1. Introduction: Video Games, the Moving Image, and The Last of Us

Video games and the “moving image”—that is, film, television, and other video formats, as per Noël Carroll—are parties to an intimate relationship. As implied in the name of the medium, video games too consist of moving images on a screen, and, throughout their history, have extensively crossed paths with the language and culture of film and television. By the 1980s, at a time when the existing technology primarily bound games to two-dimensional graphics and gameplay, the medium was already borrowing ideas and visuals from films and television series (see Wolf, Elkington). Soon, film and television companies also became directly involved in video game development; after all, as captured in a 1997 essay for Film Quarterly by influential video game scholar Mark J. P. Wolf, “video games compete for audiences at the very same sites as film and TV” (11). Perhaps the most illustrative example of this phenomenon is the foundation of Lucasfilm Games, originating from Lucasfilm in 1982.1 The company would go on to develop not only adaptations of popular film franchises owned by Lucasfilm (namely, Star Wars and Indiana Jones) but also original titles such as Maniac Mansion (1987) and The Secret of Monkey Island (1990), which are rightfully credited for pioneering the integration of elements of cinematic language and aesthetics in video games.2

The technological transition of the medium to three-dimensional polygon graphics in the mid-1990s only added to the existing intersections. Video game characters began to walk three-dimensional environments that possessed a naturalistic depth. The virtual camera could then mimic the behavior of a real camera, through the adoption of angles and movements that naturally drew inspiration from the language of film and television.3 The so-called cutscenes, video game sections that convey information through non-interactive sequences of images, grew longer, more frequent and complex, and began to resemble movie scenes. Eventually, interactive and non-interactive sections became visually indistinguishable: cinematic cutscenes would occasionally prompt the player to quickly input specific commands so that the remaining sequence of non-interactive images could proceed in the desired direction (the so-called quick-time events), while the ordinary interactions of gameplay were carried out within interactive environments possessing a level of detail that earlier, less powerful generations of computers were only capable of displaying in pre-rendered video sequences.4

The identity of video games is fundamentally contentious. Players, developers, and scholars alike disagree as to whether video games are—to put it bluntly—games or interactive stories.5 I will not delve into the many issues of these competing approaches; what matters here is that, when the story elements of video games come into focus, the term of comparison most commonly invoked is that of the moving image. A particular video game may be praised by some for its achievements in so-called cinematic storytelling and belittled by others for being more of a movie than an actual game (whatever that means).

Among the video games credited for establishing new benchmarks in cinematic storytelling is The Last of Us (see, for example, Park, Cruz), a 2013 video game developed by the American studio Naughty Dog and originally published by Sony Computer Entertainment for the video game console PlayStation 3.6 The Last of Us did indeed contribute to innovating the language of video games. In the late 2000s to early 2010s, the cinematic potential of video games was heightened by the technological advancements of new generations of hardware. Amid the pervasive influence of such titles as Call of Duty 4: Modern Warfare (2007) and Uncharted 2: Among Thieves (2009),7 so-called cinematic video games, however, most often consisted of borrowings from what was, in fact, the specific language and themes of action and adventure films, rather than film language more generally.8 In 2013, therefore, The Last of Us hit shelves with the force of a revelation: its cinematic character was not in the extent of its borrowings from selected film genres, but in its unprecedented emphasis on actors’ performances and cinematography, as well as its seamless blending of interactive and non-interactive sequences, all of which function to serve a psychologically nuanced and emotionally compelling narrative.

It is not an easy task to summarize the content of The Last of Us without risking a misrepresentation of its artistic reach. In brief, the game takes place in a vision of the contemporary United States where the population has been decimated by a fungal disease that turns the infected into zombie-like creatures. Twenty years after the initial outbreak, Joel, a working-class white man in his forties who lost his daughter in the early days of the pandemic, and Ellie, a 14-year-old girl who happens to be immune to the disease, travel across a devastated country in search of the Fireflies, an organized group of survivors that might be able to develop a cure for the disease on the basis of Ellie’s immunity.

Commonly held as an unrivalled paragon of story-driven, cinematic video games, The Last of Us was long expected to originate some sort of live-action adaptation (see Gaughan), which eventually came about in the form of a television series created by Craig Mazin, a film and television writer, and Neil Druckmann, the writer and co-director of the 2013 video game. Bearing the same name as the original game, the series is divided into nine episodes of uneven length, which aired on the American television network HBO from January 15 to March 12, 2023. The series has been renewed for a second season (see Hibberd), which is expected to cover part of the events of the video game The Last of Us Part II (Naughty Dog, Sony Interactive Entertainment, 2020).

Since the 1980s, an increasing number of video games have been adapted for cinema and television (see Picard, Knott). The characteristics of this phenomenon go somewhat hand in hand with the progressive affirmation of video games as respectable products of human culture. Thus, the child-oriented animated adaptations of the 1980s and early 1990s were followed by repeated attempts to establish some of the most popular video game franchises on the big screen—including Mortal Kombat, Tomb Raider, and Resident Evil, among others. Critics and viewers concur that these attempts have most often resulted in mediocre and less than mediocre products (see Schulzke 70, Stopel). There are indeed known difficulties in translating content from the language of one medium to the language of another.9 The bar has ostensibly been raised in recent years by adaptations such as Castlevania (2017–2021), Arcane (2021–2024), Cyberpunk: Edgerunners (2022), and The Last of Us, which have been generally met with praise by both the audience and specialized press.10 Given its cinematic character, The Last of Us must have, and has indeed, appeared as an ideal candidate for a successful translation; all that had to be done was to chisel away, so to speak, the game’s interactive component.

It turns out, however, that things are not so simple: even story-driven, cinematic video games like The Last of Us cannot be stripped of their interactive element without losing something fundamental about their story and themes. In video games, meaning is constructed through the combination of a plurality of different means, including, but not limited to, visual elements, textual elements, and elements of gameplay—that is, what the player actively does in the game, or, from another perspective, the actualization of the game’s rules through the player’s interaction.11 The Last of Us is no exception. By way of example, the change of the player-controlled character approximately three quarters into a typical playthrough—arguably one of the most defining moments of The Last of Us—cannot be reduced to the matter of its resulting visuals. In this as well as in other cases, much of the meaning is constructed through the language of gameplay.

In the present article, therefore, I seek to elucidate the relationship of the 2023 adaptation of The Last of Us with the distinctive language of its source material. Following an examination of key elements and episodes of the two works, I argue that the excision of the game’s interactive component for the purposes of its TV adaptation inevitably leaves semantic voids and narrative gaps, which are not adequately filled by means of the language of the moving image. This issue affects the series as a whole and is not limited to a few instances of oversight. Large swathes of text and, therefore, meaning are lost.12 Due to an adaptation process that systematically overlooks fundamental areas of meaning of the source material, the individual parts thus obtained cannot stand on their own but struggle in their quest for meaning and artistry.

2. Misreading Gameplay

That something may have been lost in the adaptation can already be seen in the first episode of the series. The original video game begins with a prologue set on the decisive day of the pandemic outbreak, featuring Joel, his 12-year-old daughter, Sarah, and Joel’s brother, Tommy. Its plot is quite simple, focusing on Joel’s relationship with Sarah; the spread of the pandemic to the outskirts of Austin; Joel, Sarah, and Tommy’s attempted escape; and the untimely death of Sarah, who is shot by the military. This section of the game, however, goes beyond the factual details of the plot. One of its most defining features is that the player begins the game in the role of 12-year-old Sarah. After a brief cutscene that establishes Sarah’s relationship with her father, she wakes up in the night to an eerie, empty house and starts looking for Joel. As the player-controlled Sarah freely explores the house, the outbreak slowly creeps in by means of alarming headlines in the morning newspaper, the news on television, an explosion in the distance, and, finally, a zombie-like “infected” breaking into the house. The scene continues with Sarah, her father and uncle leaving the neighborhood by car in search of safety. The child remains the player-controlled character throughout the scene. Sitting in the backseat and limited by the confined space of the car interior, the player can only look around at the surrounding havoc; Sarah responds to the player’s input by gliding from one side of the car to the other, while the dialogue indicates that she and her family are attempting to make sense of what is happening. When an accident brings the drive to an end, the player is given control of Joel. The reason becomes immediately clear: Sarah was injured in the crash and cannot proceed by foot—she must be carried by her father. Holding the child in his arms, the player-controlled Joel can no longer wield his handgun, which is given to Tommy; Joel can only run and soon ends up separated from his brother. The player is then tasked with avoiding the infected and bringing both Joel and Sarah to salvation. Their flight, however, ends shortly thereafter, as they are fired at by a soldier who is seemingly trying to enforce a quarantine zone. The soldier is shot by Tommy in return, but Sarah is dead.

As it might be clear from this brief outline, the plot per se is rather unremarkable. If the prologue manages to excel—and it does—it is thanks to precise language choices, some of which are specific to the language of the video game medium. Throughout the first interactive sequence of the game (that is, after the introductory cutscene), Sarah is not just any fictional character but the character whom the player directly controls and embodies. Both Sarah and the world around her only exist with respect to her perspective and possibilities, which are directly experienced by the player through gameplay; Sarah is what the player does, and the game world is the sum of Sarah’s interactions. Meaning is found not within the abstract story but within this particular structure. Without going into a comprehensive analysis, one of the most relevant constructions of the language of gameplay throughout the prologue is that of both Sarah and Joel as the powerless witnesses to the catastrophe, resulting not only from the specific, limited range of actions that are afforded to the duo but also from their contrast with the intrinsic agency of gameplay—powerlessness in action.13

If one were to film the exact same sequence of events as they appear on screen when playing the game for film or television, the resulting meaning would not be the same. The moving image would only be able to reproduce the visuals of an individual playthrough, whereas the video game text largely consists of interactions (which, moreover, are intrinsically not sequenced): the video game’s embodied experience of Sarah in the backseat of Tommy’s car differs from the resulting visuals of the very same sequence. To only look at the visuals is to miss large swathes of text and, therefore, meaning. Thus, to film these scenes the way they appear in an ideal playthrough would, for example, fail to convey the aforementioned powerlessness of Sarah and Joel: Sarah sitting in the backseat of her uncle’s car and looking at the surrounding havoc would be the rather trivial detail of a larger scene with three characters, were it not for the non-visual text that lies within the player’s interactions (see Figure 1). The visuals of an unarmed Joel running for his and Sarah’s life are a different text from that read by the player through the game’s controls, which, for example, also includes the experience of Joel’s death, were the player to fail to outrun the infected.

If the aim of an adaptation were to construct the same meaning and achieve the same effect of the video game, these scenes could not merely be filmed the way they appear on screen during a typical playthrough, because this is not where their meaning effectively lies. The same general meaning and effect—perfect identity, of course, being a practical impossibility—would have to be conveyed through means that are specific to the language of the moving image; the resulting visuals would not necessarily resemble those of the game. This is, precisely, where the television series for HBO largely fails, or, at the very least, fails to create an adaptation that is as meaningful and artistically outstanding as the original video game. For example, whereas the video game establishes the centrality of the character of Sarah by making her playable, the series circumvents the loss of interactivity by granting her additional screentime. On television, Joel’s house no longer offers a sufficient context for Sarah to stand out; instead, we follow her at school, at her neighbors’, and through town. This, however, does not translate into thematic complexity or psychological depth: evidently, extended screentime is no replacement for meaningful gameplay and no substitute for the conveyance of meaning through cinematography. For that matter, Sarah, whose death provides the male protagonist with background and motivation, remains deprived of agency and wholly determined by the relationship with her father, for whom she cares with motherly attitude.

Whereas the adaptation of the prologue still somewhat manages to hold its ground, the central developments of the story suffer from numerous shortcomings and contradictions due to what can be described as a general disregard for the distinctive language of the source material.

Contradictions, in particular, arise from an apparent attempt to modernize the themes and characters of the original story—which would be a legitimate enterprise, were it to be carried out consistently. Indeed, despite its artistry, the original video game is also a product of its own time, permeated not only by visions of disaster, violence, and trauma that are characteristic of post-9/11 American fiction but also by an envisioning of gender roles that remains fundamentally more conventional than subversive.14 The male protagonist, Joel, a sort of modern-day cowboy, hardened by the events of the pandemic, does not refrain from putting himself above moral and social norms and resorting to extreme violence, including torture, in order to protect Ellie, a sweet-looking young girl. Despite these seemingly unremarkable tropes, the game excels in portraying the development of the father-daughter relationship between Joel and Ellie and the maturation and gradual empowerment of Ellie herself—thanks to, in no little measure, the nuanced, chameleonic performances of voice actors Troy Baker (Joel) and Ashley Johnson (Ellie).

Be that as it may, the television adaptation appears to be guided by the desire to soften the corners of the source material’s depiction of gender roles. The rewriting of the male protagonist, competently portrayed by Pedro Pascal, becomes especially visible in episode six, which goes to great lengths to show that not only is Joel weak, frail, and fearful, but also unafraid to weepingly disclose his emotions before his brother. (Of course, he will still torture and slaughter his fellow human beings.) The rewriting of the female protagonist (portrayed by Bella Ramsey), on the other hand, is both more radical and evident. From her very first appearance in episode one, Ellie’s fierce and rebellious nature is constructed through less than subtle hints. At the moment of her first appearance, Ellie has recently been bitten by an infected, before somehow ending up in the hands of the Fireflies, an organized group of survivors, who have therefore bound her with a heavy metallic chain in anticipation of her showing signs of the disease. The zombie-like infected are powerful and bloodthirsty: the chain is therefore entirely reasonable but nonetheless cinematically contributes to the definition of Ellie’s character. Chained like a werewolf, kicking the food tray against her jailers, uncooperative, swearing, screaming, confrontational, pulling her chain in an attempt to break free—Ellie’s presentation is not very nuanced (see Figure 2). Unfortunately, this portrayal is representative of the character’s overall personality throughout the series, which often underlines her propensity for violence.

Together with these changes, the television adaptation also categorically excises several types of interactions between the two main characters that were chiefly tackled through the language of gameplay in the original video game. In general terms, one of the peculiarities of video games in their current historical incarnation is that they tend to engage the player in actions that are both minute and repetitive; The Last of Us and other games include moments of unique narrative and visuals, but also, and most characteristically, sequences of repetitive simple movements, actions, and interactions.15 Minute as they are, and all the more so by reason of their constant repetition, these actions too are elements of the video game text—that is, they carry meaning. To a large degree, the relationship between Joel and Ellie is constructed precisely through this type of language. That Ellie needs Joel’s companionship and protection is not only established through the game’s dialogues and cinematic sequences but also and especially through countless gameplay interactions. During a typical playthrough, Joel protects Ellie countless times from either the zombie-like infected or competing survivors. To put it bluntly, the player-controlled Joel does the killing and leads the way; failure to perform these tasks prompts gruesome scenes in which not only the player-controlled character, but also the child, are butchered, and the sequence has to be repeated.16 In the adaptation, combat is infrequent (see also Hodges), Joel’s protective role is not equally emphasized, and, at a comparatively earlier stage than in the video game, in episode four, the roles are already inverted, with Ellie saving Joel from getting killed. Another fundamental element of gameplay that is entirely omitted in the series relates to Ellie’s inability to swim. This detail plays a central role not only in the construction of her relationship with Joel, who has to help her traverse every stretch of water, but also in relation to the development of Ellie’s own character. Similarly lost are a number of other gameplay interactions, ranging from additional types of cooperation in navigating the rough environment, to Joel’s care in finding and collecting Ellie’s favorite comic books.

In the video game, the pair’s relationship is also developed through more conventional narrative tools, which similarly contribute to constructing Ellie’s reliance on Joel. These include pivotal moments such as their forced separation in Pittsburgh after Joel falls down an elevator shaft, leaving Ellie behind, alone and terrified. At numerous moments throughout the game, Ellie appears manifestly afraid of remaining alone; likewise, she is always overjoyed to be reunited with Joel after their frequent separations. The most defining of these situations takes place at the hydroelectric power plant in Jackson County, when the pair reunites after an enemy raid. Each of these moments is emphasized by Johnson’s intensely emotional performance, which greatly contributes to the development of Ellie’s attachment to Joel.

In the television adaptation, these relational elements disappear and are not replaced by anything else, causing a flattening effect, so to speak, on the relationship between the two. As such, in episode six of the series, it is unexpected when, as Joel attempts to pass the responsibility for Ellie to Tommy and part ways, she suddenly and vehemently protests against the prospect of being abandoned: “Don’t tell me that I’d be safer with somebody else,” she yells at Joel, “because the truth is, I would just be more scared.” In the original video game, this is an iconic cutscene that is preceded by lengthy development: the player does not need to be told by Ellie that she needs Joel, for their relationship has been thoroughly developed throughout the preceding chapters. The televised version of this scene has been the subject of exaggerated praise and has been taken as a mark of the faithfulness of the adaptation to its source material (for example, Petit), but its inclusion is in fact an indicator of the issues that plague the entire series. The original video game is not a collection of disconnected dialogues and cutscenes that can be surgically detached from gameplay and then put back together unaltered. Without adequate preparation through pertinent development in the relationship between Ellie and Joel, the resulting scene ends up lacking both psychological depth and credibility. It is a foreign body, superficially implanted in a different story.

It is not only Ellie’s relationship with Joel, but also her very character, that suffers from issues of this sort. One of the areas of excellence of the original video game, as noted above, is in the portrayal of Ellie’s development, of her maturation and gradual empowerment; The Last of Us can, to a meaningful extent, be viewed as a coming-of-age story. Ellie’s vicissitudes in abstraction do not necessarily break new ground; where the 2013 video game excels, nonetheless, is in their concrete formulation, by means of language tools that are largely specific to the medium. Among these is the change of the player-controlled character approximately three quarters into a typical playthrough. During their search for the Fireflies at the fictional University of Eastern Colorado, the duo comes under attack by a group of raiders. In the struggle against one of the attackers, Joel suffers a crippling wound in his abdomen; still controlling Joel, as the picture progressively blurs, the player is tasked with escaping the facility, until, after reaching safety, Joel finally collapses onto the ground. “You gotta tell me what to do,” Ellie squeals while trying to get him to stand up. Once again, we believe her anguish, because until then she has never really been in control. The camera cuts to black on Ellie’s calls, and a text overlay announces the coming of winter. The next scene opens on a snowy landscape: armed with bow and arrows, the girl is hunting for food; in a complete reversal of their former roles, Ellie has now to provide for Joel.

It is at this point that one of the game’s central subjects, Ellie’s coming of age, begins to fully unwrap. Ellie’s newfound agency is not merely brought about through the abstract developments of the plot, but also and especially through the specific language of the video game medium: for the first time, Ellie becomes the player-controlled character (see Figure 3).17 Evidently, the transition from controlling one character to another is not equal to a mere shifting of the scenic focus. When both Joel and Ellie are on screen at the same time, the viewer sees them both, but the player controls only one: the interactive text, so to speak, is not equal to its visuals. Whereas the moving image cannot rely on the language of interactivity, the need to translate what is specific to the language of a particular medium into the language of another precisely is what adaptations, in general, are about.

In the television adaptation of The Last of Us, the change of the player-controlled character is not replaced in any meaningful way; the meaning it constructs is lost. Episode eight resorts to essentially film the visuals of Ellie’s hunt, unchanged, save for a few minor adjustments, some of which do precisely what should have been done on a larger scale: the bow, which requires a higher degree of manipulation by the player and provides them with more feedback in comparison to firearms, is thus replaced by what, in proportion to Ellie’s body, looks like an oversize sniper rifle. Ultimately, it is questionable whether the new Ellie does, in fact, mature and evolve; flattened by her intrinsic propensity for violence and significantly less reliant on Joel, she experiences no meaningful transition from child-like innocence to disillusionment, from dependence to agency—that is, from childhood to adulthood. As a result of these many changes, Ellie’s character loses much of her original subtleness and complexity; actor Bella Ramsey can do little in the portrayal of what becomes a character short of nuance.

In the end, the television series disappoints not just as an adaptation of its source material, but precisely for its inability to convey meaning through the proper language of cinematography. Important information is often conveyed through expositional and verbose lines of dialogue. For example, to show the depth of the relationship between Ellie’s mother and Marlene, the leader of the Fireflies, episode nine resorts to two clumsy sentences: “How long have we known each other?” Ellie’s mother asks; “Our whole lives,” Marlene answers after a long pause and a sigh. In episode six, Joel essentially lectures his brother with a detailed explanation of his motives for wanting to part ways with Ellie, whereas in the video game’s corresponding dialogue, Joel’s motivations regarding Ellie remain largely unexpressed.18 In the video game’s concluding chapters, the player knows that Joel is both coming to terms with his past trauma and growing closer to Ellie from how naturally the pair speaks of a number of things, including Joel’s deceased daughter, a subject that had previously been taboo. In episode nine of the television series, on the other hand, Joel overtly explains to Ellie how, thanks to her, he has overcome his past trauma and depression, lest the viewer not understand from the change in their interactions (see Figure 4).

There is, finally, another sweeping area of meaning that is lost by means of the adaptation’s unwillingness to confront itself with the specific language of its source material. Central to the identity of the 2013 video game are also the environments within which the player’s interactions unfold. These are classic American cityscapes and natural landscapes, drawn from the collective imagination and projected 20 years into the apocalypse,19 when nature has expectedly taken over—green, indeed, dominates the color palette of The Last of Us.20 These environments, however, are not merely protagonist in a cinematic fashion: although certain objects may be given particular emphasis through either text or visuals, most of the game world receives no such treatment. Neither are environments the mere background of action. Instead, they are the space actively traversed by the player and variously manipulated in the course of exploration and combat; they are the omnipresent ruins of the old world and snapshots of the early days of the pandemic (all elements of what is commonly referred to as environmental storytelling), observed through the player-controlled camera, which obeys no predetermined sequence nor perspective. In the video game medium, that is, environments, too, are defined by the range of interactions afforded to the player. Thus structured, the environments of The Last of Us in particular overflow with visual information about both the past and present and are determinant of the game’s overall aesthetics and meaning. As they are neither cinematic protagonists nor mere background, however, their particular way of signifying does not intuitively translate into the language of the moving image; their effective translation would perhaps require the willingness to linger in a Tarkovskian glance at the seemingly peripheral and inessential.21

In the television adaptation for HBO, on the other hand, environments regress to being the mere background of action. Once again, swathes of information are lost; bland, computer-generated backdrops are given the unbearable burden of standing in isolation for the layered aesthetics of ruins of the 2013 video game. This is not unlike what has happened in the one-to-one transposition of some of the game’s most iconic cutscenes: only the most linear aspects of the original video game have survived the adaptation. Resulting from an adaptation process that systematically overlooks fundamental areas of meaning of the source material, the individual parts thus obtained cannot stand on their own; at best, they become insignificant. Given the foregoing analysis, it comes with little surprise that the series’ best accomplishments reside perhaps within an episode, the third, that does not attempt to recreate the dialogues and camera shots of the original video game but instead takes the freedom of greatly expanding upon its source material and inventing its own story, stories being native to the medium.

3. Closing Remarks

In an unfortunate comment published on December 26, 2022, in The New Yorker, series co-creator Craig Mazin, interviewed by Alex Barasch, has spoken precisely of the adaptation process and differences between the two media. In Mazin’s own words, unsurprisingly, gameplay is framed as a hindrance to the actual story rather than as the locus where meaning and the story itself may be found: “When you’re playing a section,” he says, “you’re killing people, and when you die you get sent back to the checkpoint. All those people are back, moving around in the same way.” In the television series, conversely, the killing would carry more weight, for “watching a person die,” he continues, “ought to be much different than watching pixels die.” In other words, video games may attempt to say something meaningful, but, in the end, they are just games; it would therefore be sufficient to strip them of what makes them what they are to end up not only with an adaptation but also an inherently superior product, one with real actors for that matter—not “pixels.” Such comments betray an underlying misunderstanding of the language and aesthetics of video games, accompanied by an undue opinion of the superiority of film and television. Quite obviously, these are questionable starting points for an adaptation. Unfortunately, Mazin’s comments also reflect what are, in fact, much more common misunderstandings over the identity and artistry of the video game medium, its relationship with the other arts, and its distinctive ways of creating meaning.