Semiotic Review 12: Dialogues between Continental Semiotic and Linguistic Anthropology | article published September 2025 |
DOI: https://doi.org/10.71743/tq8dwq65 | Copyright © 2025 Enzo D'Armenio CC BY-NC-SA 4.0
The Case of The Last of Us: Analyzing Identity from the Perspectives of Continental Semiotics and Linguistic Anthropology
Enzo D'Armenio
enzo.d-armenio@univ-lorraine.fr
Abstract: Starting from the perspective of continental semiotics, this article seeks to establish a dialogue with linguistic anthropology through an analysis of the video game saga The Last of Us. This game not only allows the user to adopt different styles of play, but it also pushes him or her to take an ideological stance, forcing the player to step into the shoes of characters that are in narrative and existential opposition to each other. Related to this, The Last of Us has led to a polemical polarization among players, many of whom have taken sides in opposing communities. Starting from, on the one hand, linguistic anthropological concepts of register and enregisterment and, on the other hand, from the theory of enunciation and enunciative praxis in continental semiotics, I propose the concept of co-indexical identification: the process through which the semiotic work of an identity emerges by means of linguistic enunciations—in particular, through the critical condensation of competing identities already inhabiting one or more social domains—and by an ideological and axiological positioning in relation to them. The article then discusses the opposition between enunciative subjectivity and identity-related activity pertaining to the video game semiosis produced by The Last of Us. Here, I elaborate on the way in which, within any semiotic activity, it is necessary to manage the relationship between experiential and linguistic semiosis. The signs of this management reveal the semiotic work of an emerging individual identity (what I call an identification process) and open onto practices of observation or self-observation, while the temporal sedimentation of multiple identification processes leads to the emergence of social identities (through co-indexical identification). Overall, this article aims to establish the theoretical premises of a semiotics of identity.
Keywords: identity; enunciation; experience; semiotics; linguistic anthropology
Introduction
This article adopts the disciplinary perspective of continental semiotics (Greimas and Courtés 1982; Eco 1975; Fontanille 2008; Paolucci 2020) in order to construct a methodological dialogue with particular strands of North American linguistic anthropology (e.g., Silverstein 1996; Agha 2007; Nakassis 2023). This dialogue is built around a case study, the video game saga The Last of Us, and around a specific concept that emerges from this analysis, that of identity.
In my view, the case of The Last of Us presents three major challenges with respect to semiotics and linguistic anthropology. First, video games construct a particular kind of meaning experience and employ specific expressive resources.1 On the one hand, insofar as they oblige the user to produce multimodal enunciations through a ludic-kinetic “grammar”—that is, the actions available to the user in each video game—they approximate linguistic performances: it is necessary to appropriate the system of movements specific to each video game in order to produce ludic enunciations. On the other hand, the openness guaranteed by the interactive practice allows for the adoption of heterogeneous styles and behaviors, enabling different degrees of enunciative and experiential freedom.
The second challenge is to account for the specific qualities of The Last of Us video game saga. In the second installment of the game series, in particular, users are made to take control of opposing characters in a controversial story, not only slipping into the shoes of the heroine and the anti-heroine but also, in so doing, embodying opposing existential values during their performance. This polemical structure gives rise to a complex practice: on the one hand, this gives rise to experiences of adherence and repulsion in relation to the competing characters; on the other hand, these interactive enunciations implicate users’ personal styles of play. Overall, The Last of Us differs from other video games in that it does not merely invite the user to take ownership of its movement grammars and to express them in accordance with a personal style, but above all, it pushes the user to take a position on an emotional and ideological level with respect to both.
The third challenge stems from this polemical structure: the case of The Last of Us requires us to take into account the semiotic acts and linguistic enunciations produced to comment on and criticize the interactive experience itself.2 On social networks and more generally on the web, this video game has given rise to various opposing communities. Death threats have been addressed to the game’s creators in response to events involving certain characters, and controversies have erupted around gender identity and the game’s narrative coherence.
These three challenges make it possible to illuminate the semiotics of identity, revealing the way that identities emerge through interactive practices, are socialized in debates that animate opposing communities, and hence, more globally, the ways in which any identity manifests itself and is transformed in relation to semiosis. In short, this case analysis reveals the articulations that relate identities according to different levels of pertinence (Fontanille 2008; Basso Fossali 2017) and orders of indexicality (Silverstein 1996; Nakassis 2018).
To establish a semiotics of identity,3 I proceed along two complementary paths of integration between continental semiotics and linguistic anthropology, each characterized by opposing directions. On the one hand, with respect to the emergence of one or more social identities through verbal interactions, I incorporate the linguistic anthropological approach to registers into the theory of enunciation developed within continental semiotics. On the other hand, to account for the emergence of identity work within experiential semiosis, I bring into dialogue semiotic reflections on subjectivity and specific non-verbal “grammars” (in this case, video game-related) with the study of animation in linguistic anthropology. Overall, my goal is to demonstrate that all forms of semiosis—experiential, linguistic, and social—are constitutively linked to the emergence of identity work. Moreover, it is this very process of identity work that binds together different types of semiosis and different orders of indexicality.4
In the first part of the article, I introduce the reader to The Last of Us game series, focusing in particular on its narrative development and on the way in which the two installments (The Last of Us Part I [2013] and The Last of Us Part II [2020]) force the user to control different characters, endowed with different and sometimes opposing existential paths. This montage of points of view and avatar incarnations fuels particular experiences of meaning, stimulating a sense of responsibility and affection (Part I), but also of helplessness and guilt (Part II). In the second part of the article, I address the heated debate on the ludic experience that erupted online. The analysis of this controversy shows how the theoretical and methodological tools developed by linguistic anthropology—especially those of register and of role alignment—can fruitfully complement continental semiotics, in this case, the theory of enunciative praxis. As we’ll see, this controversy not only pertains to types of persons and related registers, but also requires the recognition of what I call co-indexical identification.
Finally, in the third part of the article, I explore semiotic acts that do not concern linguistic interactions, but rather the ludic enunciations triggered by The Last of Us. In order to semiotically model video game experiences, I propose the concept of kinetic diagrams—that is, the meaningful correlation between two movement systems: the system of movements to be realized on the interface, and the corresponding system of virtual movements. In effect, in order to access the video game experience and make it work, the user must realize movements on the interface (through joysticks, virtual helmets, motion capture) that are transformed into virtual movements (running, shooting, looking, climbing). In this regard, I propose the technical concept of identification process: in my usage, this expression does not stand for identification with something or someone, but for the process of emergence and manifestation of identity activity, which takes place in semiotic events through the personal management of the relationship between experience and language meaning (here, the experience of a user in relation to the language of the video game). In the case of The Last of Us, this identification process is the management of the relationship between, on the one hand, the kinetic immersion of the user due to the control of the virtual avatars and, on the other hand, the interpretive distancing that is prompted by the controversial actions the user must embody by controlling opposing characters. Co-indexical identification and identification process thus name two coordinate dimensions of the semiotics of identity, the former turning on the emergence and circulation of sedimented social identities in public domains of communication and the latter of the experiential emergence of individual identities in and through any kind of semiosis. Both work through, but capture something beyond, the threshold of enunciation as classically described in continental semiotics.
The Last of Us
The Last of Us is a video game saga consisting, until now, of two installments (or “episodes”). Belonging to the third-person shooter genre,5 one of the main mechanics of gameplay consists in taking part in real-time gunfights, in which the avatar must take cover to defend him or herself from enemies and eliminate them using weapons and other objects available in the environment. However, in this kind of video game, and The Last of Us in particular, narrative development occupies a fundamental role in the experience, as the user follows the actions of one or more characters, whose existential conditions and relationships with other characters change dramatically.
The world in which the events of The Last of Us saga take place is a post-apocalyptic one in which American cities have been devastated by an epidemic caused by a variation of the cordyceps parasitic fungi. Most humans have been infected and mutated into aggressive zombie-like creatures, capable in turn of infecting other humans. Healthy individuals are organized into factions of varying sizes and levels of hostility.
In the first installment of The Last of Us (2013, developed by Naughty Dog and released by Sony Computer Entertainment), the user follows the actions of Joel and Ellie, whose narrative journeys have much in common with the protagonists of Cormac McCarthy’s novel, The Road (2006). The latter follows the trials and tribulations of an anonymous father and son, engaged in a desperate journey across the United States in search of food and threatened by hostile human groups. In the first episode of The Last of Us, however, Joel and Ellie are not related by family ties but adventitiously meet each other in the initial phase of the story. After their initial encounter, Joel decides to protect and take care of Ellie, a 14-year-old orphan, and the two develop a relationship that quickly takes the form of father and daughter (Figure 1).
Figure 1. A screenshot of the video game The Last of Us 2013 (Sony Computer Entertainment) showing Joel and Ellie.
This bond is not only told through cinematic sequences in which the user cannot intervene, but also through interactive actions that require collaboration between the two characters: in the first and longest part of the adventure, the user controls Joel and protects Ellie from attacks by mutant beings and hostile humans. The exploration phases, in which it is often necessary to find a way forward by manipulating the environment, require Joel, and thus the user, to open a path for Ellie, establishing an effect of responsibility and attachment that reinforces the diegetic affective bond through a coherent extradiegetic experience: the player is not just the spectator of an affective narrative but performs actions him or herself that express and build affection with Ellie.
As in McCarthy’s novel, The Last of Us poses an ethical question: in the literary work, the inexperienced son cannot accept his father’s ruthless behavior, which is necessary for their survival. In The Last of Us, Joel and Ellie discover that Ellie is immune to the virus and that a vaccine could be developed from her. A human group called the Fireflies wants to use Ellie to make the vaccine, the process of which requires a lethal surgery. Although Ellie is not opposed to the surgery, Joel, under the control of the user—who is obliged to act in this way in order to advance the game—kills the surgeon and carries the unconscious Ellie out of the hospital. In the dramatic finale, Joel lies to Ellie, claiming that the Fireflies have found dozens of other immune humans and that her sacrifice is no longer necessary.
The Last of Us Part II (2020, Sony Computer Entertainment) explores the dramatic consequences of this choice. Four years later, Joel and Ellie are living in Jackson, Wyoming, in a large human colony. Their relationship has deteriorated: the user later learns that the reason for their distance is due to the fact that Joel had to reveal his lie to Ellie.
While on patrol, Joel and his brother Tommy rescue a young woman from an attack by infected creatures. The young woman, Abby, turns out to be the daughter of the surgeon that Joel killed (and who was a member of the old Firefly faction). Abby had been tracking Joel for a long time, seeking revenge for the death of her father. She attacks Joel and, in front of Ellie, who has arrived with her girlfriend Dina, violently kills him. Ellie is spared and, after a few days, decides to leave, accompanied by Dina, to hunt Abby down and seek her own revenge (Figure 2).
Figure 2. Two screenshots from The Last of Us Part II (2020, Sony Computer Entertainment); on the left, Dina and Ellie; on the right, Lev and Abby.
From this point on, the narrative follows two parallel paths, giving the user two points of view and control over two characters. First, it follows Ellie’s actions as she pursues Abby. Ellie is ruthless in certain key moments, such as when she kills the people closest to Abby (Owen, her ex-boyfriend, and Mel, his pregnant girlfriend).
The user then takes control of Abby, discovering her past, her relationships with those closest to her, and, in particular, her father. Abby’s actions are increasingly guided by human values: She doesn’t hesitate to betray her faction to save two young people belonging to the Seraphites, another faction driven by a strong religious cult. In particular, she takes care of Lev, a young transgender boy (Figure 2, right)—whose name was originally Lily—repudiated by his mother.
Discovering the bodies of her ex-boyfriend and his pregnant partner, killed by Ellie, Abby goes to the theater where Ellie, Dina, and Jess—Dina’s ex-boyfriend, with whom she is pregnant—are camped out. Abby kills Jess but is persuaded by Lev to spare Dina and Ellie. A year later, Ellie and Dina are living on a farm and raising Dina’s son (JJ, in memory of Joel). Ellie is still suffering from the trauma of Joel’s death and decides to leave to seek revenge on Abby, despite Dina’s protests and threat to leave her. Ellie is captured by a group of marauders who have also captured and tortured Abby. After freeing herself and releasing Abby, Ellie forces her to fight but ends up sparing her. Back at the farm, she finds it abandoned.
Actant, Actors, Addressee: The Boundaries of the Semiotics of Discourse
In an attempt to account for the question of identity, we begin with a preliminary analysis carried out through the classic methodology developed within the Paris-School framework of the semiotics of discourse. This allows us to examine the structure of the narration and the modalities that animate the characters according to the conceptual pairing of actant and actor. In this framework, narrative analysis is situated within the “generative path” of meaning (Greimas and Courtés 1982); this is a theoretical device that seeks to analyze a discourse by modeling the generation of its meaning. The “deepest” levels—that is, those furthest away from the concrete manifestation of specific characters, places, and times—concern very general semantic oppositions. Subsequent levels introduce an anthropomorphic skeleton of narrative events called “narrative states” and “narrative transformations,” which in turn are organized according to “narrative programs.” Such narrative programs consist of conjunctions and disjunctions between subjects and objects of value, as well as a fundamental set of modalizations of doing (wanting, knowing, being able, having to do).
Turning our attention to this surface narrative level, it is possible to analyze The Last of Us as consisting of polemical narrative programs (NPs): (1) Abby wants and needs to kill Joel to avenge her father’s death; (2) Ellie wants and needs to kill Abby to avenge Joel’s death. This cross-revenge leads to further losses for both factions, in an escalating cycle of death and violence. The object of value here varies according to the different forms of justice in the two narrative programs. Ellie’s NP starts from a form of ethical and collective justice, namely her idea to sacrifice herself to enable the realization of a vaccine that could save humanity (realizing the vaccine and saving humanity being the object of value for Ellie). From then on, she is subject to the choices of other actors: Joel denies her the path of messianic sacrifice by killing the surgeon who was supposed to operate on her, while Abby deprives her of her father figure. These events drive Ellie to pursue a narrow and, above all, ruthless form of vengeful justice towards other humans (here, the object of value shifts to revenge: killing Abby). In contrast, Abby’s journey starts from a pursuit of particularized justice—the object of value for Abby is avenging her father, even at the expense of her relationships with those closest to her, and in particular of her relationship with her ex-boyfriend Owen—which gradually transforms into a form of universal human justice: here, the object of value for Abby transforms into protection of the weak and unjustly persecuted (Lev), even at the cost of betraying her own community. In other words, the characters undergo reverse narrative movements: Ellie transitions from universal justice (her desire to be sacrificed) to specific revenge (focused on Abby), while Abby passes from specific revenge (focused on Joel) to universal justice (defend Lev even at the price of betraying her own community).
While this narrative analysis could be articulated more finely (e.g., by invoking the discursive level and thus the figurative and thematic dressing of the game’s universe and its characters), what is important for my purposes are the two important limitations that this preliminary discussion reveals in canonical continental semiotics’ structural approach to narrative programs. The first concerns the phenomenological threshold of discourse, that is, the user’s experience. We could certainly speak of the reception effects that the text generates or elicits and refer to the user’s actions as a kind of model reader or addressee (énonciataire), as is typical in continental semiotics. However, this approach does not take into account the experience of evaluation, adhesion, or repulsion that the player co-constructs in relation to the actions of the characters, which he or she partially embodies through the movements made on the interface.
The second limit concerns the social threshold of discursive semiosis: the way in which identity-related positioning is exposed in intersubjective enunciations, in opposition to other community enunciations, carrying different values. Indeed, the empirical reception of The Last of Us Part II gave rise to a series of virulent polemics, through which groups of identity communities manifested themselves. Here, again, it would certainly be possible to treat these interactions as discourses themselves, using models already familiar in continental semiotics in terms of actantiality, actoriality, and figuration. However, as in the case of user experience, we would lose the specific communicative dynamic that drives them. Communicative interactions build a different meaning process: a social semiosis pertaining to the circulation and implementation of cultural objects—in this case, social roles and identities—in more or less specific domains.
In what follows, I analyze these two boundaries—the phenomenological and the social—through a disciplinary integration articulated in two opposing directions. First, I move from linguistic anthropology to continental semiotics to examine the social boundary of discourse, namely the polemical reception of The Last of Us. Subsequently, I take the reverse path to explore the experiential boundary, which is linked to video game semiosis.
The Social Threshold: From Enregisterment to Enunciative Praxis
The reception of The Last of Us Part II on social networks is exemplary of the social threshold and pertains to the meta-commentary about the ludic experience realized by different communities. To get a general idea of the positions related to this controversy, it is possible to consult the user reviews on the rating aggregator Metacritic. The comments reported here are only examples, and this discussion is in no way meant to serve as a systematic and ethnographic analysis, but rather as an overview of the polarization of opinions concerning The Last of Us. My aim is not to characterize the specific uses of verbal language concerning these interactions, but to examine their social meaning, related to the game experience, in a general way. A first important point to note is that on the Metacritic aggregator, while the average critics’ rating is 9.3, corresponding to universal acclaim, the average users’ rating is 5.8, precisely as a consequence of the polarization of opinions.
Some of the controversies focused on the coherence of the characters’ narrative arcs, in particular concerning Joel’s death. Here are a series of user opinions expressed on Metacritic6:
I can’t fathom why they would kill of one of the best videogame characters in such a horrendous and pointless way, then destroy the character of the other one. There is nothing likeable about this game. It deseprately tries to shove political ideas down your throat, while giving absolutely nothing you can enjoy.
Killing the main character in the most worthless way possible. Insert strong independent female character. The story looks more like a teenage reality tv episode. Ellies gay by the way. What an uplifting game. Why cant we have nice things anymore. Current gaming is in a sad state.
Simply astonishing. The bold decision to kill off the main character of the series left me in shock. Playing as the "villain" was incredibly intense, especially as I began to grow attached to her friends and witnessed them being murdered one by one. Flawless. A genuine masterpiece.
One of my favourite game comes to its end. The first game was nearly perfect, but the storyline and the emotions in the second part were deeper than the Mariana trench. The route of Ellie's revenge is tearing the player apart and you need to know Abby's past and intentions. The revenge has no meaning and has no end, just a spinning circle. Somebody has to stop it and Ellie made it. I'm very proud of Naughty Dog's team. They were brave to make decisions. Don't listen to the people who downgraded the game. It's fantastic.
Here, commenters agree on the emotional intensity of the narrative arc across the games, though they evaluate it differently, some as destroying the experience, others as making the game a “masterpiece.”
A number of comments questioned the ruthless actions that Ellie—and with her the user—must undertake to take revenge on Abby. The series of murders—in particular Abby’s ex-boyfriend Owen and his pregnant girlfriend—would not be consistent with the character traits Ellie expressed in the first installment of the saga. Here, too, the evaluation of this question is itself ambivalent:
Solid step up in gameplay mechanics and the graphics are amazing. But the writing is just the biggest let down, and the theme is in complete disconnect with the gameplay and direction. You can't make a game about "violence am bad!" and then make an on the rails experience that is purely murdering everyone you come across. There's a stealth mechanic in the game, but instead of exploring the theme of "VIOLENCE AM BAD!" as the player and maybe trying to avoid conflict as much as possible, most of the game requires you to brutally kill people and the stealth mechanics just enhance that. It's crap direction and the writing is awful. After the first one. Last of Us Part II is a brutal continuation. This is a game for adults in mind. The story is cruel, brutal and unforgiving. But still you wonder what is next. And you are not happy about it. But you want to keep going. And you will get only suffering and revenge. This is what gaming is about.
A number of players found it unpleasant to have to take control of Abby, the character responsible for Joel’s death, to the point of abandoning the game or of becoming furious when Ellie spared her:
I deleted the game when i found out im gonna play as abby. Re installed it and i just felt even worse. Worst game ever.
It is not possible for me to understand the thought process making this game. The character you play half the game is generally dislikeable and there was no way to get the ending I as a player wished for and worked for. If you want to play this game, turn of the power after the fight scene in the cinema and imagine your own **** can’t make a worse storyline than Naughtydog did. Waste of time, play another game.
The history is mediocre, the gameplay and graphics is impressive. if they make a part 3, i really hope isn t about abby.
Other comments were directed at the designers of the video game itself.7 A first reason of contestation is linked to the fact that during the promotion of the game, the designers hinted that Joel would play an important role in the game’s narrative and ludic sequences, whereas he is murdered in the very first part.
The most intense site of the polarization of opinions, however, was provoked by questions of gender identity, to the point of death threats to the game designers. According to some communities, the developers followed a kind of political correctness agenda, resulting in the forced inclusion—in their opinion—of non-binary identities: in particular, the discovery of the relationship between Ellie and her girlfriend Dina, as well the young transgender Lev. These positions were, in turn, challenged by more progressive groups who labeled them as bigots:
Neil Druckmann ruined my favorite game with his inclusivity. I'm sure Joel was killed so brutally only because a white muscular man didn't fit into his worldview.
It could have been a masterpiece if it wouldn’t be for the woke agenda. I’m felling sorry that I spent so much money. I should’ve pirated this game, this was the last drop of patience, next time I’ll do my homework and if there’s going to be at least something mutual with sweet baby Inc or any woke narrative in the game I’m just gonna pirate it or not play at all.
The story has nothing to do with the first part, except for the characters. Instead of continuing the story from the first part, the focus shifted to new, poorly constructed characters full of illogical gaps and a futile exploitation of feminist and LGBTQ+ ideas. If you can't write a good story, then introducing feminist or LGBTQ+ themes under the guise of supporting minorities is the cheapest and most pathetic thing I've ever seen. If the story had focused on the world, it would have been better. I feel like the director was forced to write a story for the first part.
The game is phenomenal. Not your standard game but an experience. It'll make you feel like a smooth criminal. It'll make you feel like a sharpshooter. It'll make you mad, yeah, that's what it MEANS to do. It'll make you sad. It'll make you think and REFLECT about the world we live in. But I can assure you once you beat it you'll feel empty because there is no other game like it. The mechanics are a work of art, the actors gave it their all, the graphical fidelity is uncanny. Bigots will complain about "the woke agenda" because they have no other argument and lack the mental capacity to conceive of one. Like it or not, GAMERS, EVERYONE FROM ALL WALKS OF LIFE games. Not just bigoted 12 year olds. If you see a review that says "woke" "agenda" "SJW" or the like, you can pretty much disregard it knowing the user who posted that review is just being emotional. Ironic, isn't it, that this game is so divisive to gamers (some so very EXTREME) when its narrative is set in a warring faction-filled post apocalyptic world... play it - the reality of their world rings true to ours... doesn't it?
More generally, the discussion that accompanied this game, while deriving from the polemical structure of the game experience, goes beyond it through acts of ideological positioning within already occupied spaces of social interaction. Polemical utterances go beyond the framework of commentary on a fictional work and occupy other spaces, in relation to which a positioning—resulting from a polemical condensation of already existing identities (“woke,” “bigot,” “SJW,” “12-year-old,” etc.)—is expressed: for example, in relation to the ethos of the directors (“brave” vs. “dishonest”) or to gender identity (“bigots” vs. “woke”).
Overall, this ideological clash between different communities, fought through opposing verbal enunciations, can be analyzed in accordance with the linguistic anthropological notion of second-order indexicality. Michael Silverstein noted that textuality is founded on a constitutive doubling: on the one hand, there are the processual operations that make it possible to detach a text from a semiotic interaction and render it autonomous (entextualization). On the other hand, there are those elements of textual utterances—themselves also processes—that point to (“index”) and thus construct the context of that very text (contextualization). Silverstein (2019:55–56) writes:
The event characteristics of language manifest as those of discourse-in-context, and to the degree that discourse is doing its sociocultural work, it precipitates for the interlocutors an organized or structured trace, one that is an interpretative conceptualization both of “what is said” and of how it counts as “what is done” in the instance. It emerges, in other words, as what we term text in a duplex condition. The process of coming to textual formedness we term entextualization; the process of how discourse points to (indexes) the context which seems to frame it we term contextualization.
For Silverstein, these characteristics of textual detachability and of the precipitation of (texts-in-)contexts in enunciative exchanges are all pragmatic processes of indexicality. As a result, the (ethnographic) study of semiotic interactions by linguistic anthropologists brings together, and attempts to analyze the patterned unfolding of, the multiple indexicalities contained in enunciative interactions: those that point to the enunciation situation, as well as those that indexically invoke broader orders of social meaning.
Here, Silverstein distinguishes between two dimensions of indexicality: as presupposing and entailing. A distal demonstrative like that, for example, is relatively presupposing in that its meaningfulness presupposes the independent existent of its referent, while, by contrast, the proximal demonstrative this (e.g., in presentative uses of introducing new information) is relatively entailing, insofar as its meaningfulness turns on the discursive creation of its focus of attention. Similarly, while a second-person pronoun (like French tu), to some degree, presupposes the independent existence of its referent, as constituting an addressee it is relatively entailing (cf. an anaphoric third-person il). Similarly, the presence of honorific formulas (e.g., plural second-person pronouns like vous used for singular addressee) involve a further entailing indexicality, one that invokes more encompassing “orders of indexicality” such as, for example, types of addressees (those to whom deference is expected/demanded) and speakers (those who are polite, impolite, and so on). A central aspect of this tradition is the study of the constitutive connections between different orders of indexicality. Constantine Nakassis writes:
Silverstein (1996, 2003a) splits indexicality into “n-th” and “n+1-th” order indexes. Their dialectical relation comprises the earlier, related splitting of “presupposing” and “entailing” indexes and the supplements “metapragmatics” and “ideology.” They are related in the following way: any n-th order index (e.g., a second-person plural pronoun) indexically presupposes something of its context as a condition of its “appropriate” use (e.g., the existence and co-presence of a potential addressee), and in doing so may, under particular (co-)textual conditions, indexically entail something in the context of its use (e.g., honorification to singular addressee). […] Hence, to continue our example, the use of a second-person plural pronoun to honorificate a singular addressee can also function to index qualities of the speaker—as gentile, polite, old-fashioned, and so on. Such an n+1-th order indexicality, thus, stands as a “virtual” contextualization of the (n-th order) token index. (Nakassis 2018:296–97)
In the case of The Last of Us controversy, we can also speak of linked orders of indexicality: the first order is that structure of indexes that function within, and precipitate, the “denotational” text of the video game (e.g., its game play and narrative); in addition, however, are those polemical enunciations that meta-semiotically link the experience of this denotational text to a second-order of indexicality, which concerns ideological and social oppositions, represented by the mobilization of semantic universes pertaining, for instance, to the figures of “bigot” and “woke.” Here, thus, the presence of a narrative feature (e.g., the indexes of a trans identity of a character) serves as a second-order index of the political positions of the game and its makers (it is “woke”), and these, in turn, of its users (depending on how, of course, they align to the “lower” indexical orders). These polemical enunciations cut out (via processes of entextualization) segments of the first-order denotational video-game text in order to insert them—through an enunciative contextualization—into an interaction pertaining to ideological oppositions that already inhabit society (“outside” of the game).
The most interesting concepts for the purposes of this analysis, however, are those of register and of enregisterment. According to Asif Agha (2007), a register is a socially typified and recognizable way of speaking (and more generally, of using signs), often designated by a meta-description (children’s language, military speech, legal jargon, etc.). This concept of register takes full account of the materiality of a language’s substrate. For instance, an Italian accent in an English verbal enunciation may be enregistered, for some social domain of persons, as a register typical of an Italian speaking English, in the same way as the typical way a soldier speaks in combat contexts may be recognized as “military jargon.” It is only as a result of enunciative sedimentation and meta-pragmatic typification that the “Italian accent in English” or “military jargon” are constituted as registers. Agha writes:
In the technical usage, however, it is necessary to distinguish registers as products (or precipitates) of human activity from the process through which they are formed. This distinction may be formulated as follows: a semiotic register is a repertoire of performable signs linked to stereotypic pragmatic effects by a sociohistorical process of enregisterment. To speak of ‘a register’ (to use the count noun) is to speak of a sociohistorical phase, a moment-interval of a process of enregisterment; the term ‘enregisterment’ (a deverbal noun of process) makes reasonably transparent the fact that a register’s existence is mediated by activities which make known (or ‘enregister’), and thus make usable, facts of semiotic value associated with signs. (Agha 2007:80)
In short, registers organically link ways of speaking, stereotypical social figures, and the possibility of recognizing and reusing them in new enunciations (i.e., recontextualizing them). Agha explains that:
Encounters with [speech] registers are encounters with characterological figures stereotypically linked to speech repertoires (and associated signs) by a population of users. Language users typify such figures in social-characterological terms when they say that a particular form of speech marks the speaker as masculine or feminine, as high- or low-caste, as a lawyer, doctor, priest, shaman, and so on. (Agha 2005:45)
This concept is important for continental semiotics because, along with the notion of indexical order, it allows us to more comprehensively describe the notion of enunciative praxis (Bertrand 1993; Fontanille 2006; Dondero 2020; Paolucci 2020; also see Padoan 2025 and Dondero 2025 in this issue).
The theory of enunciative praxis seeks to overcome the reduction of linguistic phenomena to only two modes of existence, as seen in the theories of Ferdinand de Saussure (1995) and Émile Benveniste (1970): the virtual system of language and the discourses realized through it (or utterances). The notion of a linguistic system with stable contours, which speakers appropriate to construct any enunciation, is called into question in light of the co-evolution between linguistic usage and language itself. In other words, with enunciative praxis the focus shifts to collective and historical enunciations; here, the linguistic system is conceived as a provisional and ever-changing outcome rather than as the invariable origin of discursive initiatives. For this reason, the theory of enunciative praxis postulates that the temporal and collective sedimentation of enunciations schematizes semiotic configurations through the “syntax” of four modes of existence (virtualized, actualized, realized, potentialized), rather than just two. Collective enunciations construct virtualities and cultural potentialities (i.e., are available in some historical community), which can be, by degrees presenced in particular moments, that is, actualized and realized again.
The role of these virtualities and potentialities remains vague in continental semiotic thinking, however, and is nourished by other concepts (also often vague in their description): genres, stereotypes, usages, and the like. In visual semiotics, for instance, enunciative praxis helps explain how every new image is produced in relation to a genre (portrait, landscape, etc.) and a social status (artistic, religious), through the reconfiguration of virtual and potential visual configurations. However, this mediation between linguistic enunciation and social semiosis has not yet been fully worked out. The process of enregisterment helps clarify this dynamic, as it precisely specifies the process of how enunciations and utterances are sedimented, thereby schematizing as virtual and potential, not only textual configurations and genres, but also ways of speaking/communicating and types of persons.8
There are some important corollaries that follow from these points. First of all, a register is constitutively linked to a social domain. According to Agha: “Registers have a social existence only insofar as—and as long as—the metapragmatic stereotypes associated with their repertoires continue to be recognized by a criterial population of users, that is, continue to have a social domain” (Agha 2005:46). In a way, a register is not only linked to a social domain, but also helps to co-construct and transform it: “One cannot become a doctor or a lawyer, for example, without acquiring the forms of speech appropriate to the practices of medicine or law or without an understanding of the values—both cognitive and interactional ones—linked to their use” (ibid.:51). In short, becoming a doctor or lawyer is also a matter of mobilizing the corresponding registers and understanding the values that characterize a given social practice. Registers also encompass a social range, as only individuals familiar with a particular way of speaking and specific social figures can recognize and distinguish them from other registers: “Such a model is inevitably a model for someone; that is, it involves a social domain of persons who recognize it as a model enactable through speech” (ibid.:46).
Finally, Agha highlights two aspects of the relationship between the perceptibility, enunciability, and describability of registers: the ability to perceive discursive patterns relating to registers is broader than our ability to describe them. We can easily perceive discursive differences in register, but the transition to meta-linguistic description is more complex. Moreover, even if registers are configured as pragmatic, lexical, and semantic repertoires—ways of speaking—they are always mobilized in particular events (i.e., under conditions of entextualization/contextualization) in a mixed, imperfect way; in particular, this unfolds according to complex alignments of interactional patterns, rather than precise correspondences.
In other words, registers and enregisterment processes offer—in relation to the sometimes vague labels of genres, stereotypes, and usages—an account of enunciative praxis that is already articulated in relation to social domains and, above all, an enunciative praxis that specifies and co-constructs social domains (the social threshold discussed above). If continental semiotics is concerned with all mediations of meaning, then the schematization of praxis concerns both the schematization of enunciations and socio-semiotic relations and types of persons. For this, we need the concepts of enregisterment and indexical order.
Third-order Indexicality: Co-indexical Identification
Two elements of discussion are particularly important in relation to The Last of Us and the question of identity. Firstly, the fact that registers are constitutively linked to the existence and recognition of types of social roles, and that their use in enunciations allows for the analysis of the complex alignment process realized in linguistic interactions. (Agha [2005, 2007] uses the term role alignment to speak of the way in which users of enregistered forms align themselves to the figures and cultural values invoked by those forms.)
Secondly, registers and enregisterment processes link, or rather cross, in a unitary way semiotic phenomena situated at different orders of complexity (different “orders of indexicality”): enunciation (because these are “ways of speaking”), domains of communication (the social roles associated with these ways of speaking), and perception (or rather a linguistic perception that allows, or not, the recognition of the register and the role enregistered).
However, in the case of The Last of Us, and more generally in any polemical exchange on the public scene, the quality and quantity of role alignment processes between an enunciation characterized by a set of registers and the respective types of persons often give rise to a third-order of indexicality. Indeed, the oppositions between “woke” and a “bigot” as types of persons and their respective speech patterns are already enregistered in the social scene, and gamers use them in an indexical manner to express their identities and the identities of others, while also pointing to their interactional experience with the video games. If we limit ourselves to describing this field as organized by oppositions (woke vs. bigot) and their respective ways of speaking, however, we leave little room for the complex work of alignment that each enunciation produces by personally rearranging these oppositions in relation to the ludic experience with The Last of Us video game saga.
A player can certainly counter a discourse by claiming that it expresses a bigoted attitude and dismiss with disdain the accusation that the developers followed a politically correct agenda. But he or she may at the same time express personal dissatisfaction with the depiction of violence in this video game and criticize the consistency of the narrative development of the characters. In other words, when enunciations construct different orders of indexicality starting from a domain already inhabited by contrasting positions, the speaker organizes a dramaturgy of identities in order to position him or herself in a particular way with respect to these identities. The dramaturgy of mobilized positions, to which individuals take varying role alignments, vis-à-vis different orders of entextualized indexicalities, thus gives rise to what I call co-indexical identification. With this expression, I do not mean the way in which an individual identifies with someone or something, but rather the process through which an identity manifests itself and emerges through the condensation of identities that inhabit a social space, and, correlatively, the personal positioning that takes place with respect to these social identities.9
The production format elaborated by Erving Goffman in Forms of Talk (1981) helps us to clarify this dynamic. Goffman distinguishes three (speaking-)roles and types of activity in relation to the production and enunciation of verbal utterances. The author is the person who composes a discourse (regardless of whether they deliver that discourse). The animator is the individual who performs the discourse—a “talking machine”—regardless of whether this person is the author of the speech. Finally, the principal is the entity who guarantees the values of a discourse, who embodies an institutional reputation or that of a group of people. This division into three types of actants is often realized in a complex manner in social interactions, by exploiting overlapping roles and strategies of enunciative delegation.
With respect to the controversy surrounding The Last of Us, we can distinguish three types of cases of meta-discursive enunciations, that is, enunciations that draw on other semiotic productions. The first case is illustrated with the example, discussed above, of a direct critique of the video game The Last of Us through a convocation of a Goffmanian principal: the video game The Last of Us is a discourse that animates the values of a well-established agenda, that of a political correctness typical of “woke” social groups.
A second case, as we saw above, concerns implicit judgments of The Last of Us, which openly criticize, in an indexical manner, the authors (the developers of The Last of Us) as the principals of a social group that promotes politically correct values. In both these cases, the meta-discourse connects two levels of indexicality: it pertains to an entextualized text (The Last of Us as a semiotic production) but constructs a second-order opposition by invoking the social entities concerned (the developers as “woke” authors and principals). Within this second-order indexicality, individuals producing the meta-enunciation place themselves in direct opposition to the invoked principals: thus, they stand against “woke” values and the identities that embody them and employ procedures of role alignment by adopting registers that contrast with the principal constructed as adversary: they may, in other words, adopt a politically “incorrect” register while framing the opposing identity as “woke.”
And yet, it is necessary to consider another case, where meta-enunciations likewise construct a judgment of one or multiple discourses (The Last of Us, but also the critiques the game received as an expression of “woke” values), invoking other principals (judged as bigots, for instance), but not simply occupying an opposing stance, or animating values from a typical register or social type. Instead, these (meta)critiques bend these positions to express a personal enunciation through a more complex positioning. This is the case, for example, of enunciations that denounce certain stances on the game as bigoted while still criticizing the game or its authors for other reasons: for narrative inconsistency, a misleading marketing campaign, or excessive violence. These meta-enunciations do not express simple processes of role alignment, as they allow for the emergence of an identity-related activity that does not overlap with the animation of a recognizable person type or role, but rather with the animation of a complex role and the emergence of a specific identity activity. For this reason, I propose defining this dynamic as a process of co-indexical identification, which employs a third-order indexicality through a dramaturgy of social roles in order to build, in a personal way, a different identity position. Here, pregiven enregistered identities are contextualized in practices that construct novel identities that navigate “lower”-order indexical values. The expression co-indexical identification thus captures the co-ordination of different indexical relations in this process: first, that it involves a meta-enunciation that indexes an event, a discourse, or a character (in this case, the denotational text of The Last of Us); second, that it simultaneously develops through an indexical gesture that points to types of persons that already inhabit the social space (the “woke” and “bigoted” identities that preexist the meta-comments on The Last of Us). In other words, identification is realized through a “vertical” indexical relation with an utterance or event and a “horizontal” co-indexical relation with existing social identities.
It is necessary, in this regard, to clarify the relationship between the concept of co-indexical identification and that of social identities per se, attributing to the latter, here, a technical, semiotic sense. In my view, for a social identity to be established, an isolated process of co-indexical identification is not sufficient; rather, there must be a sedimentation of these processes that allows for the recognition (i.e., presupposition) of an identity in the public sphere, from which further co-identification processes may entail further identities. In other words, three conditions are necessary for a social identity to come into being:
Enunciations that go beyond simple processes of role alignment and that involve a dramaturgy in relation to different registers and/or types of persons. In the case of The Last of Us, this means embodying both the register and role of a gamer who evaluates their experience and the role of an individual politically aligned with existing values within a complex discursive realization.
A temporal sedimentation of semiotic acts (verbal, gestural, multimodal enunciations) performed under the pressure of different social situations. In the case of The Last of Us, third-order enunciations and the related process of co-indexical identification are not sufficient, in isolation, to construct a social identity. A sedimentation (a virtualization) of social identification processes is necessary to allow for the recognition (actualization and realization) of a specific social identity in the public sphere: for example, that of a journalist who has already evaluated various video games, and assessed their strengths and weaknesses, while also enunciating explicit judgments on their political value.
An explicit (denotational) or implicit (through register modulation) condensation of identities that occupy a social space, in relation to which they express a value judgment or an identity positioning. This is the case of a journalist who chooses to evoke the accusations directed at The Last of Us to assert his or her critical stance.
The Experiential Threshold: from Kinetic Diagrams to Animation
The concepts of enregisterment and types of persons, as well as those of co-indexical identification and social identity, would function in a nearly analogous manner if, instead of examining the social semiosis related to video games, we were analyzing an equally controversial reception of a novel or a TV series. Analyzing the meta-discursive comments pertaining to the reception of a cultural product is certainly important, but perhaps even more crucial is examining the semiosis that unfolds during the reception itself. In the case of a video game like The Last of Us, it is necessary to consider the phenomenological boundary of discourse, namely enunciations that do not occur through verbal language, but through ludic interactions. What happens if we take into account multimodal semiosis that does not involve verbal enunciations? And what are the implications for identification processes?
Video games are usually studied with respect to two concepts: that of immersion and that of interaction (Ferri 2015; Calleja 2011; Pinotti 2020). The second, arguably, makes it possible to distinguish interactive media from traditional media such as literature and cinema: the user, being able to participate in the experience through his or her choices, would assume an authorial role as opposed to the more “passive” reception of traditional media (though see Nakassis 2025, in this issue). As for immersion, the fact that video games stimulate more senses and require coordination to cope with the virtual experience arguably means that they have greater effects on immersion (in narrative, emotional, or aesthetic dimensions).
With respect to these two concepts, it appears necessary to focus more carefully on the way a player accesses the virtual world of the game, and on the way his or her experience is articulated in grammaticalizations of the language of the game.10 Indeed, the user first of all accesses the virtual world visually: the two-dimensional screen presents a (two- or three-dimensional) immersive world, marked by a more or less pronounced level of figuration (photorealism, stylization, abstraction). And yet, so-called interaction emphasizes the fundamental importance of another aspect: in order to access virtual experiences, in order to make them work, the user must perform time-based movements on the interface (by means of joypads, gyroscopes, motion sensors or virtual helmets) that are transformed into movements within the virtual world (e.g., running, jumping, shooting, or looking). On the one hand, movements on the interface are usually abstract and plastic, because they are not expressed through recognizable figures or themes, but through abstract commands such as up, down, left, right, or using other activation keys (A, B, X, Y) and progressive triggers. Movements within the virtual worlds, on the other hand, often translate the abstract movement on the interface into a system of figurative and thematic movements that are specific to each video game (running, jumping, shooting, climbing, etc.).
Between these two types of movements, a kinetic diagram associates the abstract commands on the interface with more or less figurative and thematic movements within the virtual world.11 This diagram of kinetic relations establishes the rhythmic and aspectual resonance between the two systems of movement: for example, punctual movements such as jumping within the virtual world may be matched with equally punctual movements on the interface, while durative movements such as running require prolonged activation of the commands.
In other words, there are two main forms of “grammaticalization” as regards video games: on the one hand, a visual syntax that regulates the degree of figurativeness and the architecture of virtual spaces; on the other, a kinetic syntax that regulates the interactive micro-language through which the player exists (and enunciates) within each specific video game. Rather than stopping at the simple fact that these are interactive media, this approach makes it possible to analyze directly the kinetic micro-languages (offensive, defensive, contemplative) specific to each virtual production.
Visual syntax can be analyzed in the light of the criteria already formalized and stabilized in continental semiotics (Dondero 2020): in particular, vis-à-vis the configurations and oppositions of colors, shapes, and spaces, as well as recognizable figures (plastic and figurative relations; Greimas 1989), the way in which images engage in different forms of dialogue (i.e., enunciation; Marin 2002), and the way in which the point of view allows images to affirm, hide, or deny (cognitive modulations; Fontanille 1989).
As far as the kinetic syntax is concerned, in other work I have proposed two fundamental parameters (D’Armenio 2022): the form of movement and the body of movement. The form of movement refers to the qualities directly expressed by the movement in the game: plastic forms (pure trajectories, accelerations, rhythms), figurative forms (movement that constitutes figurative motifs, e.g., the motif of falling), and thematic forms (recognizable and nameable actions, e.g., “climbing”). The movement’s body is its substratum: a body can either be absent from the frame of the image or constitute itself as an indirect body, deducible from the movement, or be present in the form of a human, animal, or object (Fontanille 2004).
This results in specific meaning processes. On the one hand, the user partially embodies, through his or her actions on the interface (pressing a button on the joystick), the movements performed in the virtual world (pulling, running, climbing), thanks to an analogical transposition. On the other hand, the interpretative process enables the user to play the role of a virtual character different from him or herself, as well as to precisely inhabit a well-characterized virtual world. Overall, the user’s identity—here in the commonsense notion of the term—is combined with that of other virtual identities, thanks to the imaginative mediation between kinetic embodiment and interpretative distancing. The kinetic diagram, in other words, allows for the translation between the phenomenological threshold of the game and its discursive level.
Here, to clarify the bidirectional relationship between the player and the avatar-character we can helpfully refer Goffman’s discussion in Frame Analysis (1986[1974]), in particular to his typology of what he calls “figures.” For Goffman, a figure is animated by a speaker (animator) during a verbal interaction situated within a more or less specific social frame.12 This concept of figure includes the notion of a social type [in Agha’s terms, the “characterological figure”] that a speaker takes a role alignment to in an event of enunciating some enregistered form. Offering a typology of figures, Goffman distinguishes between different kinds of figures, most important to use being his distinction of staged figure and natural figure.13
In the case of the relationship between the avatar-character and the player, the avatar is what Goffman calls a staged figure. Compared to natural figures, which involve an identification between the enacted figure and the animator/author of the discourse—the way we, as biographical persons, speak in everyday interactions, where we are often both animator and author of our utterances—staged figures concern “fictive or biographically derived characters on the legitimate stage, screen, radio, and cartoon page” (Goffman 1986:525). In the case of a theatrical performance, for example, an actor (animator) projects a staged figure through his or her body and enunciation. However, the actor is not the exclusive source of that figure, as the actions of characters are based on a script, usually created collectively.
However, the animation of an avatar in a video game—whether human, an object, or an abstract entity—complicates this typology of figures. While Goffman briefly refers to games—“A chess player is within easy reach of his pieces, his men, his figures” (Goffman 1986:522)— his typology of figures is primarily designed to analyze verbal interaction, theater and cinema, and print reportage, media where, unlike action video games such as The Last of Us (and other media, e.g., puppetry), animation is not mediated through a medium/language-specific kinetic syntax/diagram of movements. In cases like The Last of Us, the user has to construct multimodal enunciations that appropriate a ludic-kinetic grammar that is specific to the video game in order to build his or her ludic enunciations. At the same time, the openness guaranteed by interactive practice enables the adoption of heterogeneous styles and behaviors, authorizing different degrees of enunciative and experiential freedom. That is, the ludic-kinetic grammar of a game allows for the in-game expression of different styles of play (in The Last of Us, e.g., different ways of engaging in armed struggle in clashes with other human factions or mutated beings). Rather than staged figures, we might speak of virtual, ludic figures.
Importantly, animating such figures is not only about moving a character or making them act in some particular way; it also involves the construction of what we might call a “staged world,” that is, in and by animating the character the user is also animating into being the visual and architectural features of a virtual space within which such characters exist; in such games, the visual syntax of the game thus not only provides a diagram for the animation of virtual, ludic figures; it also establishes the frame of the fictional world and thus influences the way in which the player animates the avatar (by constraining and enabling movement, but also by suggesting ways to explore movement within the space of the staged world).
Yet such virtual characters—including the avatar—are also animated according to a narrative script, making them also staged figures in Goffman’s sense. However, through their multimodal autonomous enunciations, they contribute to evolving and layering the fictional frame, thereby modalizing the player’s kinetic actions, which unfold under the influence of both the fictional figures and their narrative progression.
These considerations align with the concept of animation developed by the anthropologist Teri Silvio (2010): the animation of a character is not reducible to a simple representation or the enunciation of a fictional entity; rather, it involves relations that contribute to the user’s identity positioning. In the case of video games, the user has agency with regard to the animated avatar that is under his or her control, but even the avatar, as a figurative and narrative entity (a staged figure), shapes how the user performs in the virtual space according to diagrammatic adjustment. In a way, video game experiences can be understood as simplified rituals (Silverstein 2004; see Padoan 2025, in this issue): the user performs predetermined multimodal symbolic actions but retains a semiotic margin of maneuver with respect to the roles he or she is assigned and with respect to the ideological and narrative organization of the game. As in ritual, the user must negotiate his or her identity with respect to the identity roles he or she is partially embodying through it. The user is stitched into a (cosmically) narrated world, as in ritual (what is, in Silverstein’s terms, nomically calibrated), through the multimodal poetics of the events arranged by the video game, but has to reflexively instantiate movement in that world: that is, the user is at the same time situated through movement in the virtual world (“shifted out,” in the terms of continental semiotics), and has to manage his or her identity in the “real” world under the narrative and ideological pressure of the avatar’s role (or “shifted in”; “reflexively calibrated” in the terms of linguistic anthropology; see Silverstein 1993; Nakassis 2020, in press). In other words, just as linguistic anthropology’s reflection on registers allows for the development of the dynamics of enunciative praxis in semiotics, similarly, the association of visual syntax and kinetic syntax can also help extend and specify the concept of animation in linguistic anthropology in a way that highlights certain distinctive features of video games semiosis.14
What I wish to highlight, here, however, in connection with discussion in the first part of the article is the interplay between ludic-kinetic grammar, narrative, and play style (at all its orders of indexicality, from the idiosyncratic individual to the socially conventionalized). In particular, how visual grammar and cinematic storytelling emotionally charge action sequences with a sense of urgency and, conversely, how the outcome of kinetic actions in the game is reframed by narrative sequences that amplify their impact in accordance with specific narrative developments, and vice versa (how the narrative is emplotted only through the users’ kinetic actions).
Hence, in the first chapter of The Last of Us, as we earlier observed, the player primarily controls Joel as he protects and collaborates with Ellie throughout most of the interactive events. These actions include not only combat sequences, but also sections in which it is necessary to find passages to advance, where Joel and the player are often engaged in locating an opening that allows Ellie to move forward safely. These interactive actions are often accompanied by scripted dialogues between the two characters, aimed at reinforcing their diegetic bond during gameplay. Furthermore, the cut-scenes, in which the player cannot intervene, strengthen the bond between the two characters, progressing their kin-like relationship. In one of the later sequences, Joel is in danger, and the player takes control of Ellie as she tries to save him. The sequence is thus marked by a sense of urgency: saving Joel is essential from Ellie’s perspective—Joel is the only loved one left for her—and the player is given this responsibility, which adds an emotional surplus to the gameplay performance. In other words, through the kinetic and embodied grammar of the game, we are made interpretatively complicit in the emotional pathos co-generated by the narrative. This happens through the shifting of figures involved (the game moves between and blurs ludic figures and staged figures just as players move from being only animators to also partial authors, of a very constrained sort, of the narrative). The players are variously invited to take up role alignments to these figures (including as a kind of principal—whose ethical position is staked by the game’s play) and thus, to the wider field of enregistered positions indexed by the game and its play.
In one of the final sequences, Joel is faced with the sacrifice of Ellie, who, as the reader will recall, was not against the surgery in order to produce a vaccine for the epidemic, even at the cost of her own life. In this sequence, the player controls Joel’s movements but has no freedom in how to proceed through the experience: the player is forced to kill the surgeon or else not proceed forward in the game. This sequence thus creates a disproportion between the moral choice (saving Ellie at all costs or accepting her sacrifice, thereby condemning humanity) and the kinetic obligation to save her in a controversial manner (by killing the surgeon). The way this conflict is resolved belongs to the diegetic character Joel, but because it is partially embodied by the player, the latter is forced by the kinetic grammar of the game to participate in a paradoxical action: he or she may not necessarily want to make that choice but is compelled to do so and to embody a specific violent interpretation of that choice. This is an instance of paradoxical, active enduring: the player undergoes the choices of another entity while being obliged to partially embody it.
In The Last of Us Part II, this paradoxical coupling of kinetic control and the merciless choice by diegetic characters is further amplified: the player has no influence over the intertwined acts of revenge between Ellie and Abby; he or she can only partially embody their choices and actions. The overall effect is that the player is certainly led to identify—in the vernacular sense—with both opposing avatars, but their kinetic embodiment paradoxically reinforces an interpretive distance, prompting the user to take a reflexive stance regarding their narrative trajectories; in effect, the ludic-kinetic interface and narrative progression of the game itself prompts and thematizes the processes of role alignment noted above, motivating the production of higher-order indexicalities and the suturing of the game within wider envelopes of enregisterment (i.e., into a political field of identities). The Last of Us is thus an exemplary case because it utilizes the mechanisms of kinetic embodiment and interpretive distancing typical of video games to articulate them in a critical way.
Subjectivity and Identification
In this penultimate section, I take up the question of the relationship between subjectivity and identity. While the first term has been central to continental semiotic discussions of enunciation, the latter has figured prominently in linguistic anthropological discussions of social interaction and enregisterment. Because it is linked to enunciation, subjectivity, whether treated in terms of instances or simulacra, tends to be limited to the field of a single discourse or set of intertextual relations; it is the act (or the figuration of the act) of enunciation—the appropriation of a virtual system—through which a subject is entailed. Identity, on the other hand, as I’ve argued, is not an event-specific phenomenon; it is the result of an accumulation of positions stratified over time (Ricœur 1992).15 That is, identities involve the sedimentations—the enregisterment—of enunciated subject positions.
Another fundamental difference between enunciative subjectivities and identities is that the latter are not limited to language activities, but concern all semiotic activities, including those that escape codified, conventionalized semiotic systems (i.e., “languages”): practical behaviors (Fontanille 2008), somatic “predicates” (Coquet 2007), specific modalities through which individuals thematize (Greimas and Courtés 1982) and scenarize (Goffman 1981; Basso Fossali 2017) the activity of other individuals during experience, often without producing verbal signs but rather behavioral signs of diverse types. Identities exceed language (langage)—in the sense of codified virtualities—and their appropriation—and thus the subject precipitated therein—insofar as they include experiential semiosis. For these reasons, rather than enunciation and linguistic subjectivity, I prefer to speak of an identification process: a process through which an identity is manifested and emerges. This process is distributed beyond the subject and relates to the way in which, in any semiosis, we have to position ourselves in relation to the actors (and figures) that surround us or inhabit the environment or the text we are engaging, through operations that are not always produced through language (broadly understood). If I had to give a concise definition, I’d say that the identification process is expressed through the personal management of the circuit of meaning that links experience and language (langage). The signs of this management are the ways in which we resolve, hesitate, and undergo—and therefore, more globally, manage—the relationship between experiential semiosis and enunciation. These lived operations are at the same time an opportunity for the self-observation of both ourselves and our reactions, which can trigger a modal control and an evaluative activity. In other words, an identification process is an epi-semiotic process (épi-sémiotique) in the sense that it gives rise to a conscious or unconscious meta-semiotic activity within semiosis itself: it is an “internal” management of semiosis that opens up the possibility of meta-semiotic activity. The process of observing or controlling the identification process (for example, pretending to be worried during a poker game through a bluff) is a meta-semiotic process that takes semiosis as its object and can lead to enunciations pertaining to the identification process itself (thus meta-enunciations such as: “I was bluffing” or “I was stressed”). In identification, we relate to our own semiotic activity by taking a stance on what we are doing; we reflect on it while doing it, not necessarily in an explicit way, but in an experiential, phenomenological way (such as the disgust we feel while playing the game). This epi-semiotic activity, as we have seen above, is distributed across a process of entextualization and contextualization in a history of enregisterment processes. That is, it unfolds within an enunciative praxis that sediments identification processes.
The notion of self related to this epi-semiotic process of self-observation is in no way intended to indicate an instance of origin for semiosis, but rather is its consequence: It is through one or multiple semiotic events that the activity of an ego manifests itself and can thus be observed or self-observed. This is precisely what we see in The Last of Us. When the player, as Joel, is forced to kill the surgeon, the player faces an obvious modal (and moral) conflict: many players have hesitated for a long time before killing the surgeon; others have done so mechanically, and have also murdered the two nurses present on the scene, whose deaths are not compulsory.16 In other words, the modal and moral conflict expressed in the circuit between experience and ludic enunciation implies a more or less personal processing, even in the case of a video-game sequence whose outcome is largely predetermined. During my playthrough, by partially embodying Joel through gestures on the interface, what I was about to do seemed justified within the framework of the fictional work: save the only person with whom he has forged an emotional bond, and whose actions count in a ruthless world. This rationalization allowed for me to continue with the game by managing my own personal affective relationship to the game’s unfolding, and my complicity in it.
Similarly, recall how some players abandoned The Last of Us: Part II when the point of view changed and forced them to play the role of Abby, the character who had previously killed Joel. Among those who completed the experience, some felt that the construction of the young woman’s past was forced. In my own experience, I felt a sense of discomfort in playing the role of Abby, and although I understood her choices, I found myself wondering which of the two women the user should choose according to the semiotic logic of the text. Following this, I wondered who I felt closer to and who I would choose to support. At the start of the experience, my attachment to Ellie was clearly stronger. Then, as I partly embodied her ruthless actions, I realized that there wasn’t one actor to favor, but that we were dealing with two victims trapped in a system of violence and revenge.
Finally, consider the last part of the game. Ellie has been living with Dina and her son for a year, since Abby spared them. Tormented by panic attacks and nightmares, she decides to leave once again to seek revenge. I found this obligatory departure totally unnecessary: taking on Ellie’s perspective, it seemed a disproportionate risk compared to the responsibility of caring for a child and the exhausting journey already accomplished. And yet, at the end of the expedition, having reached, confronted, and spared Abby, I reconsidered Ellie’s actions. From her point of view, that of someone who has lost a father figure and who was psychologically and somatically tormented, I told myself that Ellie’s departure was understandable in order to restore her equilibrium. In contrast, many negative user opinions on The Last of Us: Part II expressed disappointment not with this new departure, but with the fact that it remained unfinished: in their view, it is unacceptable that Ellie did not finally take revenge on Abby for Joel’s death in the final sequence.
In these three sequences, the user is invited to combine his or her enunciative and evaluative actions as enunciated subjects in/of the game according to a personal approach; and this management allows a process of identification to emerge in which the user experiences and discovers fragments of his or her own identity in enacting that of (narrativized) others. Having to inhabit the actions of two antagonistic characters—two subject positions—within the same enunciation provokes a process of identity co-construction fueled by semiotic processes that are dynamically articulated moment by moment. This involves experiencing the events, actions, and reactions of the virtual characters as disproportionate, appropriate, amoral, virtuous, and reinterpreting these experiential judgments through a practice of self-observation, that is, through a form of role alignment and a reflection on that very alignment, one which produces evaluative assumptions of responsibility in taking on the point of view of the fictional characters. In other words, it’s a question of identification processes that have to manage the disproportions that emerge between experience and enunciation, and which are expressed through this arbitrage. This identification process is not a purely individual process (nor does it involve a purely individual identity), however, insofar as this process itself unfolds within a communicative envelope of enregisterment of already existing, sedimented identities.
Yet through the medium of this highly personal and yet also highly social process of identification, the sedimentation of these identity management processes can, in turn, lead to the emergence of more enduring individual identities that are intersubjectively observable and indexable. This identity has a variable social range and typically pertains to multiple social situations—for example, the family sphere, the professional sphere, and mixed situations—and, in general, allows one to identify a way of managing the relationship between experiential and language semiosis as characteristic of an individual.
When identification processes involve the management of different orders of indexicality and draw upon identity positions that already inhabit a social domain as a foundation for constructing a personal indexical positioning—a dramaturgy of social roles—I have referred to this as co-indexical identification. In this case as well, the sedimentation of identification processes can allow for the emergence of an identity that is intrinsically social because it emerges from the manipulation of already stabilized communicational roles.
In other words, co-indexical identification and identification processes are two ways in which an identity-related activity manifests itself through a semiotic process of meaning management.17 To differentiate these two forms of identification and their role in identity formation, we can summarize as follows. The processes of identification concern the personal emergence of an identity-related activity from signs that express a management between experiential semiosis and other types of semiosis: gestures, postures, reactions, lapses, and attitudes. The sedimentation of these management processes can in any case produce an identity endowed with a social range and linked to multiple types of situations.
In contrast, processes of co-indexical identification are linked to intersubjective semiotic productions and manipulate already existing social roles through complex alignment procedures and a dramaturgy of identities situated between the sphere of enunciation and that of communication domains. In short, these are two semiotic processes of identity emergence, which obviously overlap and feed on each other or produce conflicts.
Conclusions: Identification and Enregisterment in Semiosis
In this article, I have proposed a semiotic examination of identity, attempting to describe its functioning in relation to the epistemologies developed by continental semiotics and linguistic anthropology. Through a critical discussion of video game experiences linked to The Last of Us saga and to the controversies of its reception on digital social networks, I have presented three linked concepts: identification process, co-indexical identification, and identity. The first concerns the management of the disproportion between lived experience and language productions that take place in any semiotic activity. The second is expressed by means of enunciation and entextualization on the public stage, through a critical condensation and contextualization of competing identities already inhabiting one or more social domains, and an ideological and axiological positioning in relation to them. The third emerges from the sedimentation, or enregisterment, of identification and processes of co-indexical identification that are capable of characterizing and transforming the enduring activity of a specific identity.
In this final section, I would like to discuss the range of these operations and the research axes that open up with respect to identity within the frameworks of continental semiotics and linguistic anthropology. With respect to the first point, identification processes extend far beyond the realm of video games. Even while reading a novel, a character may unsettle us, make us think that their actions are exaggerated or overly virtuous, that their perspective is controversial, or that, all things considered, we understand them—even if we would have acted differently in their place. As in the case of The Last of Us, semiosis allows us to experience and uncover fragments of our own identity, even in the context of the reception of a textual form that does not require the active construction of a ludic enunciation.
The same happens in practical situations that are not characterized by such a clearly defined interactional framework. If we’re walking in a park and come across a group of individuals fighting, and one of them is a child, we might hesitate about what to do. Should we intervene or move on? What’s the best way to intervene? Should we talk or cry out? Should we physically protect the child? Here too, our action or inaction, as well as our hesitations, allow us to experience fragments of our identity through the management of the pressures exerted on us by the practical situation and our potential reactions or enunciations.
Finally, even in the case of a fully enunciative situation, this pressure must be managed in accordance with an identification process. If we have to speak during a talk at an academic conference, for example, our enunciations take place under the pressure of a specific type of social situation, which requires the personal execution of shared scripts: staying within the allotted speaking time, maintaining an appropriate rhythm, responding convincingly to criticisms and questions. Under this social pressure, we may experience slips of the tongue, hesitate, nervously repeat ourselves, show embarrassment, or display arrogance. These are all signs that reveal how we are managing, in real time and in a personal manner, the realization of the actualized social role—that is, how we carry out an identification process involving the relationship between experiential and verbal semiosis. Drawing from these three examples, we can conclude that identification processes are linked to all forms of semiosis.
The same applies to processes of co-indexical identification, which extend far beyond controversies over a single cultural object such as The Last of Us. The social identity of the Hollywood actor Johnny Depp, for instance, is made up of his bodily performances, which give rise to memorable characters (J. Sparrow, W. Wonka). We are here situated in a specific domain, that of fictional cinema, and this identification process includes—implicitly or explicitly—the performances of other actors, realized in different productions and potentially in different eras. But the social identity of Johnny Depp is also composed by trans-domanial enunciations such as press articles pertaining to him, or advertisements in which he has appeared. Lastly, his identity is also composed according to other social practices such as the recent trial against his ex-wife Amber Heard. The particularity of social identities is that they are simultaneously experienced by the individual who embodies and constructs them but are also observable socially. They can serve as anchors for different indexical enunciations: Johnny Depp can be indexed by various individuals as a brilliant actor (his fans, film critics), a potentially violent husband (his ex-wife), a man with a deep sense of irony, et cetera.
With respect to future research directions for continental semiotics and linguistic anthropology, I would like, in these final lines, to return to the relationship between enregisterment and enunciation. We have observed that in Agha’s analyses, enregisterment primarily pertains to ways of speaking, even though its definition more generally includes using signs. In this article, I have suggested that different players adopt variable styles as they appropriate the kinetic grammar of The Last of Us, to the extent that they constitute, through enregisterment processes, distinct ludic registers. In my view, the study of registers could be pushed further, and theorized through, multimodal modes of sign-making such as video games. What happens, for instance, when we examine the functioning of visual registers in the field of image production (see, e.g., Nakassis 2023)? How do different types of social domains—as well as the perceptibility and describability of registers themselves—take shape when we consider the praxis of image-making and world-making (as with video games like The Last of Us)?
Understanding registers as multimodal ways of enunciating that mediate and are constituted by both co-indexical identification and identification processes opens compelling avenues of research regarding the relationship between enunciation and social domains (in the sense used in Paris School semiotics to refer to domains of semiotic practice, such as art, advertising, science, politics; Dondero 2020:2), requiring a further analysis of the concepts of register, identity, and types of persons. In this context, do these types of persons correspond to types of artists or image producers (in the domain of art), types of audiences or game players (in the domain of gaming or entertainment), or perhaps individual “types,” that is, idiosyncratic persons who inhabit these social types in particular ways, ways that are recognizable to them? Or do they articulate these domains by creating a traffic and movement between them, where social types, as virtualities, are actualized (and then, in turn potentialized) across domains of semiosis?18 Further, do different media and modalities give rise to different kinds of registers? For example, what visual traits define a visual register—that is, ways of visual enunciating—that contrast from other types of registers (speech registers, cinematic registers, etc.)?
Ultimately, it seems to me that the question of identity, articulated through identification and enregisterment processes, occupies a constitutive central position in relation to semiosis, as it emerges precisely in the management of different orders of indexicality, different social domains, and different ways of producing meaning.
Acknowledgments. An earlier version of this paper was presented and generously commented on within the framework of the workshop series “Entextualization/Enunciation” dedicated to the relationship between continental semiotics and linguistic anthropology. I thank all the participants for the enriching discussions and, in particular, Maria Giulia Dondero, Tatsuma Padoan, and Constantine Nakassis for inviting me to participate. I am especially grateful to the latter for his in-depth commentary on my text, which I hope to have at least partially honored.
Endnotes
1. This article does not follow in the footsteps of the semiotics of video games but offers a general analysis of the experiences of meaning by considering exemplary video game cases. For a more specific discussion relating to the semiotics of games, see Thibault 2020; on the semiotics of video games, see Aarseth 1997 and Giuliana 2024.↩
2. In games studies, see Jørgensen 2009 and Klastrup 2009. For more discussion of the specific kinetic reading presented here, see D’Armenio 2014, 2022, 2024a. In this article, following their usage in continental semiotics, by the term “linguistic” I mean phenomena concerning verbal language (viz. langue) while by “language” (langage) I mean more or less systematically grammaticalized, or coded, semiotic systems (verbal language, visual language, multimodal systems such as the syntax of film editing, and the like).↩
3. In this article, I initially use the notion of identity in its everyday, vernacular sense. The idea is to approach a social field that already exists prior to its technical modeling. I thus speak of a semiotics of identity in the same way one can speak of a semiotics of culture or a semiotics of literature. As discussion progresses, I introduce two technical concepts: co-indexical identification and identification processes. Drawing inspiration from analogous concepts developed in linguistic anthropology—such as enregisterment and entextualization—these two technical notions aim to designate the processes through which the semiotic work of identity manifests itself. At the end of this discussion, I thus technically define identity, in accordance with a semiotic perspective, as the sedimentation of identification processes. ↩
4. Even if continental semiotics has addressed topics related to identity—organizations and institutions (Floch 2000), the body (Fontanille 2004; Marrone 2001), practices (Fontanille 2008), forms of life (Fontanille 2015), subjectivity and enunciative instances (Coquet 2007; Paolucci 2020)—this concept has not been integrated into its epistemology and has therefore only been addressed indirectly or in relation to specific fields or planes of relevance. By contrast, in the field of linguistic anthropology, the concept of identity plays an important role, as a consequence of the metapragmatic dimension which fuels its epistemology. The concept of enregisterment—the process through which enunciations conjure and collectively produce ways of speaking, or registers, by associating them with types of persons—has a clear relevance to the concept of identity that The Last of Us raises. The challenge here is to expand the notion of enregisterment (and related concepts, such as role alignment) to gaming practice.↩
5. The specialized press also often uses the broader label of action-adventure to indicate some of the main ludic activities that characterize this genre, as well as the sub-genre of survival horror. ↩
6. https://www.metacritic.com/game/the-last-of-us-part-ii/user-reviews/?platform=playstation-4, accessed on October 30, 2024. I have left typographic and spelling errors in the original uncorrected. The controversy does not only concern online user reviews and opinions on social networks and specialized forums, but also involves articles in video game magazines and less sectorial online portals such as Rolling Stone and Forbes. The developers themselves have intervened with posts on social networks to show the threats they have received. ↩
7. In a post on Twitter/X, Neil Druckmann, writer and director of The Last of Us saga, exposes some of the messages he received: https://twitter.com/Neil_Druckmann/status/1279841603843051520 accessed on November 2, 2024.↩
8. This approach makes it possible to identify as actualized a register that inhabits contemporary society (e.g., a lawyer’s jargon)—a register which could thus be realized in concrete events of speech—while a potentialized register would pertain to typical figures and ways of speaking of a bygone era that may not be relevant (actualized) in any particular moment but could be called upon (e.g., the formulas used in Morse code). The enregisterment process also makes it possible to produce more detailed analyses of particular enunciative styles, as with tropes like irony. As an enunciative style, irony is normally described as a posture that works not according to what is denotationally asserted (the first-order function of the utterance), but to the way in which it is asserted (as a second-order meaning). Registers are clearly another way of producing tropic effects: describing a phenomenon, for example the institutional steps required to form a government, using a register from another part of social life, for example childbirth as described using medical jargon, or even enunciations that adopt a particular register in a very strict way, are often productive of various pragmatic effects (e.g., comedy). ↩
9. This process seems relevant to the considerations proposed by Adi Hastings and Paul Manning (2004) with regard to the relationship between selfness and otherness. They encourage a shift from focusing primarily on authentic self-expression towards understanding how language can operate to index, mimic, and negotiate differences, thereby constructing both identity and alterity within various social contexts. ↩
10. In my view, any enunciation is specific with respect to a particular semiotic system (or “language”, in the terms of continental semiotics). The way verbal language builds meaning is different than that of still images, for instance. Of course, a given utterance, whether verbal or visual, expresses signifying connections of a general kind, but the way verbal language “says” something is different than that of other systems. The semiotics of discourse has perhaps overestimated the importance of specific grammars and textualities while neglecting the fact that a semiosis is never isolated from the chain of successive and preceding semioses. However, it seems that the opposite risk also exists: due to the fact that semiosis is constituted of correlations between utterances, social domains, and specific groups, the relatively specific way of building knowledge and semiosis through specific languages may be overlooked. The two types of analysis seem necessary for an epistemology interested in all manifestations of meaning-making. The challenge is to integrate them in a non-modular way. ↩
11. I refer to the definition of diagram proposed by Charles Sanders Peirce (CP 2.277): sensible representations that in their internal relations replicate, often in an initially unperceived way, the relations pertaining to another phenomenon. This is the case, for example, of a map that selects some characteristics of the physical space, such as distance and spatial extension, and expresses them visually by replicating some relations of the “real” object. But it is also the case of geometric demonstrations, with which it is possible to manipulate graphical marks in order to build new knowledge through visual icons instead of categorical concepts. Here, the relations between the plastic movements made on the interface are translated into the system of kinetic relations expressed by the movements in the virtual worlds. It is the specific way through which the user accesses the semiotic experience and participates in its articulation. ↩
12. Goffman writes: "When an actress takes on the stage part of Celia Coplestone, she animates a make-believe person, a stage character. By using much the same physical configuration — her own body — she can, appropriately attired, project entities of other realm status; a historical personage, a goddess, a zombie, a vampire, a fleshy mechanical woman. And, of course, if an actress does her voicing from behind the scenes, she can animate configurations not her own: a ghost, a stuffed animal, a loquacious chair, and so forth. These various configurations which an actress (or actor) can animate need a generic title: call them figures." (1986:523) ↩
13. Other figures that Goffman (1974:529–37) distinguishes include cited figures, print figures, and mockeries and say-fors. Here, we recall Judith Irvine’s (1996) argument that Goffman’s participation frameworks are essentially open ended insofar as novel participant roles (and here we would add types of figures) can be generated through their reticulation, lamination, and elaboration in discourse; these can be conventionalized/institutionalized in particular sociocultural contexts and media, as we’ve seen, for example, in the case of The Last of Us. ↩
14. For a deeper analysis of the concept of animation, see Manning and Gershon 2013 and Silvio 2019. For a richer discussion of the semiotic logic pertaining to the animation of characters, see Nozawa 2013. ↩
15. In Ricœur’s theory of identity, the focus is on constructing a narrative mediation between two identity poles. On one side, identity-idem refers to permanence, being articulated through physical appearance, character, ingrained habits, and more broadly, all that enables the recognition of a person as the same over time. On the other side, identity-ipse concerns behavioral innovation—that is, ethical conduct, the figures one considers as moral guides, and the reprogramming of values throughout a life trajectory. Within this framework, identity is thus the result of a long-term personal negotiation that can provide, or fail to provide, a coherent form—through self-narration—to the relationship between permanence and innovation. Jean-Marie Floch (2000) adapted this theory to the analysis of brand advertising, emphasizing how utterances can effectively construct an identity-related narration not only verbally but also multimodally. Compared to these theories, my approach neither seeks to encompass a life span nor to limit itself to narrative mediations: the concept of identification concerns the real-time management of the relationship between experience and language, through which identity work manifests and becomes intersubjectively observable. The identities that result from the sedimentation of identification processes are not necessarily built narratively nor necessarily through a semiotic production objectified in some medium. That said, it seems to me that one of the most important ways in which the relationship between experiential semiosis and linguistic semiosis is negotiated is through the act of narrating. ↩
16. https://www.thegamer.com/the-last-of-us-players-share-how-they-dealt-with-surgeons-ellie/, accessed May 17, 2025 ↩
17. On the subject of the formation of social identities from identification processes, see the important contribution of Wortham (2005), in which the emergence of the identity of two students with respect to the field academic learning is analyzed over the course of the year. ↩
18. Here, I use the terms virtual, actualized, and potentialized in the sense defined with discussions of enunciatve praxis; typically, in accounts of enunciative praxis the process of actualizing some virtual form (e.g., an artistic motif) is understood in historical, temporal terms, as when a stylistic motif from the past (as a virtual possibility within the cultural memory of artistic practice) is actualized (or made available) in some present (e.g., a Renaissance motif used in contemporary art or advertising), wherein it may end up being deployed, or realized (i.e., recontextualized), in some text. Here, however, I am also interested in how co-eval relations between domains and their semiotic registers (e.g., between art practice and advertising; between gaming and cinema; etc.) stand in relations of virtualization (and thus also actualization and potentialization) with respect to each other. ↩
References
Aarseth, Espen. 1997. Cybertext: Perspectives on Ergodic Literature. Baltimore, MD: Johns Hopkins University Press.
Agha, Asif. 2005. Voicing, Footing, Enregisterment. Journal of Linguistic Anthropology 15(1):38–59.
Agha, Asif. 2007. Language and Social Relations. Cambridge: Cambridge University Press.
Basso Fossali, Pierluigi. 2017. Vers une écologie sémiotique de la culture. Perception, gestion et rappropriation du sens. Limoges: Lambert-Lucas.
Benveniste, Émile. 1970. L’appareil formel de l’énonciation, Langages 17:12–18. https://www.persee.fr/doc/lgge_0458-726x_1970_num_5_17_2572
Bertrand, Denis. 1993. L’impersonnel de l’énonciation. Praxis énonciative : conversion, convocation, usage. Protée 21(1):25–32.
Calleja, Gordon. 2011. In-Game. From Immersion to Incorporation. Cambridge, MA: MIT Press.
Coquet, Jean-Claude. 2007. Phusis et logos: Une phénoménologie du langage. Paris: Presses Universitaires de Vincennes.
D’Armenio, Enzo. 2014. Mondi paralleli: Ripensare l’interattività nei videogiochi. Milano: Unicopli.
D’Armenio, Enzo. 2022. Beyond Interactivity and Immersion. A Kinetic Reconceptualization for Virtual Reality and Video Games. New Techno-Humanities 2(2):121–29. https://doi.org/10.1016/j.techum.2022.04.003.
D’Armenio, Enzo. 2024a. From the Ontology of Video Games to the Epistemology of Digital Movements. Towards a Semiotics of Virtual Practices. Games and Culture. https://doi.org/10.1177/15554120241263630.
D’Armenio, Enzo. 2024b. The Mediatic Dimension of Images: Visual Semiotics Faced with Gerhard Richter’s Artwork. Visual Communication 23(3):473–97. https://doi.org/10.1177/14703572221080529.
Dondero, Maria Giulia. 2025. Enunciative Praxis and Enregisterment in the Domain of Generative Artificial Intelligence. Semiotic Review, 12. https://doi.org/10.71743/bwf80g96.
Dondero, Maria Giulia. 2025. Enunciative Praxis and Enregisterment in the Domain of Generative Artificial Intelligence. Semiotic Review, 12.
Eco, Umberto. 1975. Trattato di semiotica generale. Milano: Bompiani.
Ferri, Gabriele. 2015. Narrative Structures in IDN Authoring and Analysis. In H. Koenitz, G. Ferri, M. Haahr, D. Sezen, and Tİ Sezen, eds. Interactive Digital Narrative: History, Practice, pp. 77–90. New York: Routledge.
Floch, Jean-Marie. 2000[1995]. Visual Identities. London: Continuum.
Fontanille, Jacques. 1989. Les espaces subjectifs. Introduction à la sémiotique de l’observateur. Paris: Hachette.
Fontanille, Jacques. 2004. Soma et sema: Figures du corps. Paris: Maisonneuve et Larose.
Fontanille, Jacques. 2006[1998]. The Semiotics of Discourse. New York: Peter Lang.
Fontanille, Jacques. 2008. Pratiques sémiotiques. Paris: PUF.
Fontanille, Jacques. 2015. Formes de vie. Liège: Presses universitaires de Liège.
Giuliana, Gianmarco T. 2024. Il videogioco come linguaggio della realtà: Introduzione a una nuova prospettiva semiotica vol. 1. Milano: Aracne.
Goffman, Erving. 1981. Forms of Talk. Oxford: Basil Blackwell.
Goffman, Erving. 1986[1974]. Frame Analysis. Cambridge, MA: Harvard University Press.
Greimas, Algirdas Julien. 1989[1984]. Figurative Semiotics and the Semiotics of the Plastic Arts. New Literary History 20(3):627–649.
Greimas, Algirdas Julien and Courtés, Joseph. 1982[1979]. Semiotics and Language: An Analytical Dictionary. Bloomington: Indiana University Press.
Hastings, Adi and Paul Manning. 2004. Introduction: Acts of Alterity. Language & Communication 24(4):291–311.https://doi.org/10.1016/j.langcom.2004.07.001.
Irvine, Judith. 1996. Shadow Conversations: The Indeterminacy of Participant Roles. In M. Silverstein and G. Urban, eds. Natural Histories of Discourse, pp. 131–59. Chicago: University of Chicago Press.
Jørgensen, Kristine. 2009. “I’m overburdened!” An Empirical Study of the Player, the Avatar, and the Gameworld. In Proceedings of DiGRA 2009 Conference: Breaking New Ground: Innovation in Games, Play, Practice and Theory. https://dl.digra.org/index.php/dl/article/view/374/374.
Klastrup, Lisbeth. 2009. The Worldness of EverQuest: Exploring a 21st Century Fiction. Game Studies 9(1):1–15.https://gamestudies.org/0901/articles/klastrup.
Manning, Paul and Ilana Gershon. 2013. Animating Interaction. HAU: Journal of Ethnographic Theory3(3):107–37.
Marin, Louis. 2002[1993]. On Representation. Stanford, CA: Stanford University Press.
Marrone, Gianfranco. 2001. Corpi sociali: Processi comunicativi e semiotica del testo. Torino: Einaudi.
Nakassis, Constantine V. 2018. Indexicality’s Ambivalent Ground. Signs and Society 6(1):281–304. https://doi.org/10.1086/694753.
Nakassis, Constantine V. 2020. Deixis and the Linguistic Anthropology of Cinema. Semiotic Review, 9. https://doi.org/10.71743/2akse703.
Nakassis, Constantine V. 2023. Onscreen/Offscreen. Toronto: University of Toronto Press.
Nakassis, Constantine V. 2025. (Im)personalizing Enunciation. Semiotic Review, 12. https://doi.org/10.71743/k99f5r76.
Nozawa, Shunsuke. 2013. Characterization. Semiotic Review, 10. https://doi.org/10.71743/rn8eyq26.
Padoan, Tatsuma. 2025. Ritual as Enunciative Praxis: Some Reflections from Katsuragi, Japan. Semiotic Review, 12. https://doi.org/10.71743/3sm36r97.
Paolucci, Claudio. 2020. Persona: Soggettività nel linguaggio e semiotica dell'enunciazione. Milano: Bompiani.
Peirce, Charles Sanders. 1931–1935. Collected Papers of Charles Sanders Peirce, volumes 1–6, Cambridge, MA: Harvard University Press.
Pinotti, Andrea. 2020. Towards An-Iconology: The Image as Environment. Screen 61(4):594–602.
Ricœur, Paul. 1992. Oneself as Another. Chicago: University of Chicago Press.
Saussure, Ferdinand de. 1995[1916]. Cours de linguistique générale. Paris: Payot et Rivages.
Silverstein, Michael. 1993. Metapragmatic Discourse and Metapragmatic Function. In J. Lucy, ed. Reflexive Language, pp. 31–58. Cambridge: Cambridge University Press, pp. 31–58.
Silverstein, Michael. 1996. Indexical Order and the Dialectics of Sociolinguistic Life. SALSA 3:266–95.
Silverstein, Michael. 2003. Indexical Order and the Dialectics of Sociolinguistic Life. Language and Communication 23(3–4):193–229.
Silverstein, Michael. 2004. “Cultural” Concepts and the Language-Culture Nexus. Current Anthropology 45(5):621–52.
Silverstein, Michael. 2019. Texts, Entextualized and Artifactualized: The Shapes of Discourse. College English 82(1):55–76.
Silvio, Teri. 2010. Animation: The New Performance? Journal of Linguistic Anthropology 20(2):422–38.
Silvio, Teri. 2019. Puppets, Gods, and Brands: Theorizing the Age of Animation from Taiwan. Honolulu: University of Hawaii Press.
Thibault, Mattia. 2020. Ludosemiotica: Il gioco tra segni, testi, pratiche e discorsi. Milano: Aracne.
Wortham, Stanton. 2005. Learning Identity: The Joint Emergence of Social Identification and Academic Learning. Cambridge: Cambridge University Press.