrew frames, we have plenty of documentation for how effects were recorded and mixed. Noting how comprehensive the reporting on sound from this film was, I am left with even more questions about the visual. How is manually drawing a shape across multiple frames any more of a historical inconsistency than using the modern recording of WWI era canons and munitions? How is allowing modern forensic lip readers to detect dialogue in silent film any less of an anachronism than drawing a CGI animation over a few frames? The answer is, for me, less compelling than the lines Jackson draws between the sacred and profane. Why is the visual more sacrosanct to Jackson than the aural and what does this distinction imply for future historical reanimations?
Conclusion
Jackson recently sold Weta Digital to the Unity game engine, which was framed by the press as a move toward creating assets in Facebook’s highly publicized (with no definitive examples) Metaverse for £1.2 billion. (Silver, 2021). This move came upon the heels of the creation of Weta Digital’s animation division in the summer of 2020. Jackson stated the following about his new division,
We are huge fans of animated storytelling in all its forms, but it can be a long, protracted, and often costly way to make movies. That’s, in part, why we have created this company – to change the model and open the doors to filmmakers and storytellers who might not otherwise be given the chance to show what they can do. (Giardina, 2020)
Weta Digital already received 6 Academy awards for their work in visual effects and 12 Academy ‘Science and Technology’ awards. They had seemed primed to continue their cinematic success before the company was sold.
CGI animation has been proven to be a brilliant tool for fiction. Jackson is right for being cautious with the signifiers he uses to define his version of history. Since ‘animation’ and ‘CGI’ connote optical trickery, Jackson’s desire to separate his work from such practices – even if in name only – is understandable, if not justifiable. When Jackson colored, reedited, and recomposited World War 1 footage for his documentary, he was reanimating footage from the Great War which suited aesthetic sensibilities of a 21st-century audience, yet when this is done without a certain level of transparency an audience cannot distinguish where an historical trace ends and the artist’s vision takes over. Reanimating historical footage as a documentary should be under far more scrutiny as a contemporary audience’s ability to recognize a ruse is rapidly outpaced by technology’s ability to hide its work. The director diving into digital animation of archival footage must acknowledge and own their reimagining of history. Broad proclamations that this is ‘every soldier’s story’ has political implications that strongly suggest historical positivism. Jackson’s erasure of ‘animation’ from his descriptive vocabulary rhetorically sidesteps this important conversation, and lulls his audience into talk about ‘restoring the humanity’ into the old footage and making the soldiers ‘come to life again’. We would certainly be indebted to the scholar who discovers how such ‘restoration’ has the potential to make audiences susceptible to nationalistic propaganda by aligning archival footage to modern aesthetics of perceptual realism.
I am also a big proponent of material artifacts. I do believe in the hauntings of objects imbued with the beautiful and the horrible narratives which accompany them. No object is a history unto itself but must be accompanied by a story which augments its trace. I would agree with Ehrlich (2020: 260) when she states: ‘The merging of the physical and virtual realities leads to an augmented view and experience of reality which exceeds the merely physical.’ Jackson also believes in hauntings, though he imbues his company, not the material object and its accompanying narrative, with the power to make soldiers ‘live again’. Every reformatted frame, paired with meticulously recorded sound effects and a soldier’s story which seems to parallel the action of the film, certainly brings something to life but, I will contend, it is not the individual soldier, but nationalism and all its familiar trappings and tropes. In fairness, Jackson has also revived the common theme about the necessary sacrifice certain young men made to maintain a quality of life for their indebted audience. Putting aside the fact that the war on the Western Front involved far more individuals than the 120 white men in the IWM’s interview archive, Jackson is admittedly looking to turn 100 men’s stories into a universal story about the sacrifices made for the British way of life. Reflection upon Jackson’s words at the completion of his film severely tests whether he ever discovered his grandfather’s individual experience at all. If he did discover William Jackson’s story, how did his audience fare in the same prospect?
After all, if Jackson’s goal was to imagine what his grandfather experienced during the war, did he not merely animate it in a way that he and his contemporary audience would find perceptually real? Jackson came much closer to understanding his grandfather’s experience, and has shown how he values material objects from the period, believing that feeling the uniforms, running the tanks, and shooting the guns would move him closer to knowing the ‘soldier’s’ story firsthand. However, I have never seen him place as much reverence on the cameras and footage produced during that same period. Jackson’s audience, on the other hand, did not engage in the process of feeling the physical traces of the Great War. Most of us have not taken the time to understand how the ‘technological fuzz’ of war media has changed over the last 100 years and how its status as ‘fact images’ has changed with the popular aesthetic. Nor did we, his audience, make the crucial editing choices required to reanimate the IWM’s archival footage. We, therefore, are no closer to understanding William Jackson’s experience, though we may see more of ourselves in it.
Footnotes
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors, and there is no conflict of interest.
