Interactive Storytelling for Videogames

by Josiah Lebowitz and Chris Klug

This will not be an unbiased review. I wrote a book on interactive storytelling, so I suppose that I’m a competitor. The reader should take careful note of that. But the authors of this book added insult to injury by failing to mention me, my book, or any of my other works. So this review will be even further biased by a bruised ego. Fortunately, my book is outselling theirs, so I can at least take some comfort in that.

I can also claim some expertise in reviewing this book; after all, I
did write a book on the same subject, right? So I probably know a thing or two about interactive storytelling.

You may be surprised, then, to learn that I give this book a qualified recommendation. I won’t trash the book overall: it does a good job of doing what the authors set out to do.

I’ll begin by pointing out that our two books operate under very different definitions of the phrase “interactive storytelling”. They never come up with a clear definition of what they mean by it, although they do present a goodly number of characterizations that make their meaning clear. To them, interactive storytelling is something you build into a videogame to heighten its entertainment value. It’s backstory, justification for the player’s actions in the game, and a dramatic context for those actions.

That’s not my definition. To me, interactive storytelling is storytelling that is interactive. In other words, the user gets to interact directly with the storyworld, and that interaction ultimately yields a story.

A storyworld is a complete, consistent, and circumscribed set of dramatic elements (actors, props, stages, verbs, traits, and the principles that govern the behavior of the actors). A storyworld is to a story as a calculator is to a number. Obviously, we’re talking about something very big and complicated. But the user interacts with the storyworld, not the story. You can’t interact with a story – it’s data, and you can’t interact with data. You can interact only with processes, and storytelling is a process.

My definition of interactive storytelling is lots more high-falutin’ than theirs. And so far, I have never actually gotten a good storyworld running, so obviously I’ve got my head in the clouds, while their feet are firmly planted on the ground.

Their notions of interactive storytelling follow the conventions of videogames; they describe five different strategies for weaving a story into a videogame. The least interactive of these simply tacks an unchanging story onto a videogame. As you play the game, you get snippets of the story, but nothing you do affects the story. This, the authors correctly note, is not interactive.

The next step up their ladder is just like the previous step, except that the player gets to choose among several endings. I consider this tiny degree of interactivity to be insignificant, but they devote a full chapter to it.

Third in the sequence is the branching tree approach. They point out that it can be expensive to provide lots of choices to the user, as each option requires lots of artwork, music, and video. They never seem to appreciate, however, that giving choices to the user is the whole idea of good software. Moreover, their discussion of branching trees is a bit primitive; they don’t seem to have developed a clear idea of all the possibilities and ramifications of branching tree structures.

They call their penultimate step “Open Ended Stories” and here they lose their way. What they’re really describing are games in which the player has a large environment to explore, but there’s still a traditional linear story that plays out as the player makes progress towards the final objective. The environmental exploration reveals bits and pieces that help the player advance, but it contains no dramatic content per se – a point the authors seem to miss. These aren’t open-ended stories; they’re conventional, non-interactive stories set inside an interactive playground.

The highest level of interactive storytelling is “Fully Player-Driven Stories”. They offer
The Sims as an example of this category. The player is free to wander around an interesting environment, doing interesting things, but there’s no story at all – at least, none in the conventional sense, and certainly no inkling of a story designed by an author.

We differ big time on the role of player agency (which they call player control). The basic question is “How do you give the player the opportunity to do dramatically interesting things without ruining the plot?” I first answered this question more than twenty years ago in a lecture in London. Then I wrote up the answer in an essay I published in The Journal of Computer Game Design, which I published on this website in 1998. In fact, I’ve written a number of essays on the question: Plot Versus Interactivity, The March of Abstraction, and Interactivity, Plot, Free Will, Determinism, Quantum Mechanics, and Temporal Irreversibility. I continued to explain the answer in numerous lectures, and then I published it in my book on interactive storytelling in 2005. I don’t want to be too harsh on the authors – everybody makes this mistake and nobody has come up with the answer I came up with. Moreover, my answer is quite esoteric, so I can understand why this cumbersome meme has not spread far. They probably wouldn’t understand it anyway, and neither would their readers. We’ll just have to wait a few decades for my more abstract approach to start making sense to people.

Another thing that bothers me is that their minds are firmly locked into the videogame gestalt. For them, the core of gameplay is battle mechanics. The “protagonist” to them is the “hero”. Games are composed of quests. These authors don’t just fail to think outside the box: to them, the box is the whole universe.

A good example of this attitude is their certainty that the only way to present a story is through lots of video, voice acting, and other expensive cosmetics. This is, of course, the only way that story is presented in the current generation of videogames; perhaps it’s the only method that the videogames market will ever tolerate. Still, it is disappointing that the authors do not even contemplate other, less expensive means of presenting a story.

I was disappointed by the authors’ failure to mention a simple and fundamental truth about storytelling: it’s about people, not things. The fundamental content of stories is about people: how people feel and interact with each other. It should be obvious that interactive storytelling should be about interpersonal activity, yet these authors seem to think that the only ways to interact with characters are 1) to fight them; 2) to get their aid in fighting somebody else; or 3) to obtain information. How about things like romance or friendship? The authors are happy to provide it non-interactively – but interactivity is for fighting, not loving, in their universe.

The authors certainly understand storytelling – probably better than I do – but they don’t know diddly about interactivity. Whenever they start getting serious about interactivity, their hands start waving around, their verbiage thickens, and their content thins. Just once in the entire book did they come close to nailing something important about interactivity: when they talked about concentrating on what the player does. But then they sped onward without realizing that they had stumbled across an important idea.

My harshest criticism of the book is that the authors did not bother to do their research. If you go to Google and enter the search phrase “interactive storytelling”, the first four links you’ll find are:
1) the Wikipedia entry, which is full of references to sources they missed;
2) my personal website, which has tons of material that they apparently never read;
3) my business website, which has a LOT of technology regarding interactive storytelling;
4) my book on interactive storytelling;

I’m not the only glaring omission in their research; they also fail to mention the following important sources:
www.grandtextauto.org, the best blog about interactive storytelling;
Facade, the first working storyworld ever created;
Michael Mateas or Andrew Stern, who created Facade and published lots of stuff about it;
the Oz Project at Carnegie Mellon University in the 1990s
The ICIDS conferences (they do reference one of the proceedings volumes, but mention absolutely nothing from any of these important conferences.

Their treatment of interactive fiction is also criminally curt. I don’t think that IF is the answer to interactive storytelling, but I at least take the time to describe it and discuss its strengths and weaknesses. And certainly anybody interested in interactive storytelling should expose themselves to some IF.

Need I go on? These fellows just didn’t bother to learn about what other people were doing. And people say that
I’m living in a cave!

Another area of mistakes is their history of interactive storytelling. For example, they declare that
Ultima was the first computer RPG. Ultima appeared in 1981; Temple of Apshai from Automated Simulations appeared in 1979. Wizardy from Sir-Tech appeared in 1981 as well, possibly earlier than Ultima.

They also claim that
Donkey Kong was the first platform game. That is incorrect. I don’t know what the first platform game was, but I do know that Space Panic was a platform game and it came out well before Donkey Kong.

They further claim that
Donkey Kong was the first game to include multiple level types. I’m not sure what they mean by "multiple level types", but the basic concept of game levels goes back at least as far as 1977, and by 1980, well before Donkey Kong, there were a zillion variations on the basic idea of different game levels. I very much doubt that Donkey Kong can claim any firsts in this regard.

They also missed an important process in the evolution of storytelling in games. As early as 1979, with
Starfleet Orion, games were trying to incorporate storytelling. At first, the stories were confined to the box copy, with a few of the more ambitious efforts having stories included in the documentation. Yar’s Revenge included a little comic book providing the backstory for the game. My own Excalibur included an entire novella. As games grew, the story was brought off the printed page and into the computer. Two of the games that played a major role in this process were Wing Commander and The Seventh Guest – neither of which the authors mention.

Despite my many criticisms, I refuse to condemn this book, because it serves a useful purpose: it shows how to write stories to embellish videogames. There’s really no overlap between their book and my own. To caricature the difference, theirs is like an automobile repair guide ("Use the 3/8" box end wrench to loosen the framajig...") where mine is like a treatise on thermodynamics ("If we integrate across all temperatures, we obtain...") They show the nuts and bolts of working with the current technology; I explain the underlying principles. If you want to get a job doing this stuff, read their book; if you want to understand interactive storytelling, read mine.

Response from Josiah Lebowitz

I invited Josiah Lebowitz to comment on this review and he had a few comments to make. I have rendered his quotations from my review in the same style as this text; his quotations of this review in purple, his reactions to them in black and my own responses in blue.

It's clear that we operate under rather different definitions of interactive storytelling. But, as I said in my book, there is certainly room for differences of opinion in that regard. That said, I do have some points I'd like to make


"But the authors of this book added insult to injury by failing to mention me, my book, or any of my other works. So this review will be even further biased by a bruised ego. Fortunately, my book is outselling theirs, so I can at least take some comfort in that."

While it didn't make it into the Bibliography, I am familiar with your book and, when negotiating a publication deal, had to repeatedly explain to my publisher how my approach differed. As for sales numbers, going entirely on Amazon rankings...I'd say it depends entirely on what day you happen to check. There have been times when my book has been in the lead.

That explains many of the things I complained about. If the publisher wants to clearly differentiate your book from mine, that makes it harder for you to address the issues I covered in detail.


"Third in the sequence is the branching tree approach. They point out that it can be expensive to provide lots of choices to the user, as each option requires lots of artwork, music, and video. They never seem to appreciate, however, that giving choices to the user is the whole idea of good software. Moreover, their discussion of branching trees is a bit primitive; they don’t seem to have developed a clear idea of all the possibilities and ramifications of branching tree structures."

Giving the user choices is the point of many kinds of software, such as an OS, photo editor, or word processor. But they're fundamentally different than games. I believe that the whole idea of a good "game" is to provide the user with an enjoyable and/or thought provoking experience, much as with a book, film, or traditional game. At times, providing a large amount of choices can be the best way to go about that but in many (or even the majority) of cases I believe that too many choices hinder and even detract from the overall experience. Of course, user preference can play heavily into this debate as well.


As for branching trees, I feel that they were covered in adequate depth. I could have gone into more detail but due to the primary target audience (university students) and page limit set by my publisher, it didn't seem appropriate to devote any additional pages to them.

"Another thing that bothers me is that their minds are firmly locked into the videogame gestalt. For them, the core of gameplay is battle mechanics. The “protagonist” to them is the “hero”. Games are composed of quests. These authors don’t just fail to think outside the box: to them, the box is the whole universe."

I'd like to point out that the title of my book is Interactive Storytelling for Videogames, so the focus on games and gameplay really shouldn't be so surprising. On top of that, I do talk about some games which don't have any battles (Monkey Island, Flower, Farmville, The Sims, etc.). I don't devote significant portion of the book to them because the book is intended as a practical introduction and guide to writing game stories and, at present, the vast majority of games include battles and/or similar types of conflict. That said, my original plans for the book were a bit more academic and would have covered such topics in more depth, but my publisher preferred something targeted more towards university students, so some of that material had to be left out.

Again, if the publisher wants it to be a textbook for use in training classes, that severely restricts the range of material you can address. This suggests another difference in goals between our books: yours is more for training, mine more for education in the classical sense.



"A good example of this attitude is their certainty that the only way to present a story is through lots of video, voice acting, and other expensive cosmetics. This is, of course, the only way that story is presented in the current generation of videogames; perhaps it’s the only method that the videogames market will ever tolerate. Still, it is disappointing that the authors do not even contemplate other, less expensive means of presenting a story."

Other methods are discussed at various points (in-game books and other writings and player imagination, for example), if not in great detail. Once again though, as the book is meant to be a practical introduction and guide, it didn't make sense to devote much time to methods that even you think may never catch on.

"I was disappointed by the authors’ failure to mention a simple and fundamental truth about storytelling: it’s about people, not things. The fundamental content of stories is about people: how people feel and interact with each other. It should be obvious that interactive storytelling should be about interpersonal activity, yet these authors seem to think that the only ways to interact with characters are 1) to fight them; 2) to get their aid in fighting somebody else; or 3) to obtain information. How about things like romance or friendship? The authors are happy to provide it non-interactively – but interactivity is for fighting, not loving, in their universe."

I'm not really sure how you missed it, but the book includes numerous examples of interactive romance and friendship, such as the case studies of Fate/Stay Night and Star Ocean. I too think they're very important elements of interactive storytelling and made sure not to overlook them.

Here’s another place where our differences in definition cause misunderstanding. My definition (the player interacts dramatically with characters), combined with the observation that the book doesn’t address how the player can interact dramatically with characters leads to the conclusion that it missed a critical element of interactive storytelling. Your definition emphasizes the drama of the story itself, and has no problem with those parts of the drama being non-interactive.


"My harshest criticism of the book is that the authors did not bother to do their research."

I'm going to take a bit of offense here. I spent months researching all manner of books, websites, and other resources, including many of the sources that you mentioned. Just because I didn't feel an item contributed enough to the book to be included in the bibliography, doesn't mean I overlooked it.

Oops. Mea culpa. I should not have made assumptions about your process; I should have confined my comments to the book itself. Still, the absence of some of these sources from the bibliography seems amiss.



"Their treatment of interactive fiction is also criminally curt. I don’t think that IF is the answer to interactive storytelling, but I at least take the time to describe it and discuss its strengths and weaknesses. And certainly anybody interested in interactive storytelling should expose themselves to some IF."

I originally planned to do a bit more on IF but it was cut due to length requirements and the fact that, in its current form, IF seems unlikely to ever break out of the little hobbyist niche it has fallen into. Once again, I'll point to the primary target audience and goal of the book. More time spent on IF would be interesting from a historical and academic perspective, but it wouldn't be very practical. I think the fact that I have an entire case study on Colossal Cave shows that I am giving IF its fair share of credit.



Another area of mistakes is their history of interactive storytelling. For example, they declare that Ultima was the first computer RPG. Ultima appeared in 1981; Temple of Apshai from Automated Simulations appeared in 1979. Wizardy from Sir-Tech appeared in 1981 as well, possibly earlier than Ultima.

You've got me there. I'll pass the blame on to the various video game history books and web sites I referenced, which made no mention of Temple of Apshai and listed Ultima as the first computer RPG.


They also claim that Donkey Kong was the first platform game. That is incorrect. I don’t know what the first platform game was, but I do know that Space Panic was a platform game and it came out well before Donkey Kong.
They further claim that Donkey Kong was the first game to include multiple level types. I’m not sure what they mean by "multiple level types", but the basic concept of game levels goes back at least as far as 1977, and by 1980, well before Donkey Kong, there were a zillion variations on the basic idea of different game levels. I very much doubt that Donkey Kong can claim any firsts in this regard.

To quote from Space Panic's Wikipedia article:
"Space Panic is a 1980 arcade game designed by Universal, which Chris Crawford calls the first ever platform game,[1] as it pre-dates Nintendo's Donkey Kong (from 1981) which is often cited as the original platform game."

I like how they mention you specifically. Anyway, I think it's really a matter of personal opinion about the definition of platform game and which history books you read.
As for multiple level types, my definition there is that the actual level layout and style changed significantly in each level of Donkey Kong, while in past games you basically just kept playing a harder and harder version of the same level. Though I'll note once again that, while I'm fairly knowledgeable about game history, I never claimed to be an expert on the subject. I got much of my information from various reputable sources. If they made a mistake, chances are I wasn't able to catch it.

It’s certainly true that many of the history books about games have glaring holes in them. Sorry for picking on you for their mistakes.



"They also missed an important process in the evolution of storytelling in games."

I'm familiar with all the games you mentioned and yes, they did play an important role. The problem was that I only had one chapter in which to talk about game history. And, in its current form, that chapter still has a higher page count that it was supposed to. I had to make a lot of cuts and skim over or ignore many games I would have really liked to talk about.


"There’s really no overlap between their book and my own."

Glad you agree. Both I and my publisher wanted the book to be different than anything that was currently on the market. As you point out, our books serve different purposes and both have their uses. I've recommended your book in the past (even after the publication of my own), with a few reservations, and am glad to see you doing the same for mine.


As a final note, I'd like to point out that you don't say anything about the last few chapters of the book, which cover my research into player story preferences in games. Perhaps because more traditional and linear story structures proved to be a fair bit more popular? Regardless, as the market has shown, there are room for all types of interactive stories. We'll only be in trouble if everyone starts thinking the same way.

I didn’t address the polling research you did because, In My Arrogant Opinion, I really don’t think that the customers can tell us where we should go in the future. Besides, these are video gamers; I want to reach the larger set of people who don’t care for video games. I agree that market research of the kind you did is of importance for the here and now, but not for anything beyond a couple of years into the future. As I wrote in the review, I’ve got my head in the clouds and you have your feet on the ground.