Lost Garden: What are game mechanics?
It’s been a while since I pointed to Dan Cook’s excellent Lost Garden blog. Today I see a post up that takes ideas from AToF and several other sources, and along the way, produces a great flowchart that I arguably should have had in the book:
This flowchart is very much like the basic outline of one of my “atoms” in the grammar of gameplay. It doesn’t include the notion of topology of space in the model (e.g., the relationships between tokens) and it doesn’t include the notion of atoms embedded in atoms. Embedded atoms would eixst right in box #1; where he says “using an available tool” there may be many whole mechanics replicating this chart exactly, as in embedded mini-games, etc. The nesting thing was one of the things that really complicated my attempt to make a decent notation, though. Finally, there’s no discussion of the mechanic that offers no choices, the case where there’s a tool you use that has no chance of failure; that’s the box that I notated with two lines at one edge.
Also something that should be caught is Tadhg’s comment in the comment thread, with its subtle and interesting distinction between “rules” and “mechanics” — the former are the physics of the game model, unalterable by users; the latter are the knobs the user can turn.
40 Responses to “Lost Garden: What are game mechanics?”
Sorry, the comment form is closed at this time.
[…] Comments […]
Embedded atoms
Iām using the concept of referencing tools, instead of embedding. When the same tool is made use of in multiple mechanics, embedding results in your diagram unnecessarily replicating mechanics. If it is an atom, it should be atomic. š
There should really be an arrow coming out of other game mechanics into stage #1. Another useful diagram would be one showing an example network of atomic game mechanics.
Mechanics that offer no choices
All mechanics that require player action require a choice. The player can perform an action or not perform an action. Even though the behavior of the simulation may be incredibly simple (ex: Press button and the blue light goes on), the player still goes through a learning cycle the first time they press the button.
In more traditional UI design, this problem comes up a lot. Many obvious UIās on consumer electronics are stymied by the fact that the user doesnāt know what button to press or sometimes doesnāt even know something is a button. You still need to use proper iconography to alert the player to its presence and consider appropriate feedback when the state of the device changes. Just because the designer thinks something is trivial doesnāt mean that you shouldnāt go through the check list to ensure that the user has everything they need to master the mechanism.
If a mechanic really doesnāt require any player interaction, I would record it as an internal rule of one of the simulation mechanisms.
Rules vs. Mechanics
This distinction is also made in the model. Action are the āknobs that the user can turn.ā The rules of the simulation are the āphysics of the game modelā.
What I find delightful about all this is that there are clear parallels across most of the major concepts that are being discussed here. Certainly some of that comes from the fact that the egg head game design community is quite small and incestuous. š But the whiff of convergence gives me hope that there is a shared foundational philosophy that can be codified from these discussions.
take care
Danc.
Personally I think the insight is:
“Games are systems. Players have fun by learning how to use those systems.”
Anything beyond that seems well, to be taking it too far. The specifics of the system, it’s “grammar”, etc., changes from project to project. As soon as you start creating the flow charts you realize that you have a widget here where you might want a thing-a-ma-jig instead if you are doing a different type of game. Maybe your underlying system is a finite state automata, a context free grammar, a Turing machine, a complex dynamic system, or whatever. Whichever it is, you will then find a grammar to describe it that is appropriate for its specifics.
I think there’s huge benefits to be had in going a little further… you yourself use the example of “…maybe your underlying system is a finite state automata, a context free grammar, a Turing machine, a complex dynamic system, or whatever.” That right there is a sort of grammar in itself, where you are using the lingo of other fields in order to describe a given mechanic. So in your example, you already DID go further, to my mind. š
I was just doing some mental wandering the other day (http://www.mxac.com.au/drt/Stickybeaking.htm) and came up with some interesting thoughts.
– Games are about learning/understanding similar to neutral networks… which means you need a feedback loop that returns the amount of success/failure… and then using it to solve a problem.
– Puzzles are about learning/understanding by deduction, without scalar feedback loops. Puzzles only have binary feedback loops, which aren’t much good to neural nets.
– Stickybeaking (Australian term) is about wandering around the world, looking at the pretty scenery, talking to the NPCs, poking your nose into other people’s business, etc. It’s about learning (not necessarily understanding) for the fun of learning.
– And some other bits not relevent to here.
In other words, games are just part of what’s going on. To understand the whole system you need to look at more than “games”, and also at puzzles and stickybeaking.
That is not a grammar. It is a set of labels denoting concepts. A grammar would be a set of rules that could be applied to express all possible concepts (in this realm). There’s a world of difference.
We can accumulate knowledge and we create CERTAIN grammars for certain, very specific systems and we can come up with sets of solutions for specific problems. But creating an all-inclusive grammar? I don’t think that makes much sense, let alone is useful.
Let me be a computer scientist for a little while. Turing Machines can be said to be a “grammar” for computing (actually in this context, they are more than a grammar but we’ll use that word). Anything that can be computed, can be computed on a Turing machine (after a fashion). Said differently, Turing Machines are a grammar that can express all computations.
First of all: we can prove that. Given a certain definition of computation, we can prove that Turing machines can express all computations. We can prove it because at this level, computation is a very formal entity. Which is half the problem. Does, for example, a knowledge of Turing machines help a person program? The answer is “no, not really” (with some slight caveats). You might learn some other skills on the way to learning about Turing machines that would help you be a better programmer, but just knowing this “grammar” of computing really does little to help you actually program.
The use of Turing machines isn’t at all to help you perform computations. It is to help you prove things about computation. The closest it can come to helping you compute things it that you can use it to prove that a certain set of things can’t be computed or that a certain set of things can’t be computed with a certain degree of efficiency (i.e. in polynomial or exponential time).
The whole study of how to actually do effective computing, to do things like run servers, store numbers in databases, calculate numbers, draw graphics, interact with a user, etc., has essentially nothing, whatsoever, to do with this “grammar” of computing. And you can’t apply a grammar to computing in general. There are lots of formal systems, or grammars, used to do things like databases, for example, but they have nothing to do with the overarching grammar. They are just defined, ad hoc, to be useful for that specific problem. There are other ad hoc systems and methods defined for graphics, programming language design and compilation, scientific computing, UI development, software engineering, etc.
That’s why the computer science field has dozens of well-established subdisciplines most with several well-established subsubdisciplines. And in 50 years, when we are computing completely different things, it will all have changed.
Game development is the same. Worse really. We can come up with a relatively straight-forward and useful definition of computing. We can’t really do that for playing games because it is a subjectie experience. What we do then is develop systems and frameworks (or grammars) for lots of different applications. And for different subimplementations of the same application. And it’s a completely iterative process that bares little relevance to any overarching “grammar” whether we had that grammar or not.
The example I am more specifically referencing is that of an overarching mechanic (kill to earn XP, build base) that may be broken into subgames, or may form part of larger games. All game systems can be seen as nested challenges, essentially, because all game systems tend to have long-term and short-term goals.
This, I agree with. But some player actions have no risk of error, no failure state. That’s an atom (like, say, pressing a button) that is therefore no longer going to nest anything within itself. There is no shorter-term goal to that action, and it’s guaranteed to succeed. Physically moving a chess piece is the example I use — it’s truly a fundamental atom, because we don’t actually allow the player to fail at it.
I did say “sort of.” š I wasn’t being very precise there (blame this head cold). What I mean is that the list of system types you created there is a categorization, and may end up being a core part of our equivalent of “parts of speech” from a systemic point of view. We’re a long way from a formal grammar here, I agree. But I’d suggest that even simple labels denoting concepts are of great use with an analogue to an analytic grammar, as opposed to a generative grammar. I’m not pretending to suggest that we have anything near a generative grammar right now.
But even analytic grammars can help in the generation process; that’s how some spellcheckers work.
There is a category error here.
Grammars have nothing to do with meaning or definitions. Just structure. The difference between a generative and analytic grammar really doesn’t have much bearing here. A grammar is a set of rules like: “a valid English sentence has a subject and a predicate” (in analytic fashion) or “sentence :: ” (in generative fashion). It cares not what those subjects and predicates might be.
Just as a grammar of gaming is not really about the types of systems that can be embedded in a game it just says that there is a system of some sorts. Which is why I say that I don’t think you can go much further than the statement:
“Games have embedded systems and players have fun learning these systems.”
There is a certain amount of insight in that statement but to go further is going to happen at a more applied level. Developing a list of systems, understanding how they can be used, etc., is not an exercise in grammar creation but definition, classification and well applied research.
Hmm, what I meant was that we’re instead at the level of defining the equivalent of subject and prediicate.
If you’re talking about specifying which are valid subjects and predicates then this isn’t an exercise in grammar-creation (rather it is definition and classification). Dictionary writers aren’t grammarians. Dictionary-writing is an applied field. It is centrally concerned with analysis of real world events (language usage).
And I do think that’s where you are. You know that your “subject” is systems. And that’s about as far as you can go with this usefully, from a grammatical standpoint.
Maybe I’m just complaining with the title of all these “grammar of gameplay” discussions. I’d be happier with “analysis of gameplay” or something that inferred that you were looking at current patterns and not trying to create an entire theory of all possible games.
That’s how existing notions of grammar started — from analysis. š
I agree. However grammer creation is the creation of a syntax from the understanding of a structure. When we get down to looking at particular systems such as a complex dynamic system we move from talking about syntax to talking semantic. The grammar bit is: games are based on underlying systems. The dictionary part is creating a glossary of these systems (as we currently use and understand them).
Great conversation….I’m working on the alphabet (data points)…of games it would appear…still you need an alphabet to make words, phrases, etc….
Conjunction Junction pops into my head for some reason….
To a degree, I’m going to agree with StGabe… but then not.
As a writer (both vocationally and avocationally), I’m always somewhat… concerned… when “the grammar of X” is whipped out as a metaphor. Most people don’t know more than the very basics of grammar, and then they don’t even know the “why” of grammar’s rules, the history, etc. I’m no fantastic shakes at it, but I’ve got some background. And I also know the most important thing there is to know about grammar when it comes to writing:
What works works, regardless of rules.
Spelling errors are one thing. If I tell you I’d like to see your wife’s ass up close… and what I meant by “ass” was “sash,” well then we have a fundamental, “atomic” problem. If on the other hand, I say:
“Lasting joy can be found in the arms of comfortable rooms here in luxurious Buena Vista hotels.”
There is, frankly, nothing wrong with that from a grammatical or spelling standpoint. It is, however, bad writing. And part of what MAKES it bad is not a breaking of grammatical *rules*, but an understanding of how grammatical constructs impact the assumed *goals* of the writing.
For example, there is no real subject. The “You” is implied, and that’s OK. But it makes it a weaker sentence. The gerund (“lasting”) is also a much weaker verb contstruct than other types of verbs, as it turns an action word into a descriptor. Also, “can be found” is passive voice, which is, again, weak writing. No rules broken. Just poor writing.
So why seek a grammar of games? Grammar is set up mainly to dictate “these forms are the ones which must be used” and “these forms may not.” The grammar will be vastly different in languages that are inflected vs. ending-based. Some languages have almost no “grammar” as we understand it, as pronunciation takes the place of position in meaning generation.
I understand the point of the exercise. Categorization of parts is an important aspect of learning our systems. But the metaphors we apply often dictate our thinking. And the common understanding of “grammar” has implications that may drive design behaviors of which you might not be so pleased.
Oh. Never mind. Ended up totally agreeing with StGabe. I had a little, niggling point to make on the other side, but talked myself out of it and forgot to go back and erase š
Hmm.
OK, to start with, I don’t think we are solely building a dictionary, in this process. We’re learning about how game systems work, learning about elements in them which seem to be necessary to successful games, learning which elements of a system are superfluous, and how these elements all work together. This stuff has already proven useful to me in trying to refine a game system, and I am pretty sure it can prove useful to others as well.
I DO actually think we are talking about grammar; I think there are implicit rules in how systems fit together in a game overall, and rules in how the elements of a system fit together, and that we currently understand them solely on an ad hoc basis, that we deal with them as black boxes without understanding why and how they fit together. This does not mean that having a better understanding of these rules will necessarily make you better at making games — as Andy said, rules are made to be broken, when it comes to making great stuff. But it can serve as a guideline, as a helper. It can also serve as a tool for understanding what has been done, so that we can learn from past work better.
Games seem like a sort of social science. Like psychology or sociology. Pscyhologists definitely do not say that they are creating a “grammar” of the brain or of human behavior. Instead they are engaging in scientific endeavor: hypothesize, experiment, conclude (then rinse and repeat).
They do use grammars along the way to write their papers and with mathematics, but these grammars are simply formal tools, part of a larger endeavor that is applied and not formal. The grammars are derivative from applied results.
So I really do think that you are misusing the word grammar here. You have a grammar, more or less, and it’s useful to a limited extent, but byeond that you’ve got to get your hands dirty and figure out details of current usage and current game systems. I.e. build up a dictionary and then expand upon the understanding of its constiuent words.
Well said Andy.
Does modern art have a grammar? Should it? I don’t think so. It has a vocabularly of techniques that people have used and understood in the past and it is constantly adding new words to this or removing ones that fall out of use.
English has a grammar, but novel-writing does not. The closest you have to a novel-writing grammar is: introduction, complication, climax, conclusion but that “grammar” has more to do with understanding what some people enjoy in a story. It’s an applied result that could be expressed as a grammar. Anything beyond that is a creative/subjective process. Someone might develop a specific technique which uses a specific grammar but that grammar would just be a tool, a way of explicating what is and is not a use of that technique, it wouldn have veyr little to do with the overall process of writing novels.
Music has a very cool system of notation that is a sort of grammar. But even that is derivative of applied results. Musical notation is a nice tool but it’s not a grammar for “writing good music”. It is simply a very useful tool that is used a lot along the way to writing good music. I don’t think you’d call musical notation the “grammar of music”. If you really want to get down to understanding modern music, instead what you do is something like the music genome project which is again: compiling a dictionary of words to talk about modern usage in music and then trying to expand upon the understanding of each of those words.
So why do we think that there is a grammar of games?
Games do not seem to me like a social science at all, at least in terms of how games work within themselves (I agree that there is huge overlap with the social sciences in terms of how they are played, but that’s a somewhat different subject). Game design has certainly proceeded on a hypothesis-test case basis, with remarkably few conclusions being drawn.
Games strike me as being more like architecture or music or programming, fields where there is plenty of room for hypotheses and experimentation and expression, but where there are also fundamental rules that are followed.
Art does have a grammar, but it is made up of very fundamental elements such as color theory, weight, negative space, and so on; that’s what the “visual design” course they make you take at the outset of studying studio art consists of. There are known and defined relationships between these elements, many relying on aspects of human cognition (warm and cool colors, how they fit together, how line weight is read by the eye, and so on). These aren’t just a bag of tricks, they have relationships to one another. It isn’t a complete grammar by any means.
Music analysis most definitely reveals a grammar of harmony at the very least. Even jazz, which is potentially incredibly complex, has been discovered to have a grammar of chord substitutions based on leading tones and a fixed set of rules. (Francois Pachet at Sony CSL in France did a lot of work on this).
All that said, of course any result will be partial, and may not even be directly utilitarian. But I do find the “therefore don’t bother” sort of thinking to be shortsighted; for one, I think the investigations that various folks have already done have already proven fruitful. For another, I think that even just the exercise teaches us more about how games work.
Oh, and as to why I think there is a grammar of games: because all games seem to work in the same way. They are all systems with certain characteristics. The bigger games break down into many smaller games glued together. The core verbs available are remarkably few. The types of challenges presented are also remarkably few. The combinatorial nature of it all makes it seem like there’s a huge variety in the field, but when you look at the actual number of moving parts, it sure seems fairly small.
Most importantly, bad games all seem to break in the same set of ways. That indicates that there’s a reliable set of rules that ought to be followed there.
See that’s not a grammar either though. At all. That is a dictionary of things that people have played around with when creating art. It is a constantly expanding and contracting dictionary as new techniques and ideas come into our out of use.
How those techniques are used depends entirely on the audience and the artist. It’s subjective and grammars are not subjective.
Again this is the same. Examples of specific rules that Jazz artists may use are not pieces of a “grammar of music” they are specific techniques that have been developed and are part of a large list of such techniques with new techniques being added all the time.
See I really think you’re on the wrong track when you say this. There is really isn’t such thing as a partial grammar. A grammar is a set of rules. That may or may not be useful. But it’s won’t be “partial”. I think that you’re making category errors like this because you are conflating semantics with syntax.
Just look at story-writing.
You can say that stories are just made up of sub stories. You can talk about how to construct a good climax. You may think that you can classify where stories fail with a certain number of flaws. Whatever.
But after writing millions of books as a culture I think we’ve understood that there is no single central theory to writing good books. Certainly no notation or grammar to it all although many authors come up with their own rules, their own notational styles for writing outlines, etc.
Aren’t games really a superset of stories? Aren’t they even more complicated?
You seem to be assuming that a grammar is not a grammar until it is complete. But grammars aren’t static objects except in the strict mathematical sense — particularly not linguistic grammar. What I mean is that we may find some of the rules, but not all. It’s a work in progress to discover the grammar.
To use the music example; there used to be a complete and thorough grammar of music, specifically four-part harmony. Everything permissible was present, everything not permissible was excluded. Then we found more rules, and the old, believed to be complete grammar was extended, and in fact some of its rules were discarded (parallel fifths and octaves, once despised, today give us the power chord; the dreaded “devil’s interval” of the tritone birthed jazz).
The thing that IS known is that even in this expansion, it’s still not “anything goes.” There’s stuff that just does not lead well to other stuff, material that simply does not connect well. When you do harmonic analysis and break down jazz through all the substitutions, you often end up right back at classic I-IV-V, the most fundamental “sentence structure” in music harmony.
Actually, I think one of the things that even the early explorations of a grammar have clearly told us is that games are NOT a superset of stories, and are significantly less complicated.
I think thats most applicable to what your talking about, as linguistics more closely resembles the system thats being described.
True, in describing data about games I came to the conclusion that may be helpful to this discussion. Since there was no broadly defined “data dictionary” (that I know of) for the gaming industry, it follows there was no robust data available. Comparatively speaking (say to other major industries) the gaming industry, as far as data is concerned, is squarely in the data stone age (Note: as far as public access is concerned).
Therefore describing gamer data required some form of classifacation on my part.
This system decribes a player’s experiance, and the player, so yes its closer to linguistics. My impression from doing this was that there are no rigid structures/systems in games (otherwise the game industry would’ve been capturing data more efficiantly) because there are no rigid rulesets for how video games are made, or about players who play them, they seem to be an amalgam of subjective and objective data.
On the other hand my objective is only to describe data about gamers, not games themselves, what goes into making them is not my focus, the end result the “who, how, what and why” of thier consumption is my concern, so that those who do make them (or study, or sell, or market, or distribute them) have the data to make them better. So maybe I’m more concerned about how the language/grammer is used and less about the constructs of the language, thats better left to others obviously.
Any work around, perhaps codifying is to strong a word, “understanding” games helps immensly. The work on Grammer of Games (and ATOF) so far was really helpful for conceptualizing, so keep it going.
Now obviously Grammer of Games is more “catchy” and I think StGabe wants you to perhaps rename it “Dictionary of Games”, and I’m not suggesting “Linguistics of Games”, however I think continued use of “Grammer” is going to lead to as many interesting discussions about the title as the message contained therin.
When I first saw this a while back, I thought of using the idea of UML diagrams (or maybe I read it somewhere…I forget) but inverting them to make the verbs the classes and the requirements or objects needed for the verbs into the nouns. Here are some of the ideas I tossed around:
Classes are verbs and the data corresponds to requirements for using the verb, and the methods are related to the different entities that can carry out the action.
Use cases show how the verb is used by different nouns.
Interaction diagrams might show how when verbs are acted on my different methods (different nouns using them), they transition to a state where a different verb can be used.
I dunno, it doesn’t really make sense to me but since UML is nice for describing interactive software, and if it isn’t good for describing games because verbs are more important than nouns, then maybe it would be good to should start with the verbs and use the tools built up for noun-based software as a starting point.
This was never the “grammar of music” it was a grammar/notational style used by some musical techniques.
What you really mean to say is “a grammar used by some musicians”. That doesn’t sound dramatic but it is far more accurate. It’s like trying to say that C or C++ is THE grammar of computing.
And what you’re really going to have to look for is techniques. Not magic formulas.
Also, the musical notations you are talking about are for a certain audience. That is the subjective bit. Different techniques/forms are going to be more or less successful on a per-listener basis. What seems like a fundamental structure for jazz may or may not have anything to do with what makes for “good” progressive rock. So you may develop some rules for one but they will likely have nothing to do with the other. Some rules (techniques actually) may seem to have a fairly broad applicability, and they will be more useful because of that, but there will always remain the outliers to which they don’t apply and that set of outliers may even grow quite large with time.
Because really you’re just listing known techniques for known. Tomorrow we may discover an entirely new way to do music or an entirely new audience. And those won’t be “wrong” because they weren’t in our current dictionary (nor will the old dictionary have been wrong for its time). We may drop some techniques for a while and then pick them back up later and neither the grammar (actually, again, it’s a dictionary at this point) “with” those techniques nor without will be right or wrong. There is no underlying “right” or “wrong” to music, stories, games or art. Just a bunch of stuff that we know works and a bunch of stuff that, at least so far, hasn’t worked, and then much, much, much more stuff that we haven’t gotten around to trying and even more stuff that we haven’t even thought of.
Contrariwise, grammars are nothing but a way of saying what is right or wrong. They are just formal systems which indicate whther something is valid (in the set of what is included in the grammar) or invalid (not in that set). A grammar can tell you that something uses correct musical notation (according to a given standards) or is valid English. It can’t tell you whether that music is good or that sentence is meaningful. It has no subjective/semantical value.
I know I’m harping on this a lot but I guess I hope that if express this enough ways then it will make sense. I’ve actually studied grammars and linguistics quite a bit and I really do think I know what I’m talking about here.
This is a very good example. You don’t look up verbs in a grammar. You look them up in a dictionary. Think about English, as an example of a grammar, it is not important what verb you use in a sentence (gramatically) but just where it is placed in relation to other sentence structures.
For example, the sentence:
“I smell blue.”
Is grammatically correct. It is semantically invalid because you can’t smell colors. But that has absolutely nothing to do with grammars. To suggest, for example, that your grammar include a rule about what things were valid to “smell” would be missing the point of grammars.
A generative grammar JUST generates all elements of a language, meaningful or not.
An analytic grammar JUST distinguishes between members of a language and non-members, meaningful or not.
That is all grammars can or will do.
You are not talking about grammars here.
We’re definitely talking at cross-purposes here. Bear in mind that I have NOT studied grammars in any depth whatsoever, so this is almost certainly a terminology thing on my side. That said, I think we still have a disagreement underlying the whole thing.
To start with, the musical analogy really is very apropos. Music DOES have underlying rules about stuff that works and stuff that doesn’t, and any musician can tell you they exist. At any given time in history, the formalized version of those rules has been inadequate, incomplete, and destined to be overridden. But stuff like leading tones simply doesn’t go away.
There are dozens of notations for music. Some of them deal with these underlying rules well, some don’t. Some approach from a harmonic point of view, others from a melodic point of view, others from a rhythmic point of view. Then again, there’s lots of notations for linguistic grammar too.
Known techniques are what I have been calling patterns. I’m after the stuff underlying the known techniques. Every pattern that I have looked at (and that IS something I’ve spent a lot of time on) has had these common structural elements, in terms of how mechanics fit together and how the patterns fit together with each other as well.
So, I am not talking about “what things are valid to smell” as you put it. I think that you can, even now with what little we have of what I am calling grammar, see structures that are “syntactically” correct but “semantically” could easily be made silly.
I think where we part ways is that you believe that all the things that I (and Ben Cousins, and Stephane Bura, and others) see as having common underpinnings, you see as being a collection of techniques. When I say, in my “grammar,” that every “game atom” takes place in a topological space, I don’t see that as a technique, I see that as a fundamental element in how games work.
[…] Also something that should be caught is Tadhg’s comment in the comment thread, with its subtle and interesting distinction between “rules” and “mechanics” — the former are the physics of the game model, unalterable by users; the latter are the knobs the user can turn. ėźøė¬źø° https://www.raphkoster.com/2006/10/24/lost-garden-what-are-game-mechanics/#comments GoPets seems to be doing well in the US Raph 2006-10-24 11:30 ģģ± | Game talk GoPets: We based our entire financial model on users in the US paying, on average, five dollars a month and weāre seeing about triple that. […]
A phrase popped into my head while thinking about this on the commute home: “Prescriptive grammars”. I did a google search on this phrase and (not surprisingly) was sent directly to wikipedia:
I think prescriptive grammar is where you are headed but I don’t think that is what most linguists think of as grammar. As I’ve tried to state and as this article probably better states: grammars are really just sets of rules, determined through observation, that tell whether something is in a language (or not), according to the current usage of that language.
A grammar “of” (prepositions are important here, using “for” would make this a very different statement) music wouldn’t be a set of rules describing certain patterns/techniques in jazz music (or classical or what have you). It would be a much broader thing describing that certain sounds are played at intervals, often together. In order to truly be a grammar “of” music it would be too general to really be of much use and would include all sorts of things that were technically “music” but practically ugly and worthless.
These patterns, techniques or rules in music are indeed words or concepts with semantical content — not grammars. At least not grammar as linguists mean “grammar”. Perhaps prescriptive grammar of a sort.
As far as “atoms” and such I do agree that it can be useful to think of games in this way. But I think there is not one way but dozens of ways. Pick up any Game Theory book (whereby I mean not video or board game theory but game theory as practiced by folks like John Nash) you’ll find several dozen different notations for talking about games. It comes down to what notation is useful for the problem you want to solve. Nash, von Neumann, et al, came up with some notations and concepts that were indeed very useful for solving certain sorts of game and economic problems. Whether there is one specific notation, pattern or technique which is going to break open the entire field the practice of looking for patterns and techniques in general is certainly fruitful.
I just wouldn’t call it grammar. I just don’t think that’s really the right word for what you are doing. If you DID come up with some pattern or technique or insight which truly did capture the essence of gaming (IMO impossible, but have it as you will) then maybe afterwards you would create a grammar (or more likely, multiple grammars) to describe parts of that technique as you used it.
[…] Raph Koster has some commentary here, with some interesting if not particularly productive discussion. […]
Nitpicking time! Oh boy.
Agreed.
Yes, there is. The subject is “joys.”
“You” is not the subject. You can say that there is an implied “by you” at the end of the sentence, but that’s not what a subject is. A subject has a certain position and some other qualities, none of which are shared by “by you.”
“Lasting” isn’t a verb. It is as you said a gerund, but as a gerund it’s an adjective modifying “joy,” not a verb.
Forget what your teacher told you, there’s nothing wrong with passive voice.
So yeah, it is bad writing, but as it turns out even people who claim to know about grammar often are just pulling stuff out their butts.
Carl? You wanna pick nits? Here we go…
First of all, whether or not “joy” is considered the subject of the sentence depends on whether or not you are deconstructing this sentence in terms of purely English grammar, or syntactic grammar. There are some linguists who would agree with you that, yes, in the passive voice, the subject receives the action of the verb. Since the “joy” is what “is found,” and the verb is passive, therefore, in this instance, “joy” could be seen as a subject. Other grammarians — working on modes that aren’t limited to English only, but working from, for example, a Latin base — will say that, by definition, saying that “the subject receives the action of the verb” is contradictory and foolish. Which is one of the reasons the passive voice is often to be avoided (we’ll get to that in a sec).
Let’s jump to the gerund for a moment. I didn’t say it was a verb; I said it was a “verb constuct [that] turns an action word into a descriptor.” Right. Not a verb. An adjective. A descriptor. What I said. So there’s one less nit to pick.
Back to the passive voice. The first issue with the passive voice being weak is, as I pointed out earlier, the reception of action by the actor. Even with a clear subject and object, the passive construct tends to be less clear, more obtuse and is often, frankly, the result of lazy writing. For example:
“The kite was flown by the boy.”
Is just silly compared to:
“The boy flew the kite.”
The oft used implication of subject in the passive voice is the second problem in many of its uses. In the example I gave, one will assume “you” as the implied subject, but it could as easily be:
“Lasting joy can be found [by penguins] in the arms of comfortable rooms here in luxurious Buena Vista hotels.” or…
“Lasting joy can be found [by rotting, animated corpses] in the arms of comfortable rooms here in luxurious Buena Vista hotels.”
You get the point.
The third weakness of the passive voice is that it sublimates the expected order of reading and understanding, often requiring more than one reading (especially in complex uses) in order to get.
And I think I said in my original comment, that no rules were broken. That there’s nothing “wrong” per se with the passive voice… that it’s just bad writing. There are times when it can be used very cleverly, especially when one is trying to achieve some of the effects that are often foisted inadvertently by incautious writers.
So… Of the nits you pick, the first may be said to be true… in part. From a certain definitional standpoint. The others? I don’t see where you’re in any way disagreeing with me. No nits.
And, by the way, many other readers of the Orwell essays to which your link refers believe that he was being sly and ironic in his overuse of the passive voice in them.
In actuality, when forums are full of people correcting each others grammar, and someone else finding an error in the original posters correction, proceeds to correct that which has been corrected, which could in fact cause embarrassment to the previous poster, they are not a grammar Nazi per se, but are in fact doing something well known in the English Lit. Departments of universities nationwide called “picking nits” which are in fact some type of bugs, although they have nothing to do with entomology and little to do with English, this whole picking or nits makes me glad I majored in Government, and hereās why: it trains you to write really long obtuse sentences that make little if any sense, that are badly formed, while raising your tolerance for bad writing that loses focus of the subject matter.
š
Which isnt to say that anyone writing is bad, but what were we talking about again?
Grammar. Which is amusing, since Wikipedia states, “The subfields of contemporary grammar are phonetics, phonology, morphology, syntax, semantics, and pragmatics.” (link) Which, if I’m not mistaken, covers all of the objections raised in this thread.
GRAMMAR, n. A system of pitfalls thoughtfully prepared for the feet of the self-made man, along the path by which he advances to distinction.
– Ambrose Bierce, The Devil’s Dictionary
[…] misunderstanding) he tries to talk them in circles. Full credit to StGabe for arguing with him in this post. StGabe is a game programmer, and computer scientist. He explains to Raph what a grammar is and how […]