Game talkAm I a game neoconservative?

 Posted by (Visited 6906 times)  Game talk  Tagged with: , ,
Aug 022011

I love arguing with Ian Bogost in public. :)

Every now and then someone objects to game design methods by arguing against “historical aberrance.” This line of reasoning claims that a particular trend is undesirable on the grounds that it is new and abnormal, unshared by historical precedent.

First, a few years ago Raph Koster invoked this argument about single player games. As Koster put it, “the entire video game industry’s history thus far has been an aberration. It has been a mutant monster only made possible by unconnected computers. … Historically speaking, single-player games are indeed an aberration.”

…following Koster’s retort, we could fault Heavy Rain for replacing human storytellers and listeners — who are good at making rapid judgments and improvisations based on different actions and their possible outcomes — and replacing them with a much coarser narrative simulation system that operates only according to the limited interpretations possible by a computer.

…Video games aren’t science. They are not a mystery of the universe that can be explained away via testable predictions and experimentation. We need to stop looking for answers

Gamasutra – Features – Persuasive Games: From Aberrance to Aesthetics.


Oddly, I am a fan of both Heavy Rain and Sleep is Death. The context of my original remark was at a business conference, not a design conference, and was aimed much more at shaking up preconceptions about the game industry than anything else.

I do believe firmly that single-player is fighting the tide, in that it works against some fundamental characteristics of the *real* canvas on which we work, which is the human brain. And I say this as a huge fan of single-player games. I think it is inevitable that single-player gaming drifts towards two poles: the interactive narrative and the puzzle, precisely because of this canvas. I also think it is inevitable that they will come to be wrapped, at all times, with multiplayer and social components — and I suspect that in the years since my original statement, this has gotten a lot less controversial than it once was!

That said, I will disagree with this statement: “Video games aren’t science. They are not a mystery of the universe that can be explained away via testable predictions and experimentation.”

I think they are, and this doesn’t preclude them also being an art. I think they are a mystery of the human brain that can be explained with greater knowledge of ourselves, and can have hypotheses proven or disproven by testable predictions and experimentation.

What’s more, I think that said predicting-and-hypothesizing is happening today at a very rapid pace, and that we are in fact learning more and more every day about an emerging science of game design.

The artists among us — a group in which I count myself! — can be and rightly should be troubled by this, because it evokes the spectre of a time when the market comes to be dominated by mathematically derived pablum designed to appeal to the lowest common denominator hindbrain triggers in our psychology, much like film (link, or see the orange-and-blue phenomenon) or music (see the soundwave analysis firms that predict hit-worthiness of music algorithmically) or graphic design or or or.

See, I am not advocating these positions. I am observing things, and arriving at conclusions. In fact, when I have engaged in advocacy, it has been to argue the case of art, for aesthetics, for broader influences and diversity — in fact, this exact topic is one I wrote about five years ago in a post called “The Algorithm or Art?” When I said at Project Horseshoe a few years ago that “I think games are math, and it worries me,” I really mean it.

I don’t think that greater understanding of color theory, golden sections, and perspective necessarily preclude there being art in the process of making paintings, though. It may well be that by taking up a given medium, though, we are choosing our shackles, choosing which constraints we limit ourselves with. Game grammar, theory of fun, social mechanics, etc, are just my attempt to explicate to myself, what the building blocks of this medium are.

That means I can enthusiastically sign on for Ian’s call “Let’s make games. Let’s make good ones. Let’s try to figure out what that means for each of us. Let’s help our colleagues and our players and our critics understand it.” But it also means that I disagree with Clive Bell, whom he cites at the end of the article, inasmuch as I do regard the tensile strength of clay as a essential and yes, exhaustible quality of the art made with said clay. My goal would be to turn that to strength rather than weakness.

  21 Responses to “Am I a game neoconservative?”

  1. I think I would agree that games are both art and science and that understanding either facet makes us better able to make them and appreciate them.

    I don’t think it’s helpful however to start throwing names like “neo-conservative” around. One’s views on the nature of games really don’t synch with one’s views on politics and let’s not compartmentalise and start hating other people in the community because we’re conflating political ideology with ludological positions.

    For example I think a lot of games are rubbish but I wouldn’t want to be labeled as if I were a person who believes a lot of people are rubbish.

  2. He seemed to think that “aberration” has the same meaning as “abomination”…

  3. Re: “Historically speaking, single-player games are indeed an aberration.”

    Well, ha to that. From 1980 (thru 93) I did the huge multi-player games, when the entire video games industry did only single-player.

    Now, I’m going to do my first epic single-player game, as it’s just what the world isn’t looking for.

    I hope to show that a single-player game (for iOS) can offer as much depth, fun and compelling long-term play as most any multi-player, MMORPG, or FB social game does.

    Just like the board game industry is thriving with games like Dominion and 7 Wonders, I think single-player games are just waiting for killer innovation (a la Minecraft) and I hope we see a bunch of great games that do this soon. Just like two-player games were brought back from the dead by Magic: The Gathering.

    Then again, a case could be made that most Facebook gamers are really single-player games. Your friends don’t really collaborate but rather help you achieve minor buffs and builds for the purposes of, essentially, viral marketing. If Farmville wasn’t really single player game, wouldn’t a group of players be working on one large farm together, rather than each having their own farm? Shouldn’t we really call these “Assisted Single-Player Games”?

  4. There’s no magic destroyed by understanding the science of your medium as thoroughly as you can, and much to be gained.

    The better you know the rules, the more effectively you can add to them, break them or totally rewrite them.

    Many students and more than a few professionals want to skip directly to the “breaking the rules” step without knowing what the rules are or why they became rules in the first place. I applaud iconoclasm, but know what you’re smashing, why you’re smashing it, how that affects the whole, and what you propose to put in its place.

  5. I realize this is a small point, and tangential to the overall statement being made, but certainly the issue of algorithmically generating content based on an understanding oh how the human mind works isn’t as big of a problem as it’d initially seem? If you make all of your content to the lowest common denominator all of your work becomes very “samey” and if that happens it loses all value as it becomes interchangeable and so there’s a race to the bottom and value falls out. Moreover, because everything becomes samey, people will start to recognize that, will master the pattern of it, and no longer enjoy it. Culture and media have a feedback loop with our brains; what we find value in is not some static value but an adaptive response to the totality of our experiences.

    So you end up with the same sorts of issues that Asimov’s prehistory generates; figuring out what everyone likes and then exposing that information to them causes what everyone likes to change. People will always crave novel experiences. So we can embrace the idea that games are math; human experience is it’s own sort of math too after all. What matters is what you’re building with that math. Whether that math is being used to benefit society, to enrich lives, spur thought. Or whether it’s pushing us into socially destructive patterns. And we have to understand the math, the science, in order to be able to tell the difference. Those algorithms need to expand from “what will sell more?” to “what effect will this have on people? What emotion will this create? What behavioral pattern will this reinforce?”

    To not look for answers means that you are willfully blind to the consequences of your work; that is deeply irresponsible.

  6. Single-player games are fighting what tide? The most-played game in the world is Windows Solitaire.

    Games are a science the way music is a science. Sure, you can get all analytical on its ass, but it don’t mean a thing if you ain’t got that swing. Anybody who makes music by looking at the MRI scans of people listening to music is no true musician. The same is true of games.

  7. Solitaire wins because of distribution not preference. If a multiplayer game with similar accessibility (similarly familiar and magically independent of an internet connection)were on every Windows pc in the world, would Solitaire still win? Maybe, but I wouldn’t bet on it.

    Using MRI scans to make music may not make you a musician, but what happens when the MRI guy becomes better at making music than the musician? I think that’s what Ralph is scared of, as we should all be.

  8. I’m not all that concerned about “mathematically derived pablum”. It seems to me that every medium goes through plateaus where most of the work becomes very fomulaic and static, because the “science” of what works and doesn’t is well-established. Then someone or something comes along and kicks over the apple-cart, and everything we thought we knew changes.

    All I’m saying is that if you know the apple cart well, you know where exactly to kick it for maximum effect.

  9. @splok, why should we be scared of that? What makes the MRI scan guy’s music less legitimate than the musician’s? The process they’d both need to go through to generate good art is about the same. The only difference is that the MRI guy has a leg up on having precise data about craft elements. Except that the scientists discovering this stuff only benefits the composers too; they can use the same understanding to improve their art and they’re probably better at making something meaningful. But even in a world in which algorithms replace humans for creative output, that only matters in so far as it means that composers are put out of work; the impact of having better music would only be a benefit to humanity as a whole. If you randomly generate the most powerfully moving song ever heard by humanity, it’s still the most powerfully moving song ever heard by humanity; how it was created is irrelevant, what it does to the people who hear it is really all that matters.

    @Yukon, Actually, that’s probably a better way to put it; algorithmic processes can refine any current trends to their apex but it’s difficult to effectively generate disruptive events with it. You need the inherently random elements of the creative process to break out of any local maximas you get stuck in, and as soon as you do that you’re back into uncharted territory again where the outcome becomes unpredictable.

    Risk aversion is the death of art, not scientific understanding.

  10. what happens when the MRI guy becomes better at making music than the musician? I think that’s what Ralph is scared of, as we should all be.

    It frightens musicians less than you might think. As said by others before, a person’s performance is as unique as a fingerprint EVEN when playing standard material. So you are talking not about a musician in general but a composer/songwriter/dittyMeister in particular. Even here, it takes talent to imitate well. The ‘science’ of music is pretty well understood and has been. Even the genre and how we can expect an audience to behave are pretty well understood, and yes, purposeful design and even algorithmic design are part of that if you want them to be. On the other hand, the guy or gal who sits down with the guitar etc. and writes the song based on emoting or reacting is still more common. I’ve listened to algorithmically generated music. Muzak is already composed to fixed rules based on measures of audience reaction.

    With all of the above, so far music hasn’t suffered although some genre are pretty well plumbed like mines in a mountain.

    What can happen is it does become muzak. Film and TV have analogs. Recently I blogged about my reactions to Eureka given someone I liked had a role. My trepidation was (in my opinion and you are free to disagree of course) is the series has already jumped the shark and adding this actor wasn’t going to help except in a burst of viewers based on the fan base. So far, I think this is right. Why? The show IS formulaic and IS demographically targeted and while that is probably working well for the demographic to which the formula is applied, it lost me. I’m not a 12 to 32 year old game player and that is what they are targeting.

    The formulas tend toward pastiche and when you slap elements into a formula without a care for fidelity of sources and resources, the formula won’t make it better. (sorry ms berg, but a redstone can’t only not do what you have it doing, it is the worst possible choice given what it could do and some of us know that). IOW, you can pretend the math will produce a compelling piece but unless you really do work the sources and resources, you will still fail.

    From this point of view, I think Bogost has a point and perhaps fear of what comes next is a little over the top, or is at least, the wrong thing to worry about. At the end of the gig, games are like any other art for the audience: if they play, you succeed. If they don’t your back in the woodshed. The MRI guy is very unlikely to beat the experienced game designer until he becomes one and then the MRI is just another source and produces more, perhaps better, resources or maybe not. The MRI is just that and likely nothing more. If you don’t know that a Redstone is simply an uprated V2 and even with magic engines can’t put a payload in orbit (crappy superstructure), you used the wrong resource because you used the wrong source. And the inconsistencies matter to the perception of the performance and the piece.

    I’m a fan of guitar tuners. Save time; improves the overall performance by removing inconsistencies. If you used and MRI to tune a game, you probably did the right thing with the source.

  11. @Ernest Adams

    We just don’t see the science in music because it was largely unintentional and spread out over centuries. But don’t be tricked into thinking that those centuries were not centuries of observation and discovery of what works (appeals to the human mind) in music.

    The fact that it may have lacked rigour doesn’t make it essentially unscientific, it just makes it really inefficient and slow :-)

    Is it a requirement to be a musician that you throw away major/minor scales? No great musician makes music without learning the rules first.

  12. All the talk about music is fine. Care to apply the same scientific analysis to the novel? Can a crack MRI scanner operator knock out Crime and Punishment or The Great Gatsby to order?

    For that matter, just returning to music, are the models derived from MRI scanning going to explain why I’ve gained so much pleasure over the years from Yoko Ono doing “Don’t Worry Kyoko, Mommy’s Only Looking For Her Hand In the Snow” on Live Peace in Toronto or all four sides of Lou Reed’s “Metal Machine Music” or Suicide’s “Frankie Teardrop”. Or any of a hundred or more other similar examples?

    Explaining how something works doesn’t explain how it works, you know!

  13. I dunno, you don’t need to be a biologist to butcher a cow.

  14. I think Italo Calvino’s novel If on a winter’s night a traveler… covers the novel-writing computer question pretty well. At some point, readers want to see a human face in the book, and they want to know some other human is reading it. Knowing that a book has an author and was written by someone of your species gives you the hope that you could perhaps write something similar, or at least gives you some feeling that a machine can’t. People’s tastes differ and do not inevitably coalesce. So, the MRI will not come up with a piece of music that will last forever – it can only read the signs of the times, for one’s tastes are a product of the times. Eventually, if the MRI gets incredibly good at “creating” songs for specific people, it will be more like a machine that allows a person to create a song by thinking about it. Or, at least, it will be nigh indistinguishable from that.

  15. @bhagpuss Changing the medium just changes the methods, not the overall point. There’s a craft to writing too. And as len says, the MRI scanner operator can’t knock out Crime and Punishment without also being a good author; science only helps us improve craft, it doesn’t so much tell us what to do with that craft.

    Now, a sufficiently advanced algorithm could replace the creative entirely, but it’d do it by behaving like the creative. And it’s output would be of just as much value as a human’s output. This is where we could be surpassed, but we’re talking singularity level AI here; it’s not really worth talking about too much till it happens.

    As to neuroscience eventually explaining why you’ve derived so much pleasure from those songs, sure. We’re not that far away from having a pretty coherent answer to that right now even. A lot of the specifics are missing still, but the general idea for how music impacts the brain is there, and that does answer your question. For your brain, that music is “shaped” correctly to generate that response. Your brain being different than everyone else’s and changing constantly, getting specific details would be highly individual, to the point in time in which they were gained, but it is in principle possible to answer that exactly.

  16. There are constants in music that transcend time and genre. And the more science understands the interactions of sonic waveforms and human perception, the more we see why these constants work.

    We can plug the constants into a computer program, select a genre, and produce a MIDI file that is complete and recognizable as a piece from that genre.

    But… it’s bland. Even if the MRI confirms that we’ve got the optimal elements to produce a pleasant sensation in the majority of people, there’s something missing.

    And that something is unpredictability. Our brains rapidly satiate on things that are predictable, to the point that optimal solutions become sub-optimal because we’re bored with repetition.

    Turning back towards games, I’m a firm believer in the potential of procedurally-generated content. But I think for it to work best, it’s going to have to be seeded and regularly updated with novel human-crafted elements. By melding the science and art, we can resolve the end-game dilemma that it’s always faster to consume content than to create it.

    Science is not the enemy of art, but an enabler. And while we don’t have to use all the new tools, mediums and knowledge it affords us, I for one am not anxious to return to banging out rhythms on a hollow log and sketching stick figures on cave walls with charcoal.

  17. Or: but fur Science, how many vibrant pigments would Art be without?

  18. As Eorlin & Yukon Sam have brought out, there’s a risk of conflating *creative* applications of science with *predictive* applications.

    Like NP problems in computer science, verifying the hitworthiness of a song (a solution) is a vastly different cup of tea than crafting one from scratch. One might be trivial and the other very hard.

    1)Are these primes factors of this number?

    2)Can you factor this number into its prime factors?

    Two very different qustions….

    Just because science can help us understand our building blocks (Raph’s original point) doesn’t in any way imply that it is an answer for creating or generating new stuff, for music, writing or game-design.

    The best AI is, at the bottom, seeded with human creativity and is regurgitative or simply random rather than really creative.

  19. Paraphrasing a current hamburger commercial: “Machines don’t enjoy mucic. They shouldn’t write it.”

    Probably true of some games.

  20. I always view game design as an art and game publishing as a science. They’re intrinsically linked though, like art and science in general, as the medium of the expression is an extension of the materials which came from science.

    Same with games. Iteration follows success of a concept. Sometimes it’s the concept itself regardless of any early business challenges. Unfortunately, more often than not, the concept needs to have business success in order for publishers to fund the iterations on it. It is lowest common denominator thinking in that the funding goes to the higher probability purchase (or, money follows money). But that’s more an insights of the audience being targeted than on any fundamental issue with game design itself.

  21. […] Critics who want to use the term interactivity to focus on the subjective experiences of players, who see interactivity in every cognitive interaction with a text, who want to write games criticism that is personal, are all focused on the inner life of the player. This is a very important thing to focus on; the much-maligned allegedly formalist game grammarian would tell you that the canvas of a game is the human mind. […]

Sorry, the comment form is closed at this time.