Most players agree that games can be art, yet act in ways that betray the fact that they do not personally approach games as art. For such players, it is as though the word ‘art’ is merely a badge that gets to be worn by things people particularly like. They see that some people seem to like novels quite a bit, and that those therefore get to be art. Well, they want to make it clear that others now like games a great deal, so of course they want games to be allowed to wear the badge too. I am certainly of the opinion that games can be art, but from my perspective the word ‘art’ does not refer to a vague and insubstantial category of preferred works.
In this article, I would like to zero in on this topic of games as art. My test case for this purpose will be the design-based puzzle game Infinifactory. This example is a very conscious choice on my part, as Infinifactory is one of the many games that I consider to be poised between conceptions of games as art and conceptions of games as not art. If that sounds strange or you’re already making assumptions about where I’m going with this, don’t worry: I’ll explain myself with considerable specificity in the sections that follow.
This article will have four primary parts. First, I will present a working definition of art that I consider to be both flexible and rigid enough to be tenable. Second, I will apply that definition to games in general. Third, I will justify my claim that players often approach games as though they are not art. Then fourth and finally, I will describe how one may conceive of Infinifactory as a work of art. In doing all of this in a systematic fashion, I aim to foster a more specific and concrete discussion of this topic than most of the extant debates and articles produced about it over the past few decades have engendered, as a small contribution to spreading the formalist revolution in the study of games to a wider audience.
A General Definition of Art:
It is my firm opinion that no one should enter into this conversation without presenting the definition of art with which they are operating. There is so much controversy and disagreement on the topic of whether certain categories of things are art or are not art, that to proceed into a discussion of this nature without first stating one’s accepted definition for art is to invite angry misunderstandings.
There are many low-level constraints that people generally accept when advancing a definition of art, as listed in the entry of the Stanford Encyclopedia of Philosophy about defining art. I will now distill those many constraints down into just two—which will necessarily differ in priorities and details from the longer list on that linked page. Here are my two requirements for a definition of art (which you may notice are abstract enough to be requirements for the definition of just about any word):
First, the definition must be flexible enough to encompass nearly all of the things which people would generally consider to be art. People disagree fervently about which objects of study are art and which are not art, about what is high-quality art and what is low-quality art, about what is even eligible to possibly be considered art, and so on. A strong definition of art will need to be wide enough to be appropriate for all such debates, and to be potentially applicable to all such candidates.
Second, the definition must be rigid enough to be used to distinguish art from non-art. For the purposes of casual conversation and brevity, it is all well and good to simply say ‘anything can be art’ or ‘art is anything that moves you.’ Under my definition, both of those sentiments are technically true. But in a formal context, leaving it at that very quickly makes the word inutile for all practical purposes. If all one says is that anything can be art or not art at any time for anyone (without addressing why), then there is no reason to use the word ‘art’ to refer to an object rather than simply expressing that one likes the object, dislikes the object, or at least finds the object interesting to pay attention to. Under a broad enough definition, the word ‘art’ would simply translate into the vague phrase ‘thing about which I have some kind of opinion.’ A strong definition of art will need to be narrow enough to ensure the word is not uselessly universal.
After much deliberation, here is the definition that I have come up with to satisfy those two requirements: Art is any experiential object perceived to have at least one discernible primary emotional or intellectual end outside of itself.
Let’s break that down. First, “experiential object.” In the course of this article, I will be regularly using the cumbersome phrase ‘experiential object’ to refer to any candidate that might be art—a painting, a film, a television program, a sport, a part of nature, an artisanal meal, a game, et cetera. It will allow me to refer to all such things that people experience, without first addressing whether they are art or are not art.
Second, “perceived to have at least one discernible primary [. . .] end outside itself.” Do not underestimate the importance of the word ‘perceived’ in that spot in the definition. Art requires an observer, and the experiential object toward which that observer’s attention is directed must seem to them to have extrinsic value in human terms. In most cases, this perceived end is an act of communication, consideration, or education. If you think I’m saying that a work of art can’t have intrinsic value, then you are mistaken; I would venture to say that most works of art have both intrinsic and extrinsic value.
Moreover, this distinction is essentially a reversal of the usual discussion of games that uses those adjectives, which is that of extrinsic and intrinsic motivation. The value provided by the experience and themes of a game do certainly play into a person’s motivation to continue playing, but the experience and themes of a game (which I’m splitting in this article into intrinsic and extrinsic varieties) both tend to fall under the purview of intrinsic motivation structures as they’re typically construed. The externally directed end or value of a work of art can provide its emotional or intellectual content through intrinsically motivated experiential progress or attention. Games with the best shot at being art to me have as their dominant strands intrinsic motivation and extrinsic value.
Finally, “emotional or intellectual.” I’m not trying to be overly exclusive here. Your favorite action movie or single-player first-person shooter probably qualifies. Don’t forget that this article will ultimately be about a puzzle game, after all. This is merely an attempt to capture what it is that distinguishes something as art for the observer. It is not enough that the work be perceived as not being an end unto itself; the end toward which it seems to be directed, in a work of art, should relate to matters of emotion or intellect.
You may be thinking, “What about truth and beauty?” Well, the gang’s all here. Works that seem to be attempting to access real truths count as having intellectual ends, regardless of whether accessing real truth is ultimately epistemologically possible. And works that seem to be primarily attempting to give people the experience of beauty count as having emotional ends, regardless of whether one thinks that beauty is a quality that inheres in objects or only exists in the mind of the observer.
And at last, the corollary: an experiential object is not art if it is perceived as an end unto itself, or its perceived end is not emotional or intellectual (e.g. an immediate physical or social benefit). Herein lies the rigidity of the definition, its practical utility, what it excludes. Basically, if the principal utility of engaging with a particular experiential object seems to be improving the immediate skills involved in experiencing the object itself (examples in the following section)—then, by my definition above, it is not art. An object which has a perceptible primary external end, but in which that end is not of an emotional or intellectual nature (such as an artisanal meal), is not art. And anything perceived as self-justifying or perceived as without a primary extrinsic purpose (such as a part of nature) is not art.
One of the most important aspects of my definition of art is the space it leaves for differing kinds of engagements with works. Its incorporation of subjectivity through the notion of perception means that it is not only conceivable but entirely expected and possibly even inevitable that one person may consider something to be art or approach something as art—which another person addresses as not art. It even leaves the door open for one person to convince another person that a given experiential object is or is not art. This definition provides a framework for determining what is art to each individual, without degrading itself into a worthless universality.
Games as Works of Art:
Alright, it is finally time to start talking about games, and how exactly they can be art. To begin in an interesting fashion, I am now going to quickly list off a few examples of categories of games which I personally do not consider to usually contain works of art, then explain this potentially controversial set of classifications.
For me, working with my definition of art, games that are typically not art include (but are not limited to): most competitive tabletop board games such as checkers and chess, most tabletop 52-card games such as Solitaire and Poker, most physical sports such as baseball and basketball, most competitive video games such as Street Fighter and PUBG, and most rhythm games such as Dance Dance Revolution and Beat Saber.
Now, why are these not art to me? You can probably guess if you were paying close attention in the prior section. The answer is that I perceive these games as primarily being ends unto themselves. When most people (myself included) actively engage themselves with the task of playing Awesomenauts, for instance, they have as their primary aim improvement and success at Awesomenauts. It is a closed loop. This is not to say there are no tangential benefits to players, such as development of physical coordination or socialization with other players, but merely that the ‘end’ toward which the game is a ‘means’ seems to me to be the game itself.
In a similar sense, a person who plays a great deal of checkers or solitaire is improving at checkers or solitaire more than they are improving at anything else. Any development of critical thinking skills that takes place is in service of improvement at playing the game. I can imagine an approach to checkers and solitaire that treats and analyzes them as art, as works that primarily elicit extrinsic emotional or intellectual ends, but it is unintuitive to me—it is not in accord with my perception of those games.
On the other hand, a person who plays a great deal of basketball may actually be getting enough physical, social, and/or monetary benefits out of the game to say that for them the game has an extrinsic aim that outweighs its self-justification. Those are indeed discernible external goals. But they are not principally intellectual or emotional goals, so those games still are not art.
It may be argued that works which seem to otherwise fail to meet my definition of art may nevertheless possess beautiful visuals which are as deserving of being called art as are most distinct works in the visual art fields. While I would agree that the visual artwork in a game is just about always art on its own, in this article I am not considering a game’s visuals as a separate element, but as one of the many details of the larger work that is the game. In that context, I would say that games in which the visuals provide an overriding portion of the game’s content (and thus ought to be considered as part of the primary end of the work) are rare, although they do exist. Abzu and Proteus are arguable examples of this rare breed.
Now I will list off some examples of categories of games which I personally consider to usually contain works of art. For me, working with my definition of art, games that are typically art include (but are not limited to): most single-player role-playing games such as Chrono Trigger and Dark Souls, most character-driven 3D platformers such as Psychonauts and Jak and Daxter, most narrative games such as The Walking Dead and Gone Home, and most single-player first-person-shooters such as Half-Life and Bioshock.
It is not enough that these games contain stories. Rather, these games tend to be art because they are evocative. They are not ends unto themselves, but (often through stories, but ideally through gameplay) they evoke intellectual or emotional ends. One can improve at one of these games as though one is playing a competitive title, and many of them have competitive multiplayer components—but as an observer of the primary single-player campaign content of these games, I perceive them as not being primarily self-justifying sport-like works. They have other ends, such as communication of themes, emotional entertainment, fostering of creativity, subjectification of other people, political argumentation, and philosophical consideration.
I suspect that many of you reading this at least feel that I’m not saying anything completely ridiculous on the topic of the categories listed thus far. But now let us consider a less clear-cut example, an edge case: a strategy game. This will really draw out the importance of the perception of the observer in my definition of art.
Edge Cases and Individual Perspectives:
The last matter to tackle in order to express my whole position on these topics before diving into Infinifactory is to justify the claims made in the introduction about the most common player interactions with games. Why do I say that I think most players approach most games as though those games are not art, despite their statements to the contrary? To answer this, I will discuss an example.
One of my most broadly misunderstood game articles in forum comments and elsewhere is my article on the deckbuilding-roguelike-strategy title Slay the Spire, in which I argue that the best experience of the game derives from playing it while using as few external resources (such as guides, videos, and wikis) as each player conceivably can. While these responses took many shapes, there is one overwhelming source of those misunderstandings that does not originate in the work: the fact that most of the responders were decidedly approaching their discussion of Slay the Spire as a discussion of something that is not art.
The crux of their response was that the game is a difficult and multifarious endeavor, which it is essentially impossible to master solely through consultation of external resources. They took umbrage at my comparison to puzzle titles, mistaking my comparison of strategy games to puzzle games as me saying that Slay the Spire can be straightforwardly solved, something I certainly do not believe (even speaking as someone who has 100% completion in the game). Thus, as no easy solution to Slay the Spire can be reached which will thoroughly carry every player all the way through to total mastery, and as players still must stay engaged throughout their playtime—they felt it was unjust of me to make an analogy to puzzle solutions, and to say content was being skipped and players were disengaging from the game to some degree by consulting external resources.
That response to my argument only has any validity, however, if one considers Slay the Spire not as a work of art, but instead as something like a sport, a musical instrument, or a competitive tabletop game. It is very telling that unironic comparisons of Slay the Spire to chess by these responders were in abundance. They were talking about the quest of players to get better at Slay the Spire, the process of which they saw as almost necessarily involving consultation of external resources. They saw Slay the Spire as an end unto itself, and thus saw improvement as a player of the game as the primary aim of playing it. Such a player could play Slay the Spire forever, watching streams, viewing guides, studying wiki pages, and having lengthy discussions about best practices in-game—slowly but surely making incremental progress toward . . . being exceptionally skilled at playing Slay the Spire. One could imagine a Rocky-style montage of them doing just that.
But if one approaches the game as a work of art, an experiential object perceived to have a discernible emotional or intellectual end outside of itself, then those objections are null. If one sees Slay the Spire as a work about fostering creativity, making difficult choices, experiencing an oppressive atmosphere, and overcoming mounting adversity above and beyond the value it has as a demonstration of self-justifying strategic skill at playing a particular card game—then any external action, however minor, that intervenes to reduce the required creativity, the difficulty of the choices, and the oppression and adversity of the game has weakened the experience of it.
Slay the Spire can still bring about those positive extrinsic consequences I’ve just listed when someone consults external resources, to be sure, but some measure less than it could have otherwise. Just as an engrossing novel can still engage a reader even if they study a chapter-by-chapter plot summary before they begin, only somewhat less than it could have otherwise. So, I argued that people should seek to minimize that kind of activity as much as possible, in order to undergo the game’s offered experience of devising strategies as undiluted as possible.
Seen this second way, one would never think of playing Slay the Spire forever; one would trace the path from their first encounter with the game as a new player to their attempts at beating its primary end goals in much the way that one would experience any given film, novel, or RPG: unless deeply enamored or analytically interested, probably just once in one’s lifetime in order to observe it as a work of art before moving on to other works.
Now, here’s the key: were those responders wrong about Slay the Spire? No, they were absolutely not wrong about it. They were only wrong to attempt to turn their perceptions into objections to the article. Whether or not they were aware of it, they were perceiving the game as an end unto itself, as an experiential object that is not art, and their remarks were entirely consistent with that point-of-view. People are absolutely free to perceive, play, and discuss the game in whatever way makes the most sense to them; I was merely advocating for my opinion of the best way to experience the game as art.
Now extrapolate out from these specific details, and you can see why I made that claim in the introduction: regardless of their instinctive or defensive proclamations that games can be art or that games are art, most people approach most games with the perspective that one approaches something that is not art.
The thematic details of the work fall away, and players become increasingly concerned with maximizing skill within the game. But this is not necessarily an indictment of that perspective. I probably play most games that way myself, especially when it comes to titles that require significant mechanical skill. You’ll find that perspective in many of the more conventional game reviews I’ve written for The Gemsbok. Loads of games seem to be consciously designed to act as non-art experiential objects, things begging to be mastered as one masters solitaire, juggling, or poker. This includes a great many games that I truly enjoy playing, and even includes a great many games I do see as art, albeit generally not complex or particularly interesting art.
The crucial takeaway here is that whether a game will fall within the category of art is dependent on one’s perception of the ‘end’ toward which a game is the ‘means.’ As it is dependent on perception, which games fall under the definition certainly will vary from person to person. But as there are criteria by which one makes these perceptual evaluations, it is not a meaningless definition accepting literally everything as art.
There are multiplayer games that I think are art to me, like Journey; and there are RPGs that I think are not art to me, like Maplestory. There are also genres, like puzzle games and platformers, that contain a broad array of art games and non-art games. It’s a fuzzy, complicated, and impossible endeavor to establish specific experiential objects as art in a way that everyone will agree with. But with the definition described here, one can at least find out what they are personally approaching as art, and why.
Infinifactory as a Work of Art:
Here we turn to a proper example to be considered from the ground up, and we encounter another troubling edge case where there is unlikely to be an overwhelming consensus on one side or the other. Infinifactory: art, or not art? As established at length thus far, the answer will differ from person to person. As you’ve probably guessed from the introduction or the title or an earlier remark about it, for me Infinifactory is a work of art.
This section will not contain much analysis of Infinifactory, but instead a description of the basis and framework that one might use to approach it as art for the purpose of analysis. (If you would like to see what actual analyses of games starting from the foundation in this article would look like, I would encourage you to check out my articles on Papers, Please, Dark Souls, and The Witness.) Now, while the aim of this section will be to showcase what it means to approach Infinifactory as art, I will also discuss the alternative approach further along. So, let’s get to it.
Zachtronics, a company helmed by Zach Barth, is a purveyor of a very unique group of puzzle games—including their titles Spacechem, Shenzhen I/O, and Opus Magnum, among others. Their puzzle games have been given a lot of special genre names: problem solving games, programming games, design-based puzzle games, and so on. To her own endless amusement, my partner refers to them derisively as ‘government work.’ However you label them, the defining features of such games are consistent: a complex, at-least-partially-visual programming language as gameplay; flexible puzzle constraints and goals which allow massive numbers of potential solutions per level; and a steeply ascending difficulty curve.
While all of Zachtronics’ design-based puzzle games incorporate their thematic content into gameplay to some degree, Infinifactory has the closest relationship between the two (slightly edging out others like Shenzhen I/O and TIS-100 in this regard), and is also unique as the only first-person 3D game among their offerings thus far.
The way that Infinifactory incorporates its thematic content into its gameplay is by having both the game’s worldbuilding and its driving conflicts play out primarily through the objectives, settings, and gameplay of the levels. Narratively speaking, Infinifactory is about a human being abducted from earth by aliens called ‘overlords’ to serve as slave labor in designing and constructing productive machinery—and the efforts of that human to survive, escape, and (if possible) end the enslaving acts perpetrated by the aliens.
The actual gameplay within the levels involves designing assembly lines out of standardized devices, in order to take raw inputs and turn them into refined outputs. As a captive, the player-character is tasked with building factories for everything from weaponry to food production, both of which have a gruesome semicomical undertone. Within the levels, one occasionally stumbles upon the corpse of a less fortunate captive, accompanied by a short audio recording of either their last words or commentary from a captive who found the corpse before you.
Now, right off the bat, there are some things that may signal a player to the perceptible details which single the game out as art in my estimation: the competent mixture of dark and comical tones in its narrative elements; the intertwining of the thematic callousness of the warlike, slaving overlords and the actual objects the player-character is forced to produce; and the inclusion of corpses with recordings of them breathing their last breaths every few levels to provide stakes and consequences to the actions of the player-character and the aliens. Beyond this, there are many subtler aspects of the gameplay which do double-duty as underscoring for the game’s themes; for example, the enslavement of people in the game by the overlords is not only a heartless waste of human lives, but also necessitates wasteful exploitation of resources through some of the demanded outputs which entirely lack (i.e. designate for immediate disposal) some of the inputs.
These elements accomplish two things: they marry Infinifactory’s thematic content to its gameplay to a significant degree, and they touch on topics of both intellectual and emotional importance to many observers. Add to all of that the lengthy, methodical process of producing a functional solution for any given level—which provides plenty of time to slowly mull over these elements and circumstances—and one may find themselves seeing how this is a work which could be fruitfully analyzed as a work of art.
But there are also elements I’ve left out of this description so far: by counting parts used and similar metrics, the gameplay lends itself quite well to obsessive repetition and optimization of the act of designing the factories, and even includes histograms and leaderboards to encourage clever optimization as a form of competition between players. Essentially, through the menu system that grants access to each level, Infinifactory breaks the fourth wall to encourage players to approach it with a similar mentality to playing a competitive game.
Viewed this way, the game seems poised between the appalling situation that the player-character is attempting to survive and the game’s mechanical reinforcement of wanting to continually repeat and improve the tasks given to the player-character, many of which were originally slave labor. So the game seems to lack cohesion; it has that dreaded buzzword, ludonarrative dissonance.
Does this destroy my evaluation of the game as art? Nope! As a matter of fact, that’s essentially why I selected it as the example game for this article in the first place. To me, the primary end of Infinifactory relates to its thematic and gamely stress on the value of human ingenuity and tenacity under duress, and to a lesser degree on the specific topics of slavery, exploitation, and armed conflict. That the work contains a few discordant notes in its menus in order to function on multiple axes of possible player engagement is not sufficient to convince me it isn’t art.
But moreover, I would have no problem with someone who places greater stress on a perceived gameplay loop of repetition and optimization (even to the exclusion of acknowledging any of its thematic content, or the detriment of their overall progress through the campaign) telling me that they either do not consider Infinifactory to be a work of art, or that they consider it to be a work of art with entirely different priorities than those I’ve assigned to it here.
In order to move beyond a naïve version of the sentiment that games can be art, it will be necessary for each person who holds to that position to be comfortable with the notions that not all games necessarily are art, and that many games that seem like art to them will seem to not be art to someone else.
The fact of the matter is that games themselves, as a whole, as a medium, are an edge case in this conversation. I would be hard-pressed to come up with even one example of a professional film or a published novel that actually fails to meet the definition of art I’ve employed here—but as you saw two sections ago, I was able to rattle off a list of such games with relatively little effort, and I think opinions on the artistic status of particular games can vary widely from player to player. In effect, at least in these early decades of game studies, one will have to be not only an ontological critic but also a patient and magnanimous critic. It’s anyone’s guess which of those it’s harder to be . . .
There is a tremendous amount of grey area remaining in everything I have argued in this article. Whatever objections sprang into your mind while reading it—be they about my definition of art, my broad statements about the gaming community, my categorizations of specific games or genres, or my discussion of Infinifactory—you can rest assured that, in my infinite paranoia, I almost certainly thought of those objections as well, and considered devoting an extra thousand words to addressing each of them.
Your objections may include: that my definition of art is too broad, and by relying so heavily on subjectivity it technically still allows anyone to say that anything is art; that my definition doesn’t actually require an intention from an artist, only that the observer perceives an intention; that my definition is too narrow, and it excludes works that you are positive count as art; that my restriction of art to emotional and intellectual ends is too arbitrary, as both are technically elements of human bodily experience and thus mechanical and physical ends should be equally valid; that I’ve called a particular work art when it doesn’t meet your definition; that I’ve denied that a work is art when it does; that my evaluation of Infinifactory, despite its dichotomous nature, still fails to capture your opinion about it; and many more!
That kind of objecting is absolutely fine by me. The purpose of this article is not to lay out an explicitly and objectively true trademarked doubleplusgood unquestionable magnificent definition of art. We’ll leave that to Oscar Wilde. The purpose of this article is to push the general, non-academic conversation about games as art in a productive direction. Formal game studies, especially under the subheading of ludology, have taken very productive steps toward building out formalist close reading strategies for games across the past 20-30 years; but compared to the older formalist strategies for reading more conventionally accepted media of art, public knowledge of these approaches is lagging significantly behind.
My personal ad hoc theory is that this lag results from the fact that so many players broadly insist that all games are art indiscriminately; this heightens the aversion of unconvinced parties who are primarily aware of high-profile non-art objects in the gaming space (like the multiplayer aspects of many popular FPS and sport video games). In order to spread game studies more widely into schools and universities, the gaming community will need to first accept a position that is compatible with two potentially uncomfortable points: (1) that certain games and even certain game genres are straightforwardly more apt candidates for broad consideration as works of art than others, and (2) that acceptance of games as art will mean tolerating the perspective that developing skill within a given game can be a form of engaged reading or engaged viewing of an art object, rather than an end unto itself. Everyone can enjoy games, but what I foresee people studying extensively in the humanities programs of the future are not just games but works of “gamely fiction.”
Again, this is not to say that all great games communicate stories, although many do. “Gamely” attributes would simply be those relating to gameplay (i.e. interactivity, and variability within a limited range). Therefore, a thorough close reading of a game as a work of art should always include a reading of the gameplay. And a work of gamely fiction would be a work of art which makes excellent use of its medium’s unique elements (as opposed to a work of literary fiction or a work of cinema, which would almost always lack gamely attributes).
At any rate, no progress can be made in the popular segment of this field when people can not get clear on what it would mean to call a game art in the first place; when the majority of players who automatically insist that games can be art seem openly hostile to approaches which treat games as works of art; and when some people carry on working with games as objects of study as though they are just like books or just like movies. Games are games, and some games are art.
 This is not a disagreement with aestheticists, those who would uphold Oscar Wilde’s famous tenet that “All art is quite useless.” In the context of the preface to The Picture of Dorian Gray in which that line appears, Wilde is conceiving of ‘use’ as a practical, non-emotional utility—like that possessed by a cooking utensil or a woodworking tool. At least in the sardonic context of that preface, he would probably object that his conception of ‘use’ includes intellectual utility, but I don’t think that difference of opinion is enough to discredit us as partial allies. And, as you will find throughout this article, we are in fervent agreement when he says, “It is the spectator, and not life, that art really mirrors.”
 This is not to say that the perceived end or aim of a game that is a work of art must be necessarily beneficial. The works that I prize as having artistic excellence are the works of art I perceive providing one or more of those virtues with special depth or insight. But a work of art may instead be perceived as offering something less praiseworthy, such as trite moralizing, objectification of others, political propaganda, et cetera.
 Non-art games will be studied as well, but mostly by game designers, sociologists, anthropologists, psychologists, and so forth. What I mean is the material that will be studied extensively by literary critics, film critics, philosophers, psychoanalysts, and (without any of the term’s preexisting journalistic context) game critics.
 I favor the word ‘gamely’—partly because of how it parallels the word ‘literary’ and partly because I came up with it myself about five years ago—but academics tend to favor the word ‘ludic’ for accomplishing the same goal. So rather than ‘gamely fiction,’ you may choose to say ‘ludic fiction’ instead.
Infinite and Individual: