Unit Operations: An Approach to Video Game Criticism

Of the several books Ian Bogost has produced so far as a video game academic, Unit Operations remains probably one of his densest and hardest to follow. There are a lot of reasons for this. For starters, it’s not so much about video games as it is a philosophy of how numerous forms of media can be interpreted like a computer program. The first 45 pages of the book don’t really even address games, instead creating a large philosophical foundation for how literature, film, and video games work like a unit operation. The book’s focus narrows to video games eventually because as an artistic medium, their unit operations are very obvious. It’s much easier to convince someone that Grand Theft Auto III’s narrative revolves around mathematical abstractions then it is to say Hamlet revolves around them. Everything from Derrida or Aarseth to T.S. Eliot’s notion of objective correlation between elements is brought into this analysis. As Bogost himself notes at the start of the text, the educational background required to fully grasp all of these elements is not only large, it’s contradictory. People with engineering degrees tend to have trouble with abstract thinking (but that’s the rule!) and people with liberal arts backgrounds tend to have trouble with objective thinking (but that’s not fair!).

It’s hard to pin down a specific definition to the concept of unit operations because Bogost ends up comparing it to so many different things. If you were an English Major, I’d tell you that essentially it’s an archetype. If you were a computer programmer, I’d say it’s a way of organizing a program’s functions. Bogost writes, “A unit operation may be observed in any artifact, or any portion of any artifact, rather arbitrarily. I insist on this broader understanding of unit operations to allow its logic to resonate across expressive forms, from literature to film to software to video games.” (14) I’m going to break down the definition of unit operations into two areas: literature and computer programming and then explain how Bogost applies that to video games.

First, Bogost compares a Baudelaire poem with a Bukowski poem. Both poems are about seeing a beautiful stranger on the street and the depressing realization that you will never see them again. He explains, “the figure that fascinates has become an effective unit operation, a tool for engaging modern life. It would be overzealous to equate this figure that fascinates with a software subsystem. [. . . instead] Bukowski’s poem relies on a consolidated version of Baudelaire’s figure, that it enacts this figure by playing by its rules.” (80) Bogost eventually acknowledges that this works a lot like an archetype. In this case, the character from either poem is defined by several rules. They are a beautiful stranger on a crowded city street, and the user will never see them again. That’s the archetype/reoccurring event that constitutes a pattern.

Unlike two similar poems, which are using the same abstract concept to communicate a feeling, a computer program communicates through the manner in which the programmer is organizing its various subroutines. Bogost writes, “Pattern creation or recognition systems…usually take the form of unit operations that perform one kind of action on data, resulting in some judgment about its worthiness as a particular pattern.” (29) I don’t have a sufficient background in computer programming to grasp all of this so I got some help from one of his grad students, Simon Ferrari. I’m going to just post a trimmed version of his explanation:

When you get into the computer science background assumed in Unit Operations, it’s talking about object-oriented programming (OOP). Like a lot of things we study in digital media programs, OOP goes back to Xerox PARC. Alan Kay’s team created this language called Smalltalk, based on the idea of “classes” that had been established earlier on. Data is encapsulated into a series of independently functioning units. I had to take a programming class to really understand what this means. First you declare a class, which is kind of like an archetype or a Platonic idea: it has a name and a set of generic traits and behaviors (this is the “constructor”). From a class you can propagate objects, individual working units which inherit characteristics from their parent class (and can also bequeath characteristics to another generation of objects, called their “children”). The big move here, the important one for Bogost, is that classes and objects in OOP also contain their own functions–if you’re a member of the class “gameJournalist,” then the action “interview” is already built into your code. This means that once you declare a few objects, you don’t necessarily need to have other, system-level functions running to get them to interact. This paradigm wasn’t very popular until the 90’s, but you can see why it caught on once programming came out of the nerd ghetto: it’s a lot more human than a series of subroutines.

To iterate this point, Bogost uses the example of a computer program tracking the background data of potential terrorists. What does it mean if a person is 70% likely to be a terrorist to a security guard deciding whether or not to search them? The program has created an abstraction because, like the poems, there are elements of uncertainty and mystery created by finite rules.

How this idea ties into video games is that a unit operation can be analyzed distinctly to gauge what the message of a video game can be. Bogost explains that computational systems “rely on unit operations as their primary mode of representation, and thus unit operations have a special role in how works like videogames function.”(65) In literature and film, unit operations are present but not nearly as central as a computer program. To make this point Bogost explains that games are organized by their technological capacities rather than their IP or even controls. We define a game by what the software allows us to do in it. This is distinct from film or books, which divide themselves by the story. He writes, “First-person shooter game engines construe entire gameplay behaviors, facilitating functional interactions divorced from individual games. Genres structure a creative approach to narrative; they describe a kind of story…game engines differ from genres in that they abstract such material [guns, villain, etc.] requirements as their primary – perhaps their only – formal constituent.” (57)

How does this work in application? Essentially, a video game asks a player to make sense of a bunch of data patterns or abstractions. Put Bogost’s way, “Video games require players to create a subjective understanding of the synthesis of one or more unit operations. Games demand that players be capable of making this synthesis palpable in their own experience.” (123) Understanding how this then leads to the creation of a message as concrete as a poem depends on two statements Bogost makes:

1) Unit Operations are biased. They always show a way of things working that are limited by human thought.

2) Unit Operations and Subjectivity have a dialogue which is simulation fever that occurs on the outside, in the viewer, not intrinsic to the game. (133)

Number two is the big one here. Subjectivity is where you, the individual, come into the equation. A video game player interacting with a large series of unit operations is generating an experience that is not intrinsic to the game. Or, put another way, that’s the key ingredient for something being a message instead of a sport. The message of a game is basically its depiction of reality (the simulation) and the subjectivity is our reaction to it based on our individual experiences.

Bogost outlines a couple of different types of simulation response or fever for a player. “Simulation resignation”, he says, “implies the blind acceptance of the limited results of a simulation, because the system doesn’t allow any other model of the source system.” “Simulation denial” implies the rejection of simulations because they offer only a simplified representation of the source system. Bogost contends, “A simulation is the gap between the rule-based representation of a source system and a user’s subjectivity.” (107) Here is where Derrida’s contention that “the only way to preserve work in an archive is to expose that work to its possible deconstruction” becomes key to unit operations. The way a simulation can communicate an idea with us is by the player being aware of its limitations and recognizing what the system is trying to communicate as a whole through its limited depiction of reality (109).

How does one apply this to video game criticism? Bogost talks about Greg Costikyan’s reaction to September 12, which is a game about bombing Iraq. The player shoots a missile that moves very slowly at a terrorist, often hitting civilians instead. The surviving civilians go on to become terrorists themselves. Costikyan, who lives right near where the World Trade Center used to be, wrote a scathing critique of the game’s political message. To summarize, the game isn’t realistic and doesn’t factor in numerous complexities about 9/11. Bogost says that this is a prime example of simulation fever or someone having a strong reaction to a depiction of reality in the game. (132) Another example is Grand Theft Auto, which makes us aware of our desire to misbehave and break the rules because our reality varies from the projected one. I can run around smashing cars and shooting civilians, which is obviously not realistic. The difference between reality and the simulation makes me aware of the game’s discrepancies and thus how I would behave in a modified reality. Bogost notes, “The reviewer who insists that once you play GTA ‘you can’t go back’ suggests that the game successfully draws attention to the player’s relationship to potential delinquency.” (168)

This is important to video game criticism simply because the concept of unit operations neutralizes a lot of the value arguments people still make to this day. A linear game is just one that is more biased than an emergent game. Both are just techniques for communicating with the player. The game’s message, the simulation fever, is still extrinsic to the system. Or as Bogost notes, “The type, and not the degree, of emergence is the deciding factor in the expressive potential of a complex system.” (151) Given the finite nature of such expression in any simulation, Bogost’s ultimate contention is that the player’s meaning is still going to come from where the simulation ends and reality begins.