‘[U]se’ is a hopelessly ambiguous or wide word, just as is the word ‘meaning’, which it has become customary to deride.
(J.L. Austin, How to Do Things With Words)
Language, it is commonly held, describes or reports on some state of affairs. Statements can thus be evaluated as being either true or false depending on whether the depiction that they offer in fact corresponds to reality. This form of evaluation, whose seeming self-evidence is the source of its everyday currency, has also been the focus of philosophical speculation as to the limits and conditions of talking about things and the modes in which things condescend to be talked about. In investigating truth claims and verifiability, philosophers of language thus assumed themselves to be posing the question of language in general.
In the 1955 series of the William James lectures at Harvard University, philosopher J.L. Austin presented an alternative view which he took to be at least partially representative of wider currents and concerns in the philosophy of language. These talks, collected in book form as How to Do Things with Words, centre around the seemingly innocuous observation that there are more uses to which language can be put than the simple description of things or relations between things. It is indeed possible to do things with words: promising, commanding, purchasing, marrying and naming, for example, are actions and not descriptions. In saying ‘Sorry…’, one does not make a representation so much as an undertaking. To apologise is to perform an action. This means that questions about truth or falsity, granting their undoubted importance in certain philosophical contexts, are not the only, or the best, or the necessarily sufficient, approach to thinking about language. The realm of action offers another valid line of inquiry and indeed has led to the development of a field of research known as ‘speech act theory’.
Emphasising action is also the focus of Alexander Galloway’s recent contribution to videogame studies, Gaming: Essays on Algorithmic Culture. He suggests that we: “Begin like this: If photographs are images, and films are moving images, then video games are actions” (Galloway 2006, 2). What results is a theory of ‘gamic action’ in which both machine and human activity are integral to the composite form of the videogame. Paralleling Austin, Galloway too indicates a trend away from a straightforward notion of representation, but the argument here relates to images rather than words. In this case the term ‘gamic’ happens to be one of Galloway’s coinages, but if we in the expansive spirit of interdisciplinarity refer to a biology textbook it seems that the term has a history; it in fact derives from the Greek for ‘marriage’ and is an adjective indicating that something pertains to sexual reproduction.
Galloway’s term ‘gamic action’ is therefore doubly appropriate (or for the less charitable, doubly a bit of a stretch), as without the cybernetic feedback loop or ‘marriage’ of player and machine action, there is no play and no videogame. However in this respect, it may also be illuminating to re-pose the nature or ontology of the videogame as a marriage between image and action or perhaps a doing things with images. Referring to Space Invaders (Taito 1978), Galloway writes ‘The aliens are nothing more than a series of byte strips stacked together. This is math made visible’ (32). The ‘math made visible’ referred to here is therefore considered part—perhaps the most important and vital part, as it is an audio-visual medium under discussion—of the computer’s overall gamic action.  Just as Austin examines performative utterances in the course of delimiting a field in the philosophy of language that will become speech act theory, perhaps it is possible to amend Galloway’s programmatic statement for videogame studies and begin like this: if photographs are images, and films are moving images, then videogames are image-actions.
Galloway’s assertion that videogames are actions is indicative of a particular tendency in videogame scholarship that attempts to uncover the unique properties of the form by separating them from putatively representational media such as film and television. The idea is that once such specific and unique features are isolated, they will naturally be the focus of a relevant critical method. For ludologists, videogames are above all games; for Galloway, they are (gamic) actions; for Ian Bogost (2006), they are comprised of unit operations. Where such endeavours venture a general theory of what videogames are (an ontology of the form), a more Austin-inspired pragmatic approach would attend to how they are used—and in the case of an audio-visual medium such as videogames, used to create some unique aesthetic effects and structures.
Fully elaborating a correspondence between speech act theory and the image-actions of videogames would require the consideration not only of the former theory but also some general treatment of the relation of images and words. Although this would be an overly extensive endeavour in the current context, it is possible to draw from Austin’s work a starting point for talking about performative images based on a pragmatic sort of attitude eminently suited to a lecture series named for William James. It should also be noted that Austin himself, despite the title of his book and the subsequent development of speech act theory, was quite happy to include within his discussion non-verbal actions such as raising a hand to vote, nodding assent or kicking a goal.  Austin’s notion of performativity is thus wider than speech, and may be applied, in some respects and with due caution, to instances of both imagery and play.
What then does Austin do with words? Initially, he notes a distinction between two types of verbal utterance. The first type (designated as ‘constatives’) can be evaluated according to a true or false scheme—if someone asserts that the cat is on the mat, it is possible to check and see if this is in fact the case. The second, which turn out to be the utterances Austin is interested in, are what he dubs ‘performatives’. It is inappropriate to assign truth values to these utterances. Instead, Austin argues that they are actions and as such can be evaluated as successful or dysfunctional depending on whether or not the utterance satisfies and enacts certain conventional criteria. They are “subject to the usual troubles and reservations about attempt as distinct from achievement, being intentional as distinct from unintentional, and the like’”(Austin 1975, 110). Already, then, we have on the face of it a good fit for videogames, in which actions are typically evaluated in terms of whether they succeed or fail—either the player times the jump well, or Mario plunges into the pit—rather than whether they are true or false, a distinction which is generally held to be irrelevant. Austin alternately calls the former kind of evaluation the (un)happiness or the (in)felicity of an action.
Austin’s performative utterances can fail or be infelicitous in various ways and he clearly took some trouble in coming up with illustrations that would most pique the sensibilities of his Ivy League audience:
Suppose, for example, I see a vessel on the stocks, walk up and smash a bottle hung at the stem, proclaim “I name this ship the Mr. Stalin” and for good measure kick away the chocks: but the trouble is, I was not the person chosen to name it… We can all agree
(1) that the ship was not hereby named;
(2) that it is an infernal shame. (Austin 1975, 23)
The distinction between performative and constative is a matter of how each type of utterance is to be judged. For constatives, the appropriate standard is truth value, which is to be ascertained by procedures of verification. This, Austin asserts, has been a highly normative conception of the function (and indeed ontology) of linguistic utterances in the philosophy of language. By demonstrating the practical existence of multiple regimes by which utterances are judged (generalised in terms of ‘happiness’), Austin effectively opens a new field for inquiry into the nature of language and our relation to it. The cultural world already sanctions many alternative procedures for judging utterances, for recognising them as events rather than descriptions.
Performatives must be carried out by someone competent to enact them (such as an authorised celebrant to oversee a marriage ceremony, or a jurist to pronounce a legal judgement). They must be heard and recognised for what they are by the relevant people—it’s not much use for a policeman to say “I place you under arrest”, unless the suspect is nearby and in a position to be so prorogued. Austin calls this ‘securing uptake’ of the utterance by interested parties. The performative acts that Austin is honing in on, then, have a constitutively social character. These actions may not bring about a change in a physical state of affairs any more or less than they describe it (hence the oft-assumed difference between action and speech, doing and saying). Rather, such utterances serve to constitute and reconfigure social relations. In a similar fashion, the most an image-action can be said to do in a videogame may be to shift a small button and the distribution of some computer’s memory. Within the game’s own logic and conventions there may be great conceptual leverage when such objectively minute action proves decisive, whether happily or unhappily for the player.
The social character or aspect of performative utterances leads Austin into territory typically inclement to level-headed pragmatists—speculation about interiority and intentionality. Although issues related to securing uptake such as competence of authority may be evidenced by various socially sanctioned signifiers, Austin also attends to a more ambiguous species of infelicity in the form of feelings or thoughts. A performative such as an apology may fail in some sense, for example, if the person issuing it is insincere and in spite of their assurances does not in fact feel sorry for the events in question. Unlike cases in which the act itself is defective or invalid, in such instances the act is carried out satisfactorily, but the intention motivating it is suspect. Someone issuing a promise they have no intention of keeping is another apposite example—the promise is issued successfully, the speech act is performed. The outcome is certainly not ‘happy’, but the manner in which it is unhappy is different from a situation in which requisite conditions for successful performance do not obtain: the promissory note is lost, defective or issued by a party lacking in authority.
An interesting subset of these ‘insincere’ performatives are a class that Austin in fact excludes from his discussion early on as ‘non-serious’. These include performatives uttered under some kind of threat or duress—“we may even say the act was ‘void’ (or voidable for duress or undue influence)”—as well as utterances encountered in literary or dramatic contexts. It is, for Austin, specious to say that Walt Whitman enjoins the Eagle of Liberty to soar in the same way that two people may conclude the sale of a commodity, even though both are, technically, actions of the type under discussion:
a performative utterance will, for example, be in a peculiar way hollow or void if said by an actor on the stage, or if introduced in a poem, or spoken in soliloquy. This applies in a similar manner to any and every utterance—a sea change in special circumstances. Language in such circumstances is in special ways—intelligibly—used not seriously, but in ways parasitic upon its normal use—ways which fall under the doctrine of the etiolations of language. All this we are excluding from our consideration. Our performative utterances, felicitous or not, are to be understood as issued in ordinary circumstances.’ (Austin 1975, 22).
Subsequent commentators have looked askance at Austin’s exclusion and the ‘peculiar way’ in which such utterances are void. One gets the impression that Austin dips his toes in here but doesn’t much like the temperature. If performative utterances enact some given social convention, issues of power (duress) and the ‘sea change’ of poetic usage can be said to lie at the very origin of those conventions, as both provide vectors for the transmission of the collective behaviours and rituals of social control that underwrite individual speech events. Parasitic utterances may indeed be extraneous to the normal usage—in the sense that they delimit it.
It certainly seems strange to characterise as parasitic etiolation an apology inscribed in Hansard for State-sanctioned violence, or Whitman’s exhortation to the eagle of liberty to soar when compared, say, to my ordinarily worthless undertaking that I will indeed remember to buy some milk on the way home. Furthermore, the assertion that such usages are necessarily or always intelligibly parasitic upon normal speech acts may be hotly contested. Austin assumes this—rather appropriately, by fiat—but is it really so simple a matter to exclude parasitic utterances or determine ordinary circumstances? At the very least this would require that insincerity or parasitism signify in such a way as to be routinely distinguishable from normal usage. On the contrary however, insincerity may function precisely through its formal indiscernibility from sincere actions: as the open secret of a power relation or the licence of poetic address. While Whitman’s imperative to a metaphoric eagle may seem parasitic upon the normative situation of issuing a command or request to a real person (or eagle for that matter) who is present at the time of utterance, Swinburne’s to Whitman is both parasitic and sincere performative utterance, both licentious poem and sober solicitation of real action from a real human being:
SEND but a song oversea for us,
Heart of their hearts who are free,
Heart of their singer, to be for us
More than our singing can be;
Ours, in the tempest at error,
With no light but the twilight of terror;
Send us a song oversea! (C.A. Swinburne, To Walt Whitman in America)
Whether or not such instances of multiple register of address and framing suggest grounds for a substantive critique of How to Do Things With Words (or speech act theory in general) has been the topic of some debate. Here might be traced a theoretical horizon along which the parasitic is not arbitrarily excluded, but somehow integral. Perhaps part of the dilemma lies in the fact that poetic and political performatives are inaugural and foundational events, whereas Austin’s performatives derive their consistency and force from established social formulas—they are events insofar as they are satisfactory repetitions.
Whatever the ramifications for a theory of language use, there are two main points arising from Austin’s doctrine of performatives that have practical relevance for our attempt to characterise image-actions. First, the exteriority of performative utterances themselves, and alongside them that of signifiers of parasitism, are emphasised. The ‘happiness’ criterion that Austin advocates for the evaluation of performative utterances ought to distinguish them from the ‘truth’ criterion of constative statements. Performatives, therefore, are not judged according to a speaker’s assumed intention to have the meaning of their statement correlate to reality. Indeed, regardless of subjective intention, the performative is judged according to how well it invokes conventional, ritualised formulas. Such judgements rely less on interpreting the inner ‘meaning’ of the utterance, and more on relational, anaphoric, deictic and diacritical signs which establish, disseminate or rehearse criteria such as competence, authority, occasion and whatever other factors impend on the felicity of a given speech act. This is in fact Austin’s starting point:
It has come to be seen that many specially perplexing words embedded in apparently descriptive statements do not serve to indicate some specially odd additional feature in the reality reported, but to indicate (not to report) the circumstances in which the statement is made or reservations to which it is subject or the way in which it is to be taken and the like. To overlook these possibilities in the way once common is called the ‘descriptive’ fallacy (Austin 1975, 3).
Indication of circumstances, reservations and intentional attitudes involve the use of words which may not have a strictly grammatical function, but rather invoke socially immanent ritual requirements. So, in order to successfully perform an action, the utterer must adopt the socially sanctioned formula—the role the utterance is intended to enact or fulfil within a given speech community—and it is this collective intentionality which informs the structure to which the speech act must accede if it is to be judged felicitous. This sort of exteriorisation of intent is also acknowledged in Austin’s requirement for ‘uptake’ by relevant parties. Rather than an encoding/decoding model in which two interiorities communicate through external signs, the performative structures utterance and interpretation, production and consumption, all of which can be more or less felicitous. In what amounts to a broadly legalistic theory of language use, form constructs content as something that is precisely what it is because another (the requirement for uptake of the utterance) is duly informed. What you say governs what you mean and the letter is taken over the spirit—just as a shot that sails over the crossbar isn’t declared a goal simply because the player meant for it to crack into the netting. Notably, this calculable and judicial impartiality is reminiscent of how a computer judges the actions of a player, with right of appeal paralleled and conditioned by the last time he or she remembered (or had opportunity) to save the game.
The second relevant point I wish to raise correlates to the first and regards Austin’s exclusion of parasitic utterances. As noted, the fact that such utterances signify their parasitism—that they are intelligibly hollow or void—is a condition of possibility for their exclusion. The class of signifiers that designate, intelligibly, that what we are perceiving is an illusion, are devices that frame or punctuate a given presentation. A title, a proscenium, a generic convention, an index, an admission price, a caption, a picture frame, diacritical marks: these disparate signs operate at the margins and interstices of artistic performance in order to orient audiences. Such framing devices may thus help to provide a theoretical passage from performative utterances to the specific case of image-actions.
Consider that Austin’s synchronic analysis of language construes normal and parasitic usage as a relation of copy to original. Insincere usage essentially operates, above all, by representing sincere uses of performative procedures. In addition to the objections previously noted, then, such a conception is odd in this context because the overly simplistic descriptive fallacy is supposedly what Austin is taking to task through his notion of the performative. In How to Do Things With Words parasitic utterances are ultimately judged according to the veridical regime of constatives rather than in terms of performative felicity. Truth value returns in a secondary role in the form of the intelligible signification of parasitism. However we all surely know from experience that artistic performances can be more or less successful, and their audiences can emerge happy or unhappy in full knowledge of a want of truth value. Accordingly, rather than an absolute demarcation between framed and unframed situations (as if an unframed statement is even a potentiality), the analysis of performatives may in some sense be extended to the judgement of parasitic utterances. In place then of Austin’s immobile distinction may be elaborated a typology of framing devices which, by virtue of their intelligible signification of intentionality, their exteriorisation of putative inner attitudes, disseminate structures of performative judgement. Insincere perhaps, but this is the insincerity of play.
Total Image Action
In order to further explore How to Do Things With Images, then, we will have to expand the discussion beyond Austin’s limitations and consider the way parasitic signs serve to ‘exteriorise’ various forms of intentionality. These limitations—the titular focus on words and the exclusion of parasitic utterances—are contestable on both methodological and historical grounds. Austin himself indicates that he was aware of the provisional status of his own conceptual boundaries, although he surely considered them necessary to circumscribe his argument. Regarding his initial goal of distinguishing performative and constative utterances, he admits that this would be possible only by isolating a given utterance from its circumstances and function. Near the conclusion of his lecture series, Austin admits: “Perhaps neither of these abstractions is so very expedient: perhaps we have here not really two poles, but rather an historical development” (1975, 146). By contrast, the historicity of image-actions can hardly be relegated to an afterthought as they are profoundly implicated in contemporary approaches to mass culture, motion and perception.
Austin appeals to a particular enunciative position in presupposing that the criteria for the judgement of performative utterances are immanent in a given speech community, always already disseminated through education and socialisation—in short, institutional forms which underlie the legalism of his theory. The focus is shifted to the context of an utterance, its role within a greater system that exceeds it, “The total speech act in the total speech situation is the only actual phenomenon which, in the last resort, we are engaged in elucidating” (Austin 1975, 148). What then can be said of a ‘total image action’? In the case of these performative images (as we have so far explored them) Austin’s default position is unavailable as the dissemination of criteria for judgement cannot be taken for granted. Videogames, in particular, demand constant re-assessment of audio-visual information on the part of players which amounts to the active apprehension of new rules in a dynamic environment. The attribution of players’ capacity to negotiate such complex virtual architectures to any naive naturalism or to an inherent capacity to intuit rules as if they were computers seems weak in comparison to the theory that performative images somehow signify their own criteria of judgement—the structures that govern success and failure, felicitous and infelicitous play. The total image actions in videogames require uptake by both a player and a technical apparatus. This might, to further paraphrase Austin, be termed the ‘total image situation’. The mechanisms by which uptake is assured would have to include both intertextual and contextual factors (generic forms, languages and visual codes), but also the aesthetic properties of particular image actions. It may be conjectured then that the parasitic, as a class of signs which (like the artistic frame) are by definition able to render intentionality intelligible, serve the same function in videogames.
If this hypothesis has merit, there ought to be some kind of correlation between types of image-actions and videogame genres because game genres tend to be organised around activity. While How to Do Things With Words may not be immediately useful for the consideration of parasitic or liminal signs, it does provide a potentially useful typology of action. Austin’s pragmatism is not just for show: he is level-headed enough to suspend the terms which have anchored his discussion once they seem to have led to impasse:
we were not always going to find it easy to distinguish performative utterances from constative, and it therefore seemed expedient to go farther back for a while to fundamentals—to consider from the ground up how many senses there are in which to say something is to do something, or in saying something we do something, and even by saying something we do something (Austin 1975, 94).
Rather than assuming there are only two distinct ways in which we formulate and judge utterances (as veridical or performative) or trying to examine what such statements are, Austin proposes a broad and nuanced spectrum of intentional performative utterances, of which the intent to describe is one particular variety.
It is fair to say that the resulting typology—with its unwieldy terms like ‘verdictives’ and ‘behabitives’—lacks the invigorating brio of the earlier lectures. These classes of action certainly seem remote from the context of videogames. Of greater interest is Austin’s shift of attention from what we do with words to how we do so. He nominates three classes: locutionary acts (in which to say something is to do something), illocutionary acts (where in saying something we do something) and perlocutionary acts (achieved by saying something). To perform a locution is to produce intelligible sounds, and is the basis of all such acts. Illocutions are locutions that carry a certain force in themselves due to social convention. Perlocutionary acts may achieve some given aim by various locutions and illocutions. Warning someone, for example, may be either an illocution or a perlocution. A warning may be carried out in the act of uttering the formula, saying “I warn you…”, issuing a yellow card in soccer or erecting a road sign, and would thus be an illocutionary act. A warning could however also be issued by an extended lecture or set of arguments, or relating a cautionary tale, or by shouting “Look out!” None of these latter actions are formal warnings which make use of a given formula, but the act of warning is still carried out by their performance—they are perlocutionary acts.
This distinction maps in an interesting (if admittedly somewhat impressionistic) fashion to that made by semiotician David Myers in his analysis of signs in gaming. Myers nominates two major videogame genres—action and adventure/RPG—by virtue of the fact that they correspond to two forms of semiosis: “The action genre is characterised by its use of first-order, denotative signs and the signification process that gives denotative signs values and meanings… The adventure genre is characterised by the use of second-order, connotative signs” (Myers 2003, 28). The adventure genre was eventually replaced by more complex roleplaying games. “Other common game genres—simulation, wargame, strategy—have histories long predating the history of electronic games, yet are, among electronic games, semiotically derivative of the action and role-playing genres”. In this scheme, games are designated by their tendency to use signs that either denote objects to be manipulated (action) or connote their context as functions within the game’s own logic (adventure). Thus, where a gun graphic in an action game denotes the physical properties and operation of a firearm, a gun in a RPG connotes a system, a certain semantically or statistically determined range of effectiveness. If we reconsider these genres from the point of image actions, the potential parallel becomes clearer. In an action FPS, the player pulls the trigger on their controller or clicks the mouse button, and in that action discharges their weapon. The denotative sign of the gun corresponds to a gaming version of the ‘illocutionary act’, and the felicity of the action is decided in the moment. On the other hand, a Final Fantasy X (SquareEnix 2001) player can command one of the characters in their party to strike a foe by selecting an appropriate icon. The character will act accordingly once their turn comes up, the attack’s degree of success being a function not of tactile adroitness so much as decisions made outside the individual action (equipment used, character stats, ‘buffs’ or ‘debuffs’ which respectively enhance or diminish capability): a sort of ‘perlocutionary’ act.
From this vantage it is possible to propose a generative account of videogame image actions which will simultaneously complicate Myers’ semiotic model and elaborate on Galloway’s assertion that ‘games are actions’. All activity in a game can be assumed to consist at the most basic level of ludic acts, for which to do something is to interact with a game apparatus. Ludic acts, like locutions, are the building blocks of the other forms of image action. There is no direct analogy for this variety of act in Myers’ semiotic account, but rather than denoting or connoting things within the game world, ludic acts are associated with iconic and indexical functions of the apparatus itself and the most basic forms of player interactivity (a keystroke, mouse movement, pulling a trigger, swinging a Wiimote—the equivalent of phonemes in language). Illudic acts are interactions in which something playful is done, and are associated with denotative semiosis. These acts tend towards tactility and an identification of audio-visual experience with the field of action, the FPS probably being the most representative genre.
Perludic acts are interactions by which something playful is done and are associated with connotative semiotic processes. Perludic acts tend to be informed more by cerebral than visceral experience as they require careful weighing up of connotations: networks and systems of signs that may not be reliant on denotation, at least in the way this latter term is used in semiotics — number games would be an example. Typically the player will perform the illudic act of selecting an icon, which will trigger a more complex action within the game world. This perludic act may be composed of an ensemble of ludic and illudic acts: instructing a Sim to tidy their room or make a meal in The Sims (Maxis, 2000) results in the performance of a set of individual and interlinked activities. This is not to imply that perludic acts are necessarily performed at a slower tempo than illudic acts: professional players of Starcraft (Blizzard Entertainment, 1996) are well known to perform hundreds of actions per minute (apm),each of which would be a perludic act issued to a unit or group of units through icons and hotkeys. This is a pace of interaction at least as fast-paced as in an action ‘twitch game’.
The typology of image actions is based on the following correspondences: Performative Utterance Semiosis Image Action Primary Generic Association Locution Iconic-Indexical Ludic Act The Gaming Situation Illocution Denotative Illudic Act Action/FPS Perlocution Connotative Perludic Act Adventure/RPG
Primary Generic Association
The Gaming Situation
The associations of image actions with game genres are only weakly correlative, especially as gaming has become more complex and techniques have been blended by designers attempting to provide varied experiences (pace Austin: ‘not two poles, but rather an historical development’). For example, in Doom (iD Software, 1993), the majority of the player’s activity is illudic gunplay and movement, but the use of blue or red keys to access various restricted parts of the level is connotatively abstracted and perludic. More recently, much of the gameplay of Resident Evil 4 (Capcom, 2005) involves illudic acts in which the player guides their avatar’s weapon in an over-the-shoulder third person view. Movements of the controller are proportionally mapped onto the way the character’s aim moves across the screen, and the firearm is discharged simultaneously with a single button or trigger press. Both are illudic acts. On the other hand, over the course of the game the player also has certain context-sensitive opportunities to interact with the environment, such as leaping a low wall or using a ladder. These are generally enacted with a single button press which causes a considerably more involved set of actions than the illudic act of pulling a trigger —essentially, they activate a mini-cinematic in which control is briefly removed from the player. These sequences, which may also appear in boss fights as life-or-death tests of timing and skill, are perludic acts.
A second type of perludic act structures Resident Evil 4‘s weapon upgrade system. The player pays a mysterious shopkeeper for modifications to their weaponry, but they do not actually do any tinkering themselves. Thus the selection of upgrades is a perludic act. The ‘same’ act can be differentiated across titles: jumping is illudic in Super Mario Brothers 3 (Nintendo, 1985) as Mario leaps in response to the player pressing the appropriate button, whereas in Resident Evil 4 the jump is a vault to clear a low wall, the button for which is indicated by a context-sensitive icon on the screen. Illudic acts can flow into a single perludic act, such as in the hacking minigame of System Shock 2 (Looking Glass Studios and Irrational Games, 1999). By manipulating nodes in a series of illudic acts, players can achieve the perludic act of hacking a computer. Even within a single game, an act can be either illudic or perludic: in Fallout 3 (Bethesda Game Studios, 2008) the player can either use their weapons in the style of an action FPS or activate a RPG stat-based targeting system.
The focus on analysing games in terms of image actions is thus capable of presenting a variegated and nuanced account of gaming experience as both player agency and ludic structure. Each game constructs and elaborates on its regime of activity, and the fluidity of our ludic categories is in this light a merit (the system being more continuum or phenomenology than typology proper). This again raises the question of how the structures of felicity are communicated in what amounts to a total image act. Austin, faced with similar problems, is able to appeal to context, and analogously a good deal of explanatory mileage can be gained in the case of the gaming situation by the concept of genre. Players expect more illudic than perludic acts in a FPS, and vice-versa in a text adventure or a 4X game. However as I have pointed out, this is hardly a watertight correlation. It may then be surmised that in order to orient players to action and communicate ludic structures, games make use of parasitic signs. These most obvious example of such signs is the game interface itself. An action title will tend to minimise the presence of such signs through a HUD (heads-up display), or like the action-horror game Dead Space (Visceral Games, 2008), integrate the interface elements into the game world through a dose of sci-fi technobabble. By contrast, a screenshot from MMORPG World of Warcraft (Blizzard Entertainment, 2004) shows a plethora of customised icons and communications. If the presence of the frame indicates that a painting is parasitic, ‘hollow or void’, the disbursement of framing devices in games is similarly indispensable to the full characterisation of image actions. The salutary effect though is that narrativity and rule structure can be thought in a single flux, rather than as elements demanding distinct forms of analysis.
This opens up a very complex terrain which can perhaps be adumbrated through looking at seminal moments in gaming history—Myers’ action and adventure games—and the ways in which these pioneering works dealt with framing. The first truly blockbuster videogame, Toshihiro Nishikada’s Space Invaders, made it impossible for observers to ignore the Japanese contribution to the videogame industry. In the early arcades science fiction was the most prevalent influence. Martin Amis’ emphatically titled Invasion of the Space Invaders: An Addict’s Guide to Battle Tactics, Big Scores and the Best Machines (1982) depicted the arcade phenomenon tout court as an assault from outer space, and was replete with pulp images of clashing starships that could only stand in stark contrast to the abstracted, blocky graphics of the time. No doubt it was easier to visually depict the void of space than the claustrophobic caverns, forests and dungeons common in fantasy narratives. Furthermore, the action-oriented nature of arcade games, which measured precise repetition of illudic acts against an abstracted ‘outer’ space (the screen and its invisible grid or visible maze) in pursuit of an abstract goal (the high score) was suited to science fiction. Space was subordinated to the task to be performed within it, as were the Invaders who came from nowhere and did nothing save invade. Amis’ association between arcade and science fiction was not absolute—Tim Skelly’s vector-based Warrior (Cinematronics, 1979) which pitted a pair of armoured knights against one another provides an influential counter-example—but even here and in the main, arcade games concerned the control of space by entities defined largely by their functional capabilities and roles in the ludic alchemy of reducing dispersed objects into homogeneous ‘points’ or whatever other currency. They appear from nowhere and plunge headlong into the nowhen of number and score, rupturing the frame.
Fantasy, by contrast, concerns lineage and lineage does not allow space or time to remain simply abstract. The various treasures to be found in Colossal Cave Adventure (Crowther and Woods, 1976) were indeed measured in terms of a numerical value—entering the command “score” brings up an evaluative “If you quit now you would score x out of a possible y”—but each object contributing to said value is discovered in a unique place (or potential series of places) and acquired by an appropriate contextualised, perludic action (as opposed to an illudic action which can be peformed regardless of context). Compared to the action genre with its sci-fi inflection, in which the punctual execution of a small repertoire of actions guaranteed survival and the domination of an abstract outer space, the fantasy-oriented adventure game conjured a more voluptuous world through textual description. Although Crowther and Woods’ prose was minimalist and functional, the exploratory aspect alone served to add depth to the experience. The game-space was not confined to one screen or set of screens: players had to map their position relative to landmarks, creatures, objects and inscribe their progress within the game’s spatial regime rather than punching a torturously derived three-letter moniker into the high-score list outside the timeframe of each individual game session. The longer play time of adventure games created a different sense of engagement than that of the constant histrionic demands of the arcade, a sense that the space was coyly yielding up its secrets rather than trying to overwhelm the player, and fostered distribution in institutional and domestic contexts rather than public and commercial space.
In each case, the total image action—an aesthetic, ludological and performative category—is a function of the relation of illudic and perludic acts to the frame. The action game punctures the frame—a paradox, because the frame’s coherency is part of the genre’s commitment to illudic illusionism. The adventure presents nested frames linked with abstract values and perludic acts, but simultaneously seeks to reinforce the magic frame of a coherent fantasy world. The gaming situation and the total image action seems to be stretched, like a canvas, between these vectors. In and by taking account of the range of gaming’s image actions, we stand to gain significant insight into How To Do Things With Images.
Amis, Martin. 1982 Invasion of the Space Invaders: An Addict’s Guide to Battle Tactics, Big Scores and the Best Machines. London: Hutchison.
Austin, J.L. 1975. How To Do Things With Words. 2nd Edition. Edited by J.O. Urmson and Marina
Sbisà. Harvard: Harvard University Press.
Bateson, G. 1972. Steps to an Ecology of Mind. New York: Ballantine Books.
Baudelaire, C. 2001. The Painter of Modern Life and Other Essays. Trans. J. Mayne. Phaidon Press: London.
Benjamin, W. 2006. Berlin Childhood Around 1900. Cambridge: Harvard University Press.
Bogost, I. 2006. Unit Operations: An Approach to Videogame Criticism. Cambridge: MIT Press.
Gadamer, H. 2004. Truth and Method. 2nd Revised Edition. London: Continuum.
Galloway, A. 2006. Gaming: Essays in Algorithmic Culture. Minneapolis: University of
Huizinga, J. 1970. Homo Ludens: A Study of the Play Element in Culture. London: Paladin.
Myers, David. 2003. The Nature of Computer Games: Play as Semiosis. Bern: Peter Lang
Nietzsche, F. (1973) Beyond Good and Evil: Prelude to a Philosophy of the Future. Harmondsworth: Penguin.
1 In Galloway’s taxonomy of action, this ‘making visible’ would be a ‘non-diegetic machine act’ which he sees as helping define and police or transgress the boundaries between a game and its outside.
2 Austin however does assume that speech acts are the ‘simplest’, paradigmatic or most straightforward examples, and the very term ‘locution’ along with the lectures series’ title leave no doubt that language is at the heart of his discussion.
3 Austin notes that a dissatisfaction with a purely representational (or for that matter purely grammatical) view in the philosophy of language considerably predates his own treatment. Wittgenstein is notable in this regard, but also the work of Immanuel Kant: First came the view, not always formulated without unfortunate dogmatism, that a statement (of fact) ought to be ‘verifiable’, and this led to the view that many ‘statements’ are only what may be called pseudo-statements. First and most obviously, many ‘statements’ were shown to be, as Kant perhaps first argued systematically, strictly nonsense, despite an unexceptionable grammatical form: and the continual discovery of fresh types of nonsense, unsystematic though their classification and mysterious though their explanation is too often allowed to remain, has done on the whole nothing but good… it was natural to go on to ask, as a second stage, whether many apparent pseudo-statements really set out to be ‘statements’ at all” (Austin 1975, 2).
 It is striking that the idea of a zone in which sincerity and insincerity are indistinguishable has extensive precedent in writing on play; commentators as diverse as Baudelaire (1995), Gadamer (2004), Huizinga (1970), Benjamin (2006) and Bateson (1972) have described the putatively frivolous activity of play in terms of a constitutive seriousness. Nietzsche goes so far as to make this attitude essential to maturity in one of his aphorisms: “Mature manhood: that means to have rediscovered the seriousness one had as a child at play” (1973, 94). The temporality posited in this re-orientation towards an anterior state acts to destabilise normatively linear notions of innocence and experience.
 In terms of rhetoric, irony in particular often does not signify its conflation of levels in terms of the meaning of the individual utterance, instead relying on contextual and intertextual positioning and structure to establish itself as a trope. In terms of politics and power relations, identical utterances are inevitably parsed by the social position of those uttering them, and many discursive formations are restricted in the first place to those privileged by education and status. A given class or other social group may only be able to speak in certain ways at all by the adoption of a form of ironic or overtly comic quotation.
 This legalism supports Austin’s insistence (presumably raised against positivism) that social conventions are ‘of course facts’, but also his assumption that in difficult or liminal cases ‘a judge could decide’ between various competing interpretations of the validity of a performative utterance.
Darshana Jayemanne researches and writes on videogame aesthetics. He is a Ph.D. candidate in the School of Culture and Communication at the University of Melbourne.