[CHWP Titles]

What you C is What You Get: Formalizing Computer Game Criticism

Patrick Finn

St. Mary's College, Calgary

patrickfinn@shaw.ca |||| http://members.shaw.ca/patrickfinn/ |||| About the Author

CHWP A.22, publ. September 2003. © Editors of CHWP 2003. [Jointly published with TEXT Technology, 12.1 (2003), McMaster University.]


[Astract / Résumé]

KEYWORDS / MOTS-CLÉS: Computer games, literary theory, ergotic text / jeux électroniques, théorie littéraire, texte ergotique.


section1.Introduction
2.Two Kinds of Theory
3.Two Kinds of Theorist, Two Kinds of Response
4.Half Life, Full Theory: A Critique that Computes
5.Critique Meets Computer: How Virtual Rubber Hits the Virtual Road
6.Conclusion
7.Notes
8.Works Cited


Introduction

This paper originated as a talk delivered on a panel co-sponsored by the Association for Canadian College and University Teachers of English (ACCUTE) and the Consortium For Computers In The Humanities / Consortium pour ordinateurs en sciences humaines (COCH/COSH) at the University of Toronto. That panel was posed the following question:  “Do we need a ‘new theory’ in order to study computer games?” To answer this question, I believe we need to determine to whom the “we” in the question refers. The most direct answer seems to be that we refers to the members of ACCUTE and COCH/COSH and those who might follow their teaching, research and publishing. The following paper argues that while we may need a new theory, that theory is already under development in a number of different forms.

The original call for papers featured a quote from Espen Aarseth sounding pessimistic about early attempts to theorize computer games that used existing critical models to analyze new media. Aarseth, one of digital culture's most insightful critics -- and greatest academic beneficiaries -- can be forgiven for being suspicious of ill-founded theory. We are all a little wary in the post-Sokal era[1]. The quote was as follows:

[. . .] the race is on to conquer and colonize these new territories for our existing paradigms and theories, often in the form of "the theoretical perspectives of <fill in your favorite theory/theoretician here> is clearly really a prediction/description of <fill in your favorite digital medium here>. (Aarseth, 1999)

Two Kinds of Theory

For the purposes of argument, I will divide theory into two groups - cultural and formal. Cultural theory I take to include studies involving gender, class, and a variety of internal and external power relations and their interactions with cultural production. From my perspective, cultural theory or theorizing finds as good a home in the world of computer games as it does in any of its other haunts. In particular, aspects of embodiment, race, class and the social or community behaviors of games and gamers have been the subject of a number of thoughtful studies. Yet, it is partially in response to these approaches that Aarseth finds a problem. Aarseth is troubled by theoretical positions that have moved from one feedstock to another without fully accounting for the shift.

One way to address these concerns is to look to the second form of critique, the one that I have connected to formal theory. While cultural studies models can be used effectively across a number of disciplines, we need to ask what aspects of a theory need to be either changed or created anew in order to address specific forms of expression. The answers involve formal elements.

Just as film studies developed formal critique to complement the transposition of cultural criticism in the mid 20th century, computer game theorists need to develop criteria for their subject. As with film, there is crossover - items such as plot, theme, perspective and character easily move between media. However, elements that are new additions introduced along with games add different terms and change the way standard formal components apply. Thus, the consideration of plot can work in each of The Faerie Queen, Macbeth, Tom Jones, Apocalypse Now and the computer game Half Life. Yet, each example entails further considerations related to its specific formal components, which in turn add to any viewing, hearing, reading or interaction.

With these formal elements in play, one can more readily apply cultural readings. Indeed, it is the set of formal elements related to a given medium that often act as the point of engagement incorporating disparate voices around the given cultural productions.

Two Kinds of Theorist, Two Kinds of Response

For ACCUTE members and those interested in literary studies, digital poetics will most likely involve the formation of a series of new formal measures of computer game functions that develop from traditions of English Literature and poetics. The incorporation of new media forms in literature departments has a history evident in the development of film studies, and the retrospective examination of performance related to dramatic works.

The type of comments that Aarseth is making are similar to those made by early critics working in film and drama who felt animosity towards scholars, primarily those in English departments, who practiced interdisciplinary work that favoured textual readings of mediums that were not themselves textual. In those cases, the attention to a media-specific formal critique served to establish a useful dialogue between cultural concerns and communicative particulars. Recent examples of this are found in the work of Theresa DeLauretis on film, and William Worthen on drama. These examples build on formal components such as plot, character and atmosphere, while incorporating the analysis of items specific to their areas of study such as montage and blocking.

Computer game theory might follow a similar path, borrowing elements from the textually based study of English and adding visual and oral components from the resident medium. What I feel must accompany these is some account of the structural or formal components that arrange these sets of expressive media.

For the COCH/COSH group and those interested in the computational aspects of humanities research, computer game theory is just one more development of computer applications in the humanities. This group knows all too well the types of critical excitement and institutional resistance that come with being the new kid on the academic block. However, this group is well prepared for a new game theory since it already has an established body of research examining the use of computers in relation to narrative and communicative expression from which to work. It is within this group that we find Espen Aarseth's concept of 'ergodic literature' a response to the questions concerning the need for a new theory for digital media.

Adding to these forms of research, computer games bring new challenges and possibilities. To demonstrate some of these I offer an analysis of the popular game Half Life developed by Gabe Newell and his company, Valve Software, and distributed by Sierra Games. What I hope to show is that each of us, whether we come down on the side of the members of ACCUTE or of COCH/COSH or both, bring with us a useful set of critical considerations that can enrich our experience of a game like Half Life, while contributing to the development of a new theoretical mode of enquiry.

Half Life, Full Theory: A Critique that Computes

For those not familiar with Half Life, I will briefly describe the game and then move into a closer analysis of some of its components. For the purposes of the example, I will follow roughly the familiar formal considerations of plot, atmosphere and perspective (or point of view).

To begin, Half Life is part of the ‘first-person-shooter’ genre. In this form of game, the player assumes the role of a character whose movement and sight lines they inhabit. Typically, the first thing gamers see is a set of arms in front of them that pick up various items, including weapons and tools, that perform functions such as ladder climbing, the operation of vehicles and door opening. As is apparent from the genre's name, the most common action is the firing of weapons. In Half Life, these weapons range from realistic replicas of grenade launchers, desert eagle pistols and M16 rifles to more ethereal implements, such as hornet guns and spore launchers. These elements are common to first-person shooters, with some games like Max Payne opting for more realistic weapons and others like Lucas Films (Jedi Knight going for more fictional constructions.

Most first-person shooters also contain some type of introduction to the plot that arrives in the form of a booklet, as introductory scrolling text, as a voice over, an introductory film which orient players in the game or a combination of these types. Half Life uses all of these, while highlighting two specific elements. The first is an interactive section of employee training, done in character. The second is a digital film that informs gamers they will assume the role of the bespectacled Gordon Freeman, a recently minted PhD who works in a large research lab called Black Mesa. Within Black Mesa Gordon encounters a variety of senior researchers many of whom are helpful and many who are grumpy and incompetent. All are at the mercy of a malicious character known only as the Administrator who always remains just out of reach.

After the initial narrative is established, Gordon faces a series of complications that further the plot. At regular intervals, he intercepts radio transmissions and encounters characters that provide information. Half Life, which has won more awards than any other first-person-shooter, is known for the strength of its story. Yet, I would wager that many among the ACCUTE group would find the storyline and dialogue a little thin if they compared it with say, Middlemarch. Still, many of the same techniques that scholars apply to the Victorian novel apply to these plotlines, which form a crucial component in the computer game experience. As with the examples of film and drama however, this narrative component is only one element of a multimedia expression.

In computer games, atmosphere is often as significant as plot. In Half Life, this is another well-known strength. The ability to create an engrossing virtual world is one of the fundamental tests for a computer game, and the makers of Half Life set industry standards with their portrayal of the Black Mesa compound and the alien netherworld that is opened as a result of experiments done at the facility. Here the talents of painters and illustrators who work with brush and canvas intersect with 3D artists and programmers. Creators that test the limits of the form while attempting to tell a story facilitate the portrayal of Black Mesa. One way for critics to formally theorize the representation of atmosphere in computer games might take account of the development of x/y axis movement, hinged to image libraries within rendered frames. These techniques first developed for Wolfenstein 3D, the original first-person-shooter, released in May of 1992, show a developmental pattern as we move through later games and up to the present day. As breakthroughs in rendering and programming changed, bringing more complex games like Duke Nukem, Doom and Quake, creators were able to explore different aspects of storytelling. With Half Life, the creation of the virtual world became so rich it is viewable from multiple perspectives.

Perspective in game critique directly involves the consideration of role-playing. As a formal component of game creation, this type of perspective was revolutionized by Half Life. In two subsequent releases, Opposing Force and Blue Shift, players change into new characters that serve different functions within the same base narrative and same series of maps. In each, gamers get an introductory film that outlines the new perspective. Once playing, the new character catches glimpses of the original Gordan Freeman character that was previously played or enacted. This shift in role-playing serves to enrich the interaction with the base story and with the world in which gamers play.

If we consider a formal analysis of Half Life then, we might include the intersection of story, atmosphere and perspective within a broader critique of the first-person-shooter genre. Still, this form of critique, while it potentially offers new thinking about games, leaves us well within the realm of standard or traditional textual critique. Although the introductory film, the inhabitable environment, and the unique aspects of first person perspective within this reading are in some sense new, they are really only a modulation of the same types of tools that are used to critique short stories and novels. In order to offer a more specific examination of computer games, there must be some direct analysis of the unique construction involved in the compositional process as well as a more focused critique of the manipulation of play.

The primary composition component of game design is programming and here the beginnings of formal critique are well underway. As with many other forms of creative expression, the creators and consumers of the form have established a dialogue around the constituent components. These initial critiques take the form of replication, parody and commentary. Thus, the gaming community is already rich with theoretical analyses of computer games. While the academy has offered cultural criticism of such things as violence in games like Half Life, it seems clear that we currently lack sufficient critical analyses of the limitations and possibilities of programming in C, C++ and Assembler that are specifically geared towards 3D graphics cards, large memory components and high speed processors.  By bringing the two forms of critique together, we might be able to address the types of concerns expressed by Aarseth.

Half Life's creator Gabe Newell and I discussed the way in which the game design process has worked. Newell sees programming as the art form that allows him to explore his interests in storytelling. Newell, who got his start as employee number 273 at Microsoft, says that gaming allowed him to pursue a form of creative expression that disappeared in his working life as Microsoft grew into a corporate behemoth. Since the popularity of computer gaming in general and of Half Life in particular, Newell has made a number of appearances at academic conferences that seek to answer questions similar to the one that I am concerned with in this article. Yet, like those who worked in cinema during the inception of film studies, he feels there is a need for a greater awareness of the way his medium actually works.

For example, the original impetus for Half Life came out of a desire to develop a greater level of complexity in the first-person-shooter genre that was at the time based on little more than tests of eye-hand coordination. Building on the story elements of earlier games such as Doom, Newell turned to contemporary fiction to establish Half Life's famous atmosphere and to his work experience for the narrative. Using the Stephen King novella The Mist as a guide, the team attempted to develop a semi-realistic world that was both believable and unsettling. For Newell, the most attractive element of the first-person-shooter genre is the potential to immerse the gamer in a separate world. He explains his attempts to highlight this aspect through the use of back-story, atmosphere and perspective. Focusing on these elements, he was able to make up for the technical shortfalls that he faced with computers, which cannot currently rival the quality of cinematic or televisual fiction.

The game engine was programmed in C with artificial intelligence components done in C++. Then the game was re-coded using Assembler, which allowed the interaction of the components in a manner transferable to the chipsets then available. For Newell and the creators at Valve, criticism that fails to consider programming begins by ignoring the computer game's most important component. I would argue that a theory done this way might miss what it is that makes computer games into computer games.

Critique Meets Computer: How Virtual Rubber Hits the Virtual Road

If we accept that a new theory should account for the primary language used in composition, a familiar question arises. How much programming should a humanist be expected to learn? Early lessons from humanities computing have shown it is unrealistic to expect students and scholars will be interested in or even capable of learning programming languages at the same time they are satisfying their other course or research requirements. One answer to this question was offered by Nancy Ide at the recent COCH/COSH meeting Malaspina University-College in Nanaimo, British Columbia. Ide, the Chair of the Computer Science Department at Vassar, and one of the original contributors to the Text Encoding Initiative, made early calls for full knowledge of computing languages. Over time, experience led her to revise that call toward a position which would see students study the way in which programming works as opposed implementing a computer language requirement. (See Ide for further discussion.)

This seems to me to be an interesting compromise. It would certainly allow computing humanists to engage in an informed critique of computer games that would be on par with the types of critiques that I mentioned in terms of film and performance studies. From my perspective, knowledge of how computers work is an essential component of contemporary critical enquiry. However, my position comes from personal research interests and I would agree with Ide that we do not need the same proficiency from everyone in the humanities - for now. In the same way that many students are able to get degrees in English departments without studying Anglo-Saxon or Medieval English, there will always be a demand for different paths towards degrees within Humanities faculties. For computing humanists however, I think the impetus towards programming acuity is somewhat stronger. As departments begin to examine computer games, there will be a need for a greater level of technical awareness among students and scholars wishing to specialize in the area. One possible approach would be to use a book like John Sowa and David Dietz's Knowledge Representation in research methods courses for humanities students, with an added design component for those wishing to examine the visual components of computer games or to become general specialists in the area.[2]

Meanwhile, away from the academy and professional critical community, this work already has its own set of protocols. Two further aspects of Half Life can help to demonstrate this point. These are the multiplayer program component and the design and modification package. Written as two separate programs that complement the central engine, these components have given rise to a community that provides critical analysis of the game, and through a system of apprenticeship, has produced the next generation of game designers.

Online gaming allows players to log on to a remote server and play against multiple players in pared-down versions of the original game. The most famous example of this is the Canadian-designed Counter Strike, which is a terrorist theme modification of the Half Life engine. The game's original construction continues to influence play, but adds new story and context. In the case of Counter Strike, this involves simulated acts of terrorism with characters from political groups that we would all recognize from today's news headlines. With real time communication facilitated by text and voice message support, a secondary dialogue occurs within the game that is often the source of programming tips and other related exchange. In my experience, games can completely stop as players engage in the discussion of various design aspects and of other programming-related issues. Most often access to these discussions is predicated on gaming competency. The formal critique that occurs within the games then finds its way back into the creative community indirectly through word of mouth and directly through game development and modification.

This development is supported by the second component. Worldcraft or the Sierra Development Kit are Half Life-specific design programs that allow users to create new game levels or modify existing ones for individual or multiplayer use.  These levels can in turn offer tiered levels of choice to players who can then manipulate options such as weapon performance, player health and the force of gravity. These 'mods' are posted and various designers visit and analyze the work. A common feature of this is a walk-through where the designer leads visitors through the creation, pointing out choices and asking for commentary. What unites these people is the original game and the formal elements of the engine and modification kits. Hence, formal critique, as is so often the case, begins in the immediate community where knowledge of the creative function is direct. This community is now so strong that more than half of the people that work with Gabe Newell at Valve come from their numbers.

Conclusion

Finally, to return to our question -- do we need a new theory for computer games?  I think the answer depends on whether or not we feel any responsibility or desire to engage with games. As I have shown, a new theory is already in development within the gaming community. Given that a number of academics have already published commentary on the cultural aspects of games, and that games are now a larger moneymaking industry than film in the United States and Canada, it seems likely that we will come to discuss them to an increasing extent and we will normalize their critique. After all, the academy, like all good bureaucracies, eventually makes everything its business. The question then becomes one of procedure - of ensuring that we are aware of the rules of each new game we enter - after that - the play's the thing.

Notes

[1] I am referring here to the controversy over Professor Alan Sokal's hoax article which lampooned literary theory even as it was published in the theoretical journal Social Text. For a bibliography related to the incident see Sokal's home page.

[2] Sowa's book is only one suggestion, but what is clear is that theorists wishing to address computer games in a formal sense must begin to grapple with some idea of what is at stake in knowledge representation and the functions of epistemological arrangement that are to be found in studies of various modal logics.

Works Cited

Aarseth, Espen. Cybertext: Perspectives on Ergodic Literature. Baltimore: Johns Hopkins University Press, 1997.

----. Home Page. www.hf.uib.no/hi/espen/.

De Lauretis, Theresa. Alice Doesn't: Feminism, Semiotics, Cinema. Indianapolis: University of Indiana Press, 1984.

--- ed. Technologies of Gender: Essays on Theory, Film, and Fiction. Indianapolis: University of Indiana Press, 1989.

Ide, Nancy. Home Page. www.cs.vassar.edu/~ide/.

Sierra Games. Home Page. www.sierra.com.

Socal, Alan. Home Page. www.physics.nyu.edu/faculty/sokal/.

Sowa, John and Dietz, David. Knowledge Representation: Logical, Philosophical and Computational Foundations. Toronto: Thomson Learning, 1999.

Valve Software. Home Page. www.valvesoftware.com.

Worthen, William. Shakespeare and the Authority of Performance. Cambridge: Cambridge University Press, 1997.