Center For

Literary Computing

Exploring and advancing the digital humanities

Looking Glass Project

The Looking Glass Project

How can we use immersive virtual reality to explore literature? And what do literary concerns such as narrative, textuality, and authorship show about VR? The LookingGlass project is a collaboration between the CLC and WVU’s Virtual Environments Lab. The project explores Lewis Carroll’s Through the Looking Glass, combining the novel’s richly visual setting and linguistic play with an immersive 3D environment used the game engine from the popular first-person shooter Half-Life.

Looking Glass: Literary Research into Virtual Environments

Project Justification

Our goal is to create a prototype for virtual literature, that is, both a virtual environment that explores literary concerns and a literary artifact that explores VEs. Lewis Carroll’s Through the Looking Glass is ideally suited for such research: the novel merges but never collapses physical spaces, fantasy worlds, and simulated environments. The novel is composed of vivid scenes, mediated by extreme transformations. The spaces between scenes are marked by gaps and abstractions, by the loss of any sense of reality or reference. The effect is of radically discrete “spaces” (as on the chess board which Carroll claims for the underlying algorithm of the novel). The experience of reading the book is not one of rapid jumps but hazy and weird fading in and out.

Through the Looking Glass narrates a space that is unstable and shifting. In turn, the novel is framed by the question of whether Alice dreamed the whole thing, or whether the Red King is in fact dreaming it all, including Alice. There’s an undeniable relation between this framework and the narration: the events are in some way the outcome of Alice’s interpretive process, a process that is always conjectural and hypothetical. Certainly there’s a rigid “superstructure” of chess, but the role of this structure is not always clear: while chapters largely follow chess moves, there is considerable narrative development that does not fit a chess game. In what way is the narrative an outcome of the chess game?

Now, a first person shooter or ego shooter computer game raises related questions. The player/user scans across texture maps to make sense of space and orientation. There’s a sense of immersion (thus first person or ego), but what is the relation between claim that this is a geometrically perfect, coherent, “real” space (thus immersion), on the one hand, and the user’s visual clarification and apprehension of details and outlines, on the other? In fact, there’s a constant process of interpreting of up, down, forward, back, etc. Forms of spatial representation are constantly invoked and applied. While the events of the game – the game play and interactions – rely on the presupposed stability of this space, there is nonetheless an ongoing cognizing construction (“navigation” is still the best metaphor for the process involved).

We propose to build a game-like application, where the game’s objective is to explore and experience certain aspects of Carroll’s novel; and, at the same time, to explore our experience of virtual environments. The goal here is not to shoot or to solve clues per se, though there may be some of the latter, but to move through the environment. Through the Looking Glass is easily adapted to the notion of discrete spaces of levels employed by computer games, while at the same time constantly drawing attention to the artifice of this construct. In fact, many of the philosophical questions explored in Carroll’s narrative are already a part of the user/players experience of the game. A working hypothesis: the difference (epistemological, phenomenological, and ontological) between levels of narrative – between the frame tale and Alice’s adventures in looking-glass land – corresponds to the difference between the user and the game space. Both involve immersion and the transference of the self (the first person or ego) into the character in the narrative or game. Of course, it’s easy too forget this transference – forgetting is part of the point of “immersive” environments – but part of the interest of Through the Looking Glass is the way it reminds us of this process.

The four parts of the project, as currently envisioned

  • Interface
  • Labyrinth/Game Space
  • Event Spaces
  • Characters/Behaviors


We want to involve the physicality of the user, to activate an awareness of the many dimensions of the user’s physical experience. The solution, in the meeting of the novel and the immersive game, is to force disorientations or at least attention the interface. There will be three choices for navigating the application: a first person shooter-like interface, a scrolling version of the novel’s text, and a chess game. (Our initial set-up will follow the chess problem in the novel, but by imagining a much larger space and characters with certain autonomous behaviors, eventually other of chess problems could be constructed.)

We imagine an interface using head and wrist mounted sensors, which is launchable both on a desktop and a larger projection system (e.g. Immersadesk). We’d like to integrate other senses – e.g. the user’s walking as a way of navigating the labyrinth – but perhaps that comes later? In the descriptions that follow, we imagine the user’s movement in real space to be mirroring those in virtual space because of the glove and head-mount, etc.

We want the look to be immersive 3D but maintaining the familiar feel of computer games. We imagine the space as both environment and “window.” Towards this end, we imagine the possibility of bringing up “sub-windows” or pop-up windows for display, toggled on or off. These include:

  • Pop-up windows corresponding to individual events (see below). These use our familiar sense of the pop-up window to convey the many levels of reality in the novel. All pop-ups first appear with the overlaid words “Loading.”
  • The text of the novel will scroll down the screen as if on top of the visualization. The text can be toggled on or off. The text displayed and scrolled will correspond to the visualization currently or most recently visited. However, the user may select the text and go to particular scenes, which in turn make the viewer leap to a new space.
  • A chessboard view can toggle on or off. Here the user can play against the computer. The visualization and text will jump or scroll in turn.
  • Finally, a section of the screen show the score, based on what encounters are achieved and how far along in the level the user progresses.

Game Space

The Looking Glass world is a vast chessboard, but the “map” of the navigable game space is as if the chessboard has been stretched, exploded, with mazes added between each space. A given space, meanwhile, must be assumed to have complexity and possibility of movement: while chapters in the novel in some way correspond to chess moves, many other events occur, indicating a much more labyrinthian and complex space. Thus, movement into a “space” in the chess game can correspond to considerable action in the first-person world.

We want to use the engine from one of the popular first person shooter games, probably Quake or Unreal. The project will work in the tradition of the game mod, but we want to pursue questions far from the usual concern of mod designers, for example: How does the graphical space relate to coding, on the one hand, and to the behavior and experience of the player or user on the other? Can we understand the game space as the figuration of an underlying narrative issue? Our hypothesis: we take these games as vision machines, creating movement, illusions of space and depth, all through certain codes. The content of the game is perhaps secondary to this mechanism, though it may be more correct to say that the first person shooter content is the particular cultural form taken by our immersion in this machine.

It’s important to note that Carroll’s works are out of copyright, a major advantage for our project. We can work directly with the text and the familiar Tenniel graphics. Carroll worked closely with Tenniel, and these artworks offers vivid illustrations extending the textual space – the illustrations are a fundamental part of our experience of reading the novel, of our memories of the book, and so on. The visual space of Tenniel’s illustrations extends the textual space of the novel. At the same time, they are fascinating in their abstractions: while figures may be very detailed, backgrounds fade into lines and curious patterns. We’ll replace the graphics of the game with the Tenniel illustrations. We want to start by removing all the graphical markers of the game (wall, door, etc.), leaving only geometric projections. Certain places we’ll replace with line-drawn trees and other elements of Tenniel’s illustrations. Other parts of the labyrinth will simply be empty.

The looking-glass world explores our cognitive movement from such a geometric projection to a fully realized “world.” The absence of images leaves only “space” or only “virtual environment,” not a particular virtual environment but the environment itself (though this can not be exactly right, as this will always be a geometrical spaces, however lacking in markers). We want to create a navigable labyrinth without familiar markers. We’ll use the vertical pencil lines and abstractions of Tenniel’s drawings to demarcate parts of this labyrinth. And there will be spaces where suddenly there is vision and event, as the user moves into the space of the novel’s chapters. But we imagine the reader experiencing a kind of disorientation and frustration, a looking glass experience. Alice spends much of the book wandering, moving through a flowing, hazy wilderness. Only gradually to spaces come into focus. This is also a question of effort on the user’s part, an integration of the user’s physicality: how to negotiate an invisible labyrinth to arrive at a space that can be seen and interacted with.

This question is further extended to the objects of the virtual world. Through the Looking Glass focuses on spatiality and transformation rather than stable objects – in fact, foreground objects are more often than not mutating, flowing, morphing, etc. The fluid object is the result of the fluid space. Interactions with these objects occur on the presupposition of a stable and coherent space. The objects are seen in the stable light and space of this presupposition. We will also replace the gun of the first person shooter with a hand. The novel offers numerous images and descriptions of picking up objects, chess pieces, etc.

The spaces of the labyrinth will be roughly as follows. Generic descriptions are in parentheses.

  • Environment (empty, void, labyrinth). This is the Looking Glass world, uniform and featureless. It should look somewhere between a blank screen and a blank page.
  • Brook (dotted lines in empty space): These mark the movements between spaces on the chess board.
  • Chess board spaces (parallel vertical lines): These correspond to movements in the game.
  • Event Scenes (described below: these are spaces with certain interactions/events, and more detail than the generic “chess board space”).

Event Spaces

The game starts with a first-person/ego view of Alice’s sitting room, in color and 3D. The user steps through the looking glass into the looking glass world, which is uniformly black and white – appearing like Tenniel’s drawings. The first space is always the “house,” by default. The user can then move into the labyrinth through various exits from the house.

  • House. An elaborate Victorian room. In this space, using the controls to look makes the point of view oscillate wildly. Only by using the controls wildly, as it were, can the point of view be stabilized. Exits are to the garden and to generic “environment.” Any attempt to move from the house to the garden renders the user’s point of view weightless, floating like a balloon. You can’t exactly control it, just guide like a balloon.
  • Garden. If you arrive at the Garden of Live flowers, you find that in order to go forward you must move backwards, and so on. All direction is reversed.
  • Shop. If you arrive at the shop, you find numerous objects that can not be grasped. If you look at them they are displaced. If you chance to grasp at one, it floats like a balloon. Eventually you find an egg, and with some effort can grab it. If you go outside the shop space with the egg, it will grow to gigantic proportions; grow into Humpty Dumpty (TBD).
  • Dining room. If you make it to this concluding scene, you will begin floating and growing larger than the setting. Eventually the game will end.


Other spaces are determined by the character present. While the default will set these characters roughly where Carroll’s chess problem puts them, there could be other settings. Actions and attributes describe how characters act when encountered. The appearance of characters will be based on Tenniel’s illustrations; we hope to import the actual drawings and wrap around a character in a product like Poser or 3Ds Max. Encounters below are assumed to occur when Alice enters the square adjacent to the character.

Currently the project is being developed using the Half-Life game engine, Hammer, and 3D Studio Max. Contact the CLC to learn the current status of the project.