Perspective on Text Adventures

I'm writing this in order to reduce the number of tangents I go off on in the Text Adventures of Note and Classic Text Adventures articles. This is more of a memoir (or something) than a curational list. At any rate, expect it to be a bit rambly.

I grew up in the 80's. I had a Commodore 64. I really liked video games. I wanted to write my own video games, but BASIC was too slow. So when I wrote my own games, they were often text adventures.

I say "text adventure" here because that's what they were called at the time, before Infocom began labelling its wares "interactive fiction".

But even nowadays I generally avoid the term "interactive fiction". I still call them "text adventures". Probably mainly because it doesn't carry the connotations that "fiction" does. There isn't an assumption that they have any plot beyond perhaps "collect all the treasures", nor any character development or other narrative elements.

It is a more basic form. It involves a computer giving a player some text, being a description of some kind of imagined environment, and then the player giving the computer some text, being a description of what the player would like to do in this environment. With this cycle repeating until some end goal is reached.

The word "adventure" also connotes that there is some kind of space being explored, that the player moves around in this imagined environment (in virtue). This is not a strong connotation, but the activity in most text adventures does span a variety of discrete locations.

I don't even usually call text adventures "games", but that's for reasons that probably aren't relevant here. One can make a strong argument that a text adventure is a kind of game, but it's also not necessary to look at them that way.

The requirement that the player communicates their intentions in a textual form disqualifies things like "Choose Your Own Adventure" books (or the modern equivalent, vanilla Twine games) from being text adventures, even though one could argue that they are a form of "interactive fiction", at least if one takes that term on its face value.

But one typically does not; interactive fiction is its own thing. I would define interactive fiction as the subset of text adventuredom which is assumed to have narrative elements (such as plot and character development and so forth) in order to be more like conventional, non-interactive fiction; to be more like a story.

But this property doesn't interest me greatly, and in fact I often feel that it detracts from the somewhat dreamlike effect that a narrative-lacking, unmotivated, and slightly incoherent world can have.

This definition I've given also does not exclude text adventures or interactive fiction from having graphics. When the graphical content consists of still pictures (possibly with minor animations only loosely connected to the gameplay), I do not consider this to be much different from a book with illustrations. The exchange between computer and player is still primarily text-based.

I don't remember which text adventure I played first. It was either Zork I or African Adventure: In Search of Dr. Livingston. It was likely the former. I think I had known about text adventures before playing one; possibly I had read about them in a magazine.

These two adventures were both very influential on me, but of particular influence was the Usborne book Write your own Adventure Programs for your Microcomputer. It was probably also partly responsible for setting me off in the direction of programming languages, because, among other things regarding text adventures, it describes how to write a simple one-or-two-word command parser.

And, once again on the point of nomenclature, I'd like to note that this book was frustratingly difficult to find again in adulthood, because I was convinced that it had the term "text adventure" in its title, when it in fact does not; it has the term "adventure program" which is not something anyone, as far as I know, ever calls them.

That isn't the only idiosyncratic thing about this book. It claims unequivocally that the territory through which the player travels in the game must be mapped onto a Cartesian grid. "Haunted House", the type-in adventure which appeared in the book and which the bulk of the book was dedicated to analyzing, had 64 locations in an 8-by-8 grid. Further, the book claimed that if you wanted to model a 3-dimensional world, with passages going over and under other passages, you'd need a 3-dimensional array, and that this takes a lot of memory ("48K is probably the minimum to make it worthwhile").

This was probably an intentional simplification, but it did not jibe with my experience, even at the time. The games I had played had quite irregular maps that would not comfortably fit in a grid, with one-way routes and up-and-down routes and even "bendy" routes (where west and south are "opposite directions", for example.)

I had seen a map of Zork I published in Electronic Games magazine, too, and it made this visually obvious.

I realized somehow (through I did not know the term "graph" at that point in time, and indeed the term "graph" seems to not have been standardized at that point in history, with some authors calling them "plexes" or "mazes"; see Microprocessor Programming for Computer Hobbyists) that the map could be represented by a graph with degree 6 for the four cardinal directions plus up and down (or degree 10 if you wanted the 4 diagonal directions too), and that such a map could be implemented in BASIC as a 2-dimensional array, like DIM EX(30,6), where one of the dimensions is the "location number" and the other dimension is the "exit number". Which is what I did, when I went to write my own games. (Footnote 1)

But other than that, there is a lot of sound advice in this book, for instance

It is no good putting the light behind a locked door and then putting the key in a dark room.


Another piece of advice was to put objects where they would be expected to be found - "knife in the kitchen, book in the library, axe near the woodpile".

However, that gets back to why I feel that text adventures are a different beast from interactive fiction. Discovering objects in places they don't belong can induce a frisson of surrealism. I remember several such instances in Zork I and African Adventure.

Along these lines, I could go even further — although I fear it might not make much sense, as I am still developing this theory. But you could bear with me, if you like.

A major part of most text adventures is interactively exploring the environment. They are, in some sense, conductors of psychogeography. In fact, even a linear, non-interactive novel, or a first-person graphical video game like DOOM or Myst, is a "conductor of psychogeography" in this sense. But text adventures seem to be better conductors than these media.

Why this is, is probably because a textual description of the environment engages the player's imagination and sensibilities in a way that visuals of it do not; and having the player wander across the map, in a largely undirected way, also engages the player's imagination and sensibilities.

It is not unlike a kind of virtual dérive.

Giving it a plot and other narrative elements, though, does not improve this property; it does does not improve the text adventure as a mode of exploration.

But as I said, I'm still developing this theory, and I'm not entirely sure what I mean by it yet, so take it for whatever it's worth. I just thought I'd raise the point.

What I do know is that, long before I had heard of the word "psychogeography", I used a term I made up myself, "locationness", to describe something extremely closely related. And one of the ways I could try to describe locationness is to say that a place has locationness if it would make a good location in a text adventure.


Footnote 1

Compare and contrast Gregory Yob getting fed up with the omnipresence of the Cartesian grid in seek-and-locate computer games, and designing Hunt the Wumpus to break out of that pattern.